<?xml version="1.0" encoding="UTF-8"?><rss xmlns:dc="http://purl.org/dc/elements/1.1/" xmlns:content="http://purl.org/rss/1.0/modules/content/" xmlns:atom="http://www.w3.org/2005/Atom" version="2.0" xmlns:cc="http://cyber.law.harvard.edu/rss/creativeCommonsRssModule.html">
    <channel>
        <title><![CDATA[Stories by LETTRIA on Medium]]></title>
        <description><![CDATA[Stories by LETTRIA on Medium]]></description>
        <link>https://medium.com/@lettria?source=rss-4f5319276610------2</link>
        <image>
            <url>https://cdn-images-1.medium.com/fit/c/150/150/1*2423xA-yF-GzkU7sRKCu6g.png</url>
            <title>Stories by LETTRIA on Medium</title>
            <link>https://medium.com/@lettria?source=rss-4f5319276610------2</link>
        </image>
        <generator>Medium</generator>
        <lastBuildDate>Thu, 19 Nov 2026 20:51:29 GMT</lastBuildDate>
        <atom:link href="https://medium.com/@lettria/feed" rel="self" type="application/rss+xml"/>
        <webMaster><![CDATA[yourfriends@medium.com]]></webMaster>
        <atom:link href="http://medium.superfeedr.com/" rel="hub"/>
        <item>
            <title><![CDATA[Explore Lettria’s New Look & Feel with Bettina D’ávila, Senior Product Designer at Mozza — Lettria]]></title>
            <link>https://lettria.medium.com/explore-lettrias-new-look-feel-with-bettina-d-%C3%A1vila-senior-product-designer-at-mozza-lettria-7718760d9cab?source=rss-4f5319276610------2</link>
            <guid isPermaLink="false">https://medium.com/p/7718760d9cab</guid>
            <category><![CDATA[product-design]]></category>
            <category><![CDATA[ux]]></category>
            <category><![CDATA[nlp]]></category>
            <category><![CDATA[ai]]></category>
            <category><![CDATA[product-management]]></category>
            <dc:creator><![CDATA[LETTRIA]]></dc:creator>
            <pubDate>Tue, 21 Mar 2023 16:54:18 GMT</pubDate>
            <atom:updated>2023-03-22T13:40:27.397Z</atom:updated>
            <content:encoded><![CDATA[<h3>Explore Lettria’s New Look &amp; Feel with Bettina D’ávila, Senior Product Designer at Mozza — Lettria</h3><figure><img alt="" src="https://cdn-images-1.medium.com/max/1024/0*0Kkn8xI8NGzW-4au.jpg" /></figure><p>With Lettria’s new platform currently being rolled out, we realized it’s the perfect time to spotlight how our team hit so many milestones and were able to implement our latest updates. In this interview, <a href="https://medium.com/@medinous">Bettina D’ávila</a>, discusses her experience working on the project on behalf of <a href="https://mozza.io/">Mozza</a>, our Product Studio, and how we aim to help our partners leverage natural language processing outside of data teams. Bettina shares her exclusive insights about the Lettria platform, including unique challenges of designing a text-focused platform and the importance of democratization.</p><h3>Our Team-building Backstory</h3><p><em>Bettina talks about how she got brought into the project and what was similar and different about the Lettria project compared to other tech projects she has worked on.</em></p><p><strong>Lettria:</strong> I think the best way for us to start would be going through the origins of the Lettria x Mozza partnership. How did you get brought into the project?</p><p><strong>Bettina:</strong> I’m relatively new at Mozza — I got brought into work with Lettria as my second project — but I specifically asked to join because I’ve spent the last few years tackling new sectors of the tech industry. I had previously worked as Product Design Manager at Doctolib (a massive French healthtech player) after spending three years working exclusively in e-commerce. When I joined Mozza as a product design freelancer, I decided to take on projects that help me branch out and learn up-to-date trends in the tech industry.</p><p>Bettina D’ávila, Senior Product Designer at Mozza</p><p><strong>Lettria:</strong> What was similar and what was different for the Lettria project, compared to health tech work for example?</p><p><strong>Bettina:</strong> The discovery phase, of course. I was constantly learning new information about healthcare and the medical industry in my previous job, and now I have become equally immersed in machine learning and artificial intelligence. But considering the larger mission, bringing something that was previously super complex and inaccessible to users is part of my job. I think that the barriers and complexities in these industries both need to be brought down and made accessible at a human level.</p><p>Before my time as a manager at Doctolib, I spent years working for AB Tasty, a SaaS tool to optimize user journeys on major websites. We took similar approaches for our client’s B2C websites, often the same tactics for categorizing information and simplifying complexity for our end users. For instance, Intricately presented graphs and data were blockers for certain segments of users and we wanted to build an inclusive, common user experience. On the other hand, at Lettria you have tons of amazing features that we want to spotlight, but they often require some background knowledge to fully grasp certain terms and concepts and actually get the most out of the product.</p><p>So, the path that we took was to build a digital experience that reduces the learning curve for the user, by introducing NLP concepts and guided information in a contextual way — since <a href="https://www.lettria.com/approach">the ultimate goal for Lettria</a> is to make NLP technology more democratic. A major priority was ensuring that users could naturally progress and evolve through the platform and develop their skills and understanding no matter their backgrounds.</p><h3>Flattening the Learning Curve</h3><p><em>Bettina discusses her personal experience with the learning curve of working with NLP for the first time and how she tackled it.</em></p><p><strong>Lettria:</strong> How was your personal experience with the learning curve? What was it like working behind the scenes with NLP for the first time?</p><p><strong>Bettina:</strong> Well, Natural Language Processing was an entirely new frontier for me. So I remember getting super lost when I first started using Lettria since I was missing some of the background and experience. But I’m naturally curious, and I took it upon myself to get immersed in the concepts and fully understand the steps of what was going on. This process was important to the work since from a designer’s perspective, breaking everything down into basics is super helpful.</p><p>Those baby steps and early understanding impressions helped us define how pretty much every end user will eventually use Lettria. We started jotting down the questions and context for these people, and I was able to look at the tool from different vantage points. Basic stuff like “What are my resources?” or “Where do I need to go?” helped me not only understand the user journey, but also the platform and what it offers overall.</p><p>Once you know the actions that are required to reach a specific goal, even complex topics like NLP can be broken down into simple segments. Then, over time, bit by bit you start to get a better grasp on what’s really going on regarding the tech side.</p><p>A taste of some of the updates Mozza helped bring to the table</p><h3>User Research Insights</h3><p><em>Bettina describes how they interviewed people from Lettria and clients using the platform to gather insights for the project. She explains how they defined three product pillars, and ultimately decided to prioritize Customer Feedback Analysis in terms of where to start first.</em></p><p><strong>Lettria:</strong> That’s a perfect way to launch us into my next question. You interviewed people from Lettria but also users on the client side as well. What were some of the best takeaways you got from that process?</p><p><strong>Bettina:</strong> Yeah, so we did our internal audit, which I was talking about before. After working with Lettria and getting to understand the fundamentals of the service, we took a step back and pinned all of the “Aha” moments that we had coming in on the project. Then, to really cement our understanding, we took that information and began approaching internal users and clients, finding out what their needs and experiences were while working with the platform.</p><p>All interviews were super technical, focusing on the takeaways from scientists and engineers and they had a lot to teach us about building up the self-service aspect of Lettria. Plus, each persona we interviewed had their own need and way of looking at the service so it was tricky in terms of defining a universal user roadmap since the needs and expertise get to be so diverse and vast.</p><p><strong>Lettria:</strong> So from there, how did you get to the point where you could prioritize a specific entry point? Which applications did you wind up tackling first?</p><p><strong>Bettina:</strong> We spent the first month and a half or so exploring the platform with the Mozza and Lettria teams. Here, we got to explore product vision and overall values that Lettria wants to espouse. We basically got to an understanding that democratizing the use of NLP outside of data teams was an important business goal, therefore we developed with Lettria team product initiatives to support an actionable and no-code strategy that we can adapt to all sorts of use cases.</p><p>Based on that, we defined three product pillars, which are essentially the most appreciated parts of the product. We decided that first and foremost, what’s important to Lettria’s users is the time to value, meaning: how early in the platform users can see the benefits of using Lettria on their projects and thus, how much time and money they can save. Secondly, how easy to use the platform is. And thirdly, the performance, so people are able to derive using Lettria and see how powerful the tool itself can actually get.</p><p>Once we settled and all agreed on the set of pillars, we tackled what we call “ <a href="https://www.lettria.com/solutions">platform stories</a>,” which are basically sets of ways to pitch and launch the product based on its different use cases. We talked a lot about the potential regarding Voice2CRM but ultimately decided to prioritize Customer Feedback Analysis in terms of where to start first. In terms of the fundamental scope of solutions, after analyzing the general feedback we’d developed early on, Customer Feedback Analysis helped us create a formula to model the rest after.</p><p>The Pipeline of a NLP Project with Lettria</p><h3>Design Choices and Next Steps</h3><p><em>Bettina discusses the challenges of proposing a new UX/UI and navigation system for the platform, and what she is most proud of in the new version.</em></p><p><strong>Lettria:</strong> From there, were there any hiccups in proceeding? Was there any moment where you had to change strategies or change what you’d started early on?</p><p><strong>Bettina:</strong> Honestly, it was incredibly smooth from that point of initial audit. We’d really done a ton of legwork in terms of understanding the platform and defining the user journey, at that point we just had to start adapting our solutions and putting everything into place. Working alongside the Lettria team was pretty fast-paced and productive because, at a high level, we were all on the same page objective-wise. There were some design choices that we would have to debate and a few smaller details on specifics to hash out, but nothing out of the ordinary.</p><p><strong>Lettria:</strong> So what are you most proud of in the new version, now that the new beta’s officially come out?</p><p><strong>Bettina:</strong> Really the entire structure and navigation system is something I’m proud of because it was truly a launch that I’ve seen grow from the ground up. This is my area of expertise, and it’s always super challenging because information architecture and navigation fundamentally define how people are going to understand and enjoy the entire platform. I love working on that stage because it’s where the logical and structural changes come in, but it’s a lot of work because it sets the stage for everything you end up modifying next.</p><p>There’s a lot of logic that’s been built into the platform with this new design, things like child view/ parent view and rethinking the overall hierarchy. Also, design details like the typography makes a huge difference in the way the platform is appreciated and understood. As a text-focused platform, we’ve made headways in terms of presenting information that’s easy to scan quickly or read intensively.</p><p>It’s been fun working on this project and there’s been a ton of flexibility in the Lettria x Mozza collaboration. I really think that users will appreciate the progress we’ve made in making the platform as approachable as we can. There’s still a lot of work to do in terms of accessibility and future integrations, but this new rollout definitely marks a huge improvement in terms of the values we defined and Lettria’s product vision overall.</p><h3>Discover the new version yourself!</h3><p>Overall, the Lettria x Mozza collaboration has resulted in a platform that is more approachable, easy to use, and powerful. Check out all the latest features in this tutorial video, and <a href="https://app.lettria.com/">create your Lettria account </a>on the website to start exploring our toolkit for free.</p><p><em>Originally published at </em><a href="https://www.lettria.com/articles/explore-lettrias-new-look-feel-with-bettina-davila"><em>https://www.lettria.com</em></a><em>.</em></p><img src="https://medium.com/_/stat?event=post.clientViewed&referrerSource=full_rss&postId=7718760d9cab" width="1" height="1" alt="">]]></content:encoded>
        </item>
        <item>
            <title><![CDATA[What is sentiment analysis and how does it work? — Lettria]]></title>
            <link>https://lettria.medium.com/what-is-sentiment-analysis-and-how-does-it-work-lettria-ed7b1a64c87c?source=rss-4f5319276610------2</link>
            <guid isPermaLink="false">https://medium.com/p/ed7b1a64c87c</guid>
            <category><![CDATA[nlp]]></category>
            <category><![CDATA[ai]]></category>
            <category><![CDATA[emotion-analysis-api]]></category>
            <category><![CDATA[sentiment-analysis]]></category>
            <dc:creator><![CDATA[LETTRIA]]></dc:creator>
            <pubDate>Mon, 23 Jan 2023 18:26:59 GMT</pubDate>
            <atom:updated>2023-01-23T18:26:59.775Z</atom:updated>
            <content:encoded><![CDATA[<h3>What is sentiment analysis and how does it work? — Lettria</h3><p>So you want to know more about <a href="https://help.lettria.com/what-is-sentiment-analysis">Natural Language Processing (NLP) sentiment analysis</a>? Well, then you’ve come to the right place. In this article, we’ll help you understand how sentiment analysis works, explain how the Lettria platform has been developed to make this process easier and more powerful, and provide you with some examples of business use cases and applications for sentiment analysis.</p><p>The TL;DR explanation? Sentiment analysis (sometimes referred to as opinion mining or emotional artificial intelligence) is a natural language processing technique that analyzes text and determines whether the data is positive, negative, or neutral.</p><p>The applications and use cases are varied and there’s a good chance that you’ve already interacted with some form of sentiment analysis in the past. But before we get into the details on exactly what it is and how it works, let’s (all too) quickly cover the basics on natural language processing.</p><h3>Natural Language Processing</h3><p><a href="https://www.ibm.com/cloud/learn/natural-language-processing">Natural language processing allows computers to interpret and understand language</a> through artificial intelligence. Over the past 50 years it has developed into one of the most advanced and common applications for artificial intelligence and forms the backbone of everything from your email spam filters to the chatbots you interact with on websites.</p><p>Through <a href="https://mitsloan.mit.edu/ideas-made-to-matter/machine-learning-explained">machine learning</a> and algorithms, NLPs are able to analyze, highlight, and extract meaning from text and speech. In its simplest form, this could simply be using an artificial intelligence (AI) model to scan text to identify particular words or phrases, but in its more advanced applications, as we will touch on later, it allows computers to identify the emotion of the speaker themselves — sometimes better than even a human can manage.</p><p>NLPs are typically employed when an organization can benefit from analyzing sets of data that are far too large to process manually. Whether it’s trying to monitor social media or emails or extract information from reviews or exchanges with customers, there’s a good chance that having humans analyze the information will be too time consuming and resource heavy.</p><p>Syntactic analysis (sometimes referred to as parsing or syntax analysis) is the process through which the AI model begins to understand and identify the relationship between words. This allows the AI model to understand the fundamental grammatical structure of the text, but not really the text itself. For example, sentences can be grammatically correct and not make any sense, or it could fail to identify the contextual use of some words as a result of the sentiment or emotion within the text (sarcasm being a common issue).</p><p>We won’t go into too much detail about exactly how an NLP works in this article, but they use text vectorization to transform the data into a format that the AI model can understand and then process the textual data using machine learning algorithms that allow the computer to make associations between an input and the correct output.</p><p>This process means that the more data you feed through your NLP the more accurate it becomes. With each new analysis allowing it to build a more complete knowledge bank that helps it to make more accurate and complete analysis.</p><p>NLPs have now reached the stage where they can not only perform large-scale analysis and extract insights from unstructured data (syntactic analysis), but also perform these tasks in real-time. With the ability to customize your AI model for your particular business or sector, users are able to tailor their NLP to handle complex, nuanced, and industry-specific language.</p><p>You can find out a lot more about the benefits and applications of an NLP in other sections of our website, but a platform like Lettria allows users to go beyond that simple textual analysis and actually identify the sentiment and emotion expressed within the text itself.</p><h3>‍What is sentiment analysis?</h3><figure><img alt="" src="https://cdn-images-1.medium.com/max/953/0*3hKD7JNBNw2rkdP3.png" /></figure><p>So we’ve given you a little background on how natural language processing works and what syntactic analysis is, but we know that you’re here to have a better understanding of sentiment analysis and its applications.</p><p>AI models in natural language processing can begin to gain a more complete understanding of language by using sentiment analysis to understand both the polarity of text (positive, negative, or neutral) and the specific emotions expressed within a text.</p><p>Sentiment analysis is the foundation of many of the ways in which we commonly interact with artificial intelligence and it’s likely that you’ve come into contact with it recently. Have you started a conversation with customer support on a website where your first point of contact was a chatbot? Sentiment analysis is what allows that bot to understand your responses and to point you in the right direction.</p><p>Have you tried translating something recently and wondered how the program is understanding your original? Well, if it works well, then that will be relying on Natural Language Processing (NLP) with sentiment analysis to help identify the contextual meaning and nuance of what you are trying to translate.</p><p>Sentiment analysis is all around us. So, if you’re new to the game and yet to start using it to your advantage, this article will help you to better understand its various applications and explain how you can start using sentiment analysis to gain invaluable business insights.</p><p>Not all sentiment analysis applies the same level of analysis to text, nor does it have to. There are times when simply understanding the polarity of a piece of text is sufficient, but most use cases (a topic that we go into in a later section) for sentiment analysis require natural language processing that is capable of identifying the emotional context of the text.</p><p>Sentiment analysis is very much a long-tail problem. That is to say that there are many different scenarios, subtleties, and nuances that can impact how a sentence is processed.</p><p>If you’re only concerned with the polarity of text, then your sentiment analysis will rely on a grading system to analyze your text. This might be sufficient and most appropriate for use cases where you are processing relatively simple sentences or multiple choice answers to surveys or feedback.</p><p>This graded sentiment analysis requires that you train your AI model to associate a certain score to different words and terms. From there, it can apply a plus or minus grading system to each piece of text to determine if it is positive, negative, or neutral.</p><p>For example, you might train your AI model to associate the word fast as being positive and the phrase time-consuming as being negative. This means that the following sentence:</p><blockquote>The delivery time was really fast</blockquote><p>Will be labelled as positive. Whereas this sentence:</p><blockquote><em>‍</em>Filling in your return form was really time-consuming</blockquote><p>Would be determined to be negative. These are both pretty simple examples, but even for polarity of text you might need a more powerful NLP once you start to deal with sentences that contain multiple sentiments. Say, for example:</p><p><em>Filling in your return form was really time-consuming, but the refund was handled very quickly.</em></p><p>This sentence contains both positive and negative statements. If your AI model is insufficiently trained or your NLP is overly simplistic, then you run the risk that the analysis latches on to either the start or the end of the statement and only assigns it a single label.</p><p>Fortunately, with a platform like Lettria you can ensure that your NLP can identify multiple sentiments from within the same statement. But what if you want something a bit more complex than just polarity of text and you actually want to get into real emotion detection and analysis?</p><p>Emotion detection systems are a bit more complicated than graded sentiment analysis and require a more advanced NLP and a better trained AI model.</p><p>Emotion detection requires more complex machine learning algorithms or detection systems that use lexicons with assigned emotional meanings or associations (in many ways, these detection systems are not that dissimilar to the approach used in graded sentiment analysis).</p><p>This obviously presents a number of monumental challenges and understanding and interpreting the emotional meaning behind a piece of text is not easy. Even humans make mistakes when it comes to analyzing the sentiment within text or speech, so training an AI model to do it accurately is not easy.</p><h3>Challenges to sentiment analysis</h3><p>Language is an incredibly complicated thing. Just in writing this article I’ve managed to confuse myself on several occasions — and that’s when I’m faced with the relatively simple challenges of analyzing my own text.</p><p>As we have already discussed, an NLPs AI model has to be fairly advanced in order to begin to identify the sentiment and emotional message expressed within a text. Some sentences are relatively straightforward, but the context and nuance of other phrases can be incredibly challenged to analyze.</p><p>Take for example:</p><blockquote>The hotel has comfortable beds.</blockquote><p>It would be relatively easy to label this sentence as positive. There’s a single adjective, it is clearly associated with a positive sentiment, and the sentence only conveys one message with no great complexity.</p><p>But things can get complicated very quickly…</p><h3>Context</h3><p>The first challenge is to understand the context of the sentence. The meaning of the same set of words can vary greatly depending on the context in which they are said. It could be impacted by the previous sentence or the specifics of certain technical language.</p><p>Take for example, the following customer response:</p><blockquote>Every single feature</blockquote><p>If your support agent has just asked a customer what they like about your product then this is clearly a positive sentence, but if they just asked your customer what they didn’t like about your product then it would fall firmly into the negative category.</p><p>That’s why it’s important that your NLP is capable of not only analyzing the individual statements, sentences, and words, but also being able to understand their placement and usage from a contextual standpoint.</p><h3>Figures of Speech</h3><p>Figures of speech can also greatly change how sentences and words should be interpreted. The most obvious examples are with irony and sarcasm, where their presence can completely flip the meaning of a word or phrase.</p><p>Let’s take another example from an exchange between a support agent and a customer. Say the support agent asks the customer if they are enjoying their experience with your product and they get the following response:</p><blockquote>Well I’m here speaking to your customer support, so you can tell that I’m really loving it.</blockquote><p>The contextual clues within the sentence and previous exchanges would allow an NLP capable of sentiment analysis to identify this statement as being sarcastic and so not label the sentence as positive, even though it would appear to be so when viewed in isolation.</p><h3>Comparisons</h3><p>Another area where sentiment analysis can ensure that natural language processing delivers the correct analysis is in situations where comparisons are being made.</p><p>Comparisons can sometimes be simple and straightforward. Take the following examples:</p><blockquote><em>‍</em>This product is better than any of the competition</blockquote><blockquote>This was the best experience that I’ve ever had</blockquote><p>Both statements are clearly positive and there’s no real requirement for any great contextual understanding.</p><p>But things can get more complicated in a situation like this:</p><blockquote>I suppose it is better than nothing</blockquote><p>The statement would appear positive without any context, but it is likely to be a statement that you would want your NLP to classify as neutral, if not even negative. Situations like that are where your ability to train your AI model and customize it for your own personal requirements and preferences becomes really important.</p><p>Lettria allows users to get their project up and running and customize their AI model 75% faster than the off-the-shelf NLPs. This means that not only are you able to start seeing results more quickly, but you’re also able to dedicate more time to really defining how you want your AI model to analyze the text and allow it to perform more accurate sentiment analysis by providing it with more examples and more clearly defining positive, negative, and neutral statements.</p><p>That additional information can make all the difference when it comes to allowing your NLP to understand the contextual clues within the textual data that it is processing.</p><p>So, on that note, we’ve gone over the basics of sentiment analysis, but now let’s take a closer look at how Lettria approaches the problem.</p><figure><img alt="" src="https://cdn-images-1.medium.com/max/961/0*sjBa7gbICRaM473L.png" /></figure><h3>‍How does Lettria handle sentiment analysis?</h3><p>We’ve already hinted at the fact that not all NLPs are created equal, and Lettria has put itself into a unique category by providing users with a low-code or no-code platform that specializes in customizable textual data processing.</p><p>This platform-based approach allows users to benefit from Lettria’s 15 pre-trained and customizable multilingual models to employ an NLP to start with a stable foundation and focus on training their AI model to be highly specialized for their specific business, industry, and use case.</p><p>Lettria offers all of the benefits of an off-the-shelf NLP (implementation and production time) with the power and customization of building one your own (but 4 times faster). Alright, that’s the sales pitch done, now let’s take a closer look at how Lettria actually handles sentiment analysis.</p><p>Lettria’s API uses resources from psychology and the 8 primary emotions modelled in Putichik’s wheel of emotions (joy, sadness, fear, anger, attract, surprise, and anticipation). From there, we break the analysis into two steps.</p><h3>‍Step 1</h3><p>Our algorithm analyzes the text to identify the adverbs and adjectives that are modifiers of meaning within a text. Once this is complete and a sentiment is detected within each statement, the algorithm then assigns a source and target to each sentence.</p><figure><img alt="" src="https://cdn-images-1.medium.com/max/697/0*8coM-qbn6PfKRoT6.png" /></figure><p>The emotional value of a statement is determined by using the following graded analysis.</p><p>Values are calculated by either using the sentiment objects or, if they are not available, by a prediction model at the sub sentence level. Values are normalized to stay within the -1 : 1 interval between the element, sub sentence and sentence level comparisons should (again) therefore be made with elements of the same depth.</p><figure><img alt="" src="https://cdn-images-1.medium.com/max/697/0*rEGSm2ygrPY9H8UK.png" /></figure><p>With the sentiment of the statement being determined using the following graded analysis.</p><p>All sound a bit confusing? I know how you feel, but let’s use a real-world example to make things a bit clearer.</p><p>Take a simple sentence like ‘I like reading’ (at least, I hope you do if you’ve decided to make your way through this article).</p><p>The statement contains an overall positive sentiment, an emotion of joy as defined by the 8 primary emotions, and an emotional intensity of .46 (on a scale of -1 to 1).</p><p>Now, say you’re really enjoying this article and decide to leave a comment saying ‘I really like reading’ then you would still return a positive sentence, but the addition of ‘really’ would increase the value of the emotion to .66.</p><p>This first step essentially allows Lettria to carry out the graded sentiment analysis and polarity of text analysis that we discussed in the previous section. The second step is where we start to process the context and the real emotion expressed within the text.</p><h3>Step 2</h3><figure><img alt="" src="https://cdn-images-1.medium.com/max/1024/0*A_XP4JM-l9g2Vzc2.png" /></figure><p>To gain a more complete understanding of the emotions of a sentence, Lettria uses deep learning to identify the context of the sentiments within a text. This is done using the framework of 28 different emotional values.</p><p>These emotional guidelines help the AI model to understand the context of the sentiments being expressed. When you combine steps 1 and 2, Lettria is not only able to determine the polarity of a statement, but also the emotional context and value within a sentence.</p><p><em>Your product is so unreliable, why does it always fail so easily.</em></p><p>So, when we then get a slightly more complicated example that looks something like this:</p><p>The analysis would be able to determine that this statement is still negative, in spite of the fact that you could have trained your AI model to associate ‘easily’ with being a positive.</p><p>In many ways, you can think of the distinctions between step 1 and 2 as being the differences between old Facebook and new Facebook (or, I guess we should now say Meta). At first, you could only interact with someone’s post by giving them a thumbs or thumbs down. Which essentially meant that you could only react in a positive way (thumbs up), negative way (thumbs down), or natural way (no reaction).</p><p>But, they eventually introduced the ability to use a wide range of different emojis that allowed you to express a variety of different emotions and reactions. This meant that the original poster had to think a bit more deeply when they wanted to interpret your reaction to their post (and account for the possibility that you might have been sarcastic or ironic).</p><p>This increase in a user’s ability to react to posts is actually an example of a use case where sentiment analysis can be used to help better understand large data sets of social media interactions (more on that later), but it is a good way of thinking about when you might want a more advanced NLP to process your sentiment analysis and deliver more nuanced results.</p><p>The reality is, for all of the use cases and applications that we are about to touch on, you need an NLP that is capable of doing more than just graded sentiment analysis.</p><p>The Lettria platform has been specifically developed to handle textual data processing and offers advanced sentiment analysis. Delivering a high level of accuracy and the ability to customize your AI model to suit all of your specific business and industry requirements, Lettria is able to address all of the use cases where sentiment analysis is applied.</p><p>So let’s take a closer look at when and where organizations are typically using natural language processing sentiment analysis and you can see if there are applications that could help you to improve your performance and your understanding of your customers and business.</p><h3>Sentiment analysis use cases</h3><p>Although the applications for natural language processing sentiment analysis are far-reaching and varied, there are a few use cases in which the analysis is commonly applied. Sentiment analysis can help organizations to gain invaluable business insights in virtually any situation in which they are faced with the task of handling large quantities of unstructured textual data, but here are a few examples of areas where your company or project could immediately benefit from sentiment analysis.</p><h3>Social Media Monitoring</h3><p>If you aren’t listening to your customers wherever they speak about you then you are missing out on invaluable insights and information. That means that social media platforms are areas where your leads, customers, or former customers will be sharing their honest opinions about your product and services.</p><p>You might have a social media team and some social media managers who track conversations and try to engage with users on various platforms, but it’s unlikely that they have the time, resources, and capabilities to not only track and interact with every conversation, but also accurately record the exchanges.</p><p>That’s where natural language processing with sentiment analysis can ensure that you are extracting every bit of possible knowledge and information from social media.</p><p>Your knowledge and understanding of your leads and customers can be significantly increased by analyzing their social media posts and comments and, perhaps more importantly, real-time social media monitoring can help you to manage by exception and identify the comments and trends that require the most attention.</p><p><a href="https://www.mckinsey.com/capabilities/operations/our-insights/social-media-as-a-service-differentiator-how-to-win">Research from McKinsey</a> shows that customers spend 20 to 40 percent more with companies that respond on social media to customer service requests. Not only that, but companies that fail to respond to their customers on social media experience a 15 percent higher churn rate.</p><p>The even more concerning statistic for unresponsive companies? Customers expect to hear back from you quickly. 40 percent of consumers expect a response within the first hour and 79 expect a response within the first day.</p><p>What does all of that mean? You risk losing business, and lots of it, if you’re not able to identify the social media posts and comments that require your attention and meaningful attention.</p><p>Employing natural language processing sentiment analysis to analyze your social media in real-time will ensure that your social media and customer support teams will be able to identify the customers that require immediate human attention.</p><p>It increases efficiency, improves resource allocation and time management, and, most importantly again, improves customer experience and brand loyalty.</p><h3>Customer Service</h3><p>We’ve already touched on how sentiment analysis can improve your customer service on social media, but it can also improve your customer service performance through other channels.</p><p>For starters, natural language processing sentiment analysis is a key element for high-performing chatbots. You may be employing an off-the-shelf chatbot that applies basic filters to your customer conversations, but you also have the ability to train an AI model that will be customized for your specific business needs and language.</p><p>What’s more, sentiment analysis can help you to filter incoming customer support tickets and ensure that they are labelled correctly, passed on to the appropriate team or department, and assigned the correct level of urgency.</p><p>Much like social media monitoring, this can greatly reduce the frustration that is often the result of slow response times when it comes to customer complaints. It is also another example of where sentiment analysis can help you to improve resource allocation and efficiency.</p><h3>‍Brand Reputation</h3><p>Our first couple of use case examples help companies to improve their short-term performance in a way that can have a lasting impact on long-term results, but our next few examples are much more focused on long-term thinking, planning, and performance.</p><p>Social media monitoring and customer service responses can play a key role in improving brand loyalty, but it also helps you to identify the areas of your brand that are performing the best and those that require attention.</p><p>Sentiment analysis allows you to train an AI model that will look out for thoughts and messages surrounding particular topics or areas. To monitor in real-time all of the conversations that relate to your brand and image.</p><ul><li>Perceived brand value — if your products or services offer competitive value within your industry</li><li>Perceived brand quality — do your products or services match your competition when it comes to actual quality</li><li>Perceived brand trust — do your customers feel comfortable with their decision to be loyal to your products or services</li></ul><p>Brand loyalty is typically broken down into three key areas:</p><p>Understanding how your customers feel about each of these key areas can help you to reduce your churn rate.<a href="https://media.bain.com/Images/BB_Prescription_cutting_costs.pdf"> Research from Bain &amp; Company</a> has shown that increasing customer retention rates by as little as 5 percent can increase your profits by anywhere from 25 to 95 percent.</p><p>We can all fall in love with the idea of a new customer, but making sure that you take care of your existing customers is just as important. Real-time monitoring through sentiment analysis will improve your understanding of your customers, help you to have more accurate net promoter scores, and ensure that your existing customers become loyal customers.</p><h3>Market Research</h3><p>There’s a good chance that you’ve already run campaigns that have included surveys and other initiatives to help you get feedback from leads and customers. That kind of market research can provide invaluable insights into your company, how you’re perceived, and your competition; however, if those interactions involve anything more complex than multiple choice answers then there’s a good chance that you’ve left some knowledge on the table.</p><p>As with social media and customer support, written answers in surveys, product reviews, and other market research are incredibly time consuming to manually process and analyze. Natural language processing sentiment analysis solves this problem by allowing you to pay equal attention to every response and review and ensure that not a single detail is overlooked.</p><p>Why put all of that time and effort into a campaign if you’re not even capable of really taking advantage of all of the results? Sentiment analysis allows you to maximize the impact of your market research and competitive analysis and focus resources on shaping the campaigns themselves and determining how you can use their results.</p><h3>Voice of Customer and Net Promoter Scores</h3><p>The brand reputation use case made mention of how sentiment analysis can help you to have a more accurate net promoter score, but it’s worth taking a closer look at how it can improve your understanding of your NPS and Voice of Customer (VoC).</p><p>Similar to the market research use case, if you’re interested in tracking your net promoter score than sentiment analysis may well be the only way that you can carry out large-scale analysis without overlooking some of your feedback and results.</p><p>Open-ended questions have long been a nightmare for surveys and feedback, but sentiment analysis solves this problem by allowing you to process every bit of textual data that you receive.</p><p>Sentiment analysis will not only help you to get every bit of detail, knowledge, and information out of all of these responses, but real-time tracking will mean that you’ll no longer have to wait weeks, months, or potentially even years to be able to take that understanding and apply it to your business.</p><p>As you can see, sentiment analysis can provide meaningful results for companies and organizations in virtually any sector or industry. It can improve your understanding of your business and customers and increase efficiency and performance.</p><p>So, the question isn’t really whether or not natural language processing and sentiment analysis could be useful for you. It’s simply a question of how you can make sure that your NLP project is a success and produces the best possible results.</p><h3>Start your own sentiment analysis project</h3><p>If you’ve made it this far then it’s fair to say that there’s a strong possibility that you’re interested in exploring the benefits that Lettria’s sentiment analysis could bring to your project or organization. It might be because you’re frustrated with your existing NLP project or you’re only beginning to explore the world of natural language processing.</p><p>Lettria’s platform-based approach means that, unlike most NLPs, both technical and non-technical profiles can play an active role in the project from the very beginning. This means that your work will not suffer from the silo effect that is the undoing of many NLP projects.</p><p>All too often, NLP projects are thought of as being the exclusive domain for data scientists and developers. It is true that they may play a crucial role in getting the project up and running, but most of the time it is other teams and profiles that benefit from the results and insights that natural language processing produces.</p><p>Lettria solved that problem by making the platform accessible to non-technical profiles and focusing on developing an approach that provides all of the power of a customizable NLP with the added benefits of increased collaboration and knowledge sharing.</p><p>Not only can your data scientists and developers work on an NLP that offers incredibly accurate and efficient results, but business profiles can be directly involved in the project itself and contribute their own knowledge and expertise and gain a better understanding of the insights that your analysis is revealing.</p><p>This engagement and ability for technical and non-technical profiles to collaborate and easily contribute to the same project increases buy-in, reduces failure rates (85% of NLP projects fail before they begin producing any meaningful results), and improves results.</p><p>So, whether you’re reading this as a data scientist or developer with an advanced understanding of natural language processing, or you’re a sales person or customer support specialist with absolutely no background in NLP platforms, Lettria is the platform for you.</p><p><em>Originally published at </em><a href="https://www.lettria.com/articles/what-is-sentiment-analysis-and-how-does-it-work"><em>https://www.lettria.com</em></a><em>.</em></p><img src="https://medium.com/_/stat?event=post.clientViewed&referrerSource=full_rss&postId=ed7b1a64c87c" width="1" height="1" alt="">]]></content:encoded>
        </item>
        <item>
            <title><![CDATA[How To Enrich Your CRM With Customer Feedback Analysis — Lettria]]></title>
            <link>https://lettria.medium.com/how-to-enrich-your-crm-with-customer-feedback-analysis-lettria-6dde4e2d7a56?source=rss-4f5319276610------2</link>
            <guid isPermaLink="false">https://medium.com/p/6dde4e2d7a56</guid>
            <category><![CDATA[data]]></category>
            <category><![CDATA[ai]]></category>
            <category><![CDATA[nlp]]></category>
            <category><![CDATA[crm]]></category>
            <category><![CDATA[enrichment]]></category>
            <dc:creator><![CDATA[LETTRIA]]></dc:creator>
            <pubDate>Mon, 23 Jan 2023 18:22:25 GMT</pubDate>
            <atom:updated>2023-01-23T18:22:25.466Z</atom:updated>
            <content:encoded><![CDATA[<h3>How To Enrich Your CRM With Customer Feedback Analysis — Lettria</h3><figure><img alt="" src="https://cdn-images-1.medium.com/max/1024/0*zmzJ1nuMeyqdVaxu.png" /></figure><p>Over the past few decades, Customer Relationship Management (CRM) has become the cornerstone of businesses looking to leverage data: according to statistics, more than 91% of companies with ten employees or more use CRMs to manage customer information. How many of them gain actionable insights from their CRM database, though?</p><p>This is not many, say studies. The <a href="https://www.validity.com/wp-content/uploads/2021/04/State-of-CRM-Data-Management-2022.pdf">2022 State of CRM Data Health Report</a> reveals that about 95% of companies face data quality issues that prevent them from fully leveraging their CRM database. Poor data quality has caused about 44% of companies to lose 10% of their annual revenue, while 75% of organizations have faced customer loss that could have been avoided through better data.</p><p><strong>CRM data enrichment</strong> — i.e., improving, cleaning, and refining your CRM data using tools — is the best way to address such data quality issues and usability roadblocks. However, enriching highly qualitative and unstructured information, such as customer feedback, and deriving insights from it can be challenging. That is where <strong>customer feedback analysis</strong> gains significance as a powerful information enrichment tool.</p><p>So, what exactly is customer feedback analysis? How does it help you make the most of your CRM data? How do you analyze feedback from customers — should you do it manually, or are there any tools you can use to automate the process? If there are tools, how do you choose the right one?</p><p>Read on to find answers to all these questions and more. In this post, we talk about how to use customer feedback analysis for <strong>data enrichment</strong>.</p><h3>What Is Customer Feedback Analysis?</h3><p>Simply put, it is the process in which companies break down customer feedback using tools and turn it into meaningful, actionable, and useful insights. It opens a direct doorway to customers — you can make better decisions based on what you learn from them, offer them products and services they want, and improve the overall customer journey.</p><p>Let’s try and understand it further. <a href="https://go.odaseva.com/web-resource-forrester-2021-state-of-enterprise-crm-data-management-lp">Studies</a> show that using the insights obtained from customer data for decision-making is a priority for more than 85% of businesses. And, what is one of the largest pools of customer information readily available to them? Yes, you guessed it right, feedback from customers.</p><p>You receive customer feedback in different formats, such as text, speech, videos, graphics, images, etc., through multiple channels. These include:</p><ul><li>Support conversations</li><li>SMS and email</li><li>Feedback forms</li><li>In-app and on-site surveys</li><li>Customer reviews on third-party websites</li><li>Social media posts</li><li>Live chats</li></ul><p>Despite being abundant, customer reviews might not be meaningful or usable in its raw form. Imagine that you are running an eCommerce company. The product feedback you receive from customers through calls, chats, and online reviews will be in cluttered language or informal speech, making it hard to decipher the emotions, intentions, and sentiments behind it.</p><p>In other words, though you have large volumes of unstructured, qualitative feedback data in your CRM, you lack the methods or tools to analyze and process these data sets for insights and patterns. Customer feedback analysis is one such knowledge-structuring tool to extract relevant details and gain usable knowledge from customer reviews collected in different formats.</p><h3>Why Is CRM Data Enrichment from Customer Feedback Analysis Important?</h3><p>As we have already said, customer content in its original form — natural speech or text — would rarely give you any actionable insights. Why? Here are a few reasons.</p><p>A group of customers who give the same feedback about a product or service might do it in strikingly different ways. Check out the examples below:</p><ul><li>Customer 1: <em>Your payment process is too complicated.</em></li><li>Customer 2: <em>It took me ages to complete the payment.</em></li><li>Customer 3: <em>Your payment page is difficult to navigate.</em></li><li>Customer 4: <em>I have never seen a payment page as non-intuitive as yours.</em></li></ul><p>Here, all the customers point out the same issue and express the same sentiment — frustration. However, their choice of words and syntax vary, which makes it hard to extract useful information from such feedback.</p><p>Customer feedback is qualitative, and hence its quality differs. Some customers might give you specific and instantly usable feedback, while others offer generic, vague, and unusable responses. Here are two examples:</p><ul><li>Example 1: <em>Your checkout process was a nightmare.</em></li><li>Example 2: <em>I couldn’t add more than three products to the cart. The coupon code was also not working. Then again, I couldn’t use credits from previous purchases. I am disappointed!</em></li></ul><p>Here, the first response doesn’t add any value to your analysis, while the second feedback gives you relevant information on the issues or glitches in your system. Sorting loads of such data sets and identifying which are insightful and which are not can be a laborious task.</p><p>Customers might choose to leave their feedback in different languages. For instance, a France-based company is likely to get customer feedback in both French and English. To make it usable, you must first translate the text or speech from one language to another.</p><p>When your business has a presence on multiple platforms, you receive customer feedback through more than one channel. While some of these channels, such as support calls, are speech-based, others are text-based. Collating and streamlining feedback data spread across channels can be tricky unless you use tools designed for that.</p><p>Customers don’t thematically structure their feedback. A customer giving feedback might talk about billing issues, product quality, and refund issues in the same call or message. Without extracting these themes/topics and labeling them, you cannot take any action for improvement.</p><p>Text-based feedback, such as social media posts, chats, and online reviews, is often cluttered. Typos, emojis, structural mistakes, grammar errors, repetitive content, improper use of punctuations and cases, usage of slang, presence of URLs, stop words, and figures of speech make it complex to understand. Such information in your CRM database needs cleaning before you can use it.</p><h3>How Does Customer Feedback Analysis Work?</h3><p>The process of customer feedback analysis makes large and cluttered bodies of information usable by categorizing them into insightful and non-insightful data. There was a time when companies used to do it manually, but today there are AI-based Natural Language Processing (NLP) tools to automate the whole process.</p><p>Such a tool comes with a suite of analytical tools that help identify key data patterns and trends from qualitative and unstructured feedback. The information and knowledge thus extracted automatically enrich your CRM data during the process.</p><p>Listed below are some of the key tasks that an AI-based feedback analysis can automate for you:</p><ul><li>Data collection from various sources and channels</li><li>Data cleaning</li><li>Text categorization</li><li>Machine-based translation</li><li>Survey analysis</li><li>Sentiment analysis</li><li>Speech-to-text and text-to-speech conversion</li><li>Keyword analysis</li><li>Topic analysis</li><li>Contextual extraction of texts</li><li>Synopsis generation for large texts</li><li>Dictionary and vocabulary building</li></ul><h3>Benefits of CRM Data Enrichment from Customer Feedback Analysis</h3><p>The primary objective of customer feedback analysis is to pull out usable and granular-level insights from a largely unstructured data set. You can use the insights thus gained to enhance your operations, sales, and revenue. Here are some of the ways you can benefit from the process:</p><p>Performing CRM data enrichment through feedback analysis, through surveys for example, is a sure-shot way to improve the quality of your products and services. An enriched CRM database will indicate bugs that need immediate fixing, features and functionalities that need improvement, and features that are missing from your product or service suite.</p><p>Customer expectations regarding personalization are soaring high, creating pressure on businesses to offer curated experiences for users. According to a <a href="https://www.mckinsey.com/capabilities/growth-marketing-and-sales/our-insights/the-value-of-getting-personalization-right-or-wrong-is-multiplying">2021 report by Mckinsey</a>, 71% of customers expect personalized experiences based on their past interactions, while 76% of them feel frustrated when brands do not offer them.</p><p>Customer feedback analysis is a great way to enrich your CRM data for personalization. It helps create comprehensive and updated customer profiles based on which you can offer curated experiences and differentiate your company in a highly competitive market.</p><p>Sentiment analysis, an integral part of customer feedback analysis, opens up opportunities for your sales team. You can optimize your sales strategies based on the insights it gives and identify the right time to sell a product or service. AI-based analysis indicates whether a customer is in a good mood to buy, thus allowing your sales team to nurture leads and pitch a product or service at the right time.</p><p>Customer service matters; quality customer service matters more. <a href="https://d1eipm3vz40hy0.cloudfront.net/pdf/cxtrends/cx-trends-2020-full-report.pdf">A report by Zendesk</a> reveals that nearly 50% of customers are likely to switch to a competitor after one bad experience from a business, and the number shoots up to a staggering 80% with more than one bad experience.</p><p>Analyzing the feedback, surveys and tone of a customer will equip support agents to adopt suitable strategies that increase customer satisfaction. For instance, if you notice that customers are unhappy with delayed deliveries, you can offer them a promo code for the next purchase as a gesture. Likewise, for a customer who is happy with a purchase, you could try upselling products that further enhance their journey.</p><p>All businesses must track a few metrics to understand what customers feel and experience about their offerings and brands. Net Promoter Score (NPS), Customer Satisfaction Score (CSAT), and Customer Health Score (CHS) are among the most crucial metrics. Analyzing feedback is a great way to measure and track these metrics and improve your offerings based on them.</p><p>{{line}}</p><h3>How to Choose a Feedback Analysis Tool for CRM Data Enrichment?</h3><p>There is a good range of market-leading tools for customer feedback analysis, each with different features, functionalities, and pricing. How do you choose one that is worth your investment? Ask yourself these five questions before selecting a tool:</p><p>The tool you choose must suit the size of your company, the volume of feedback, the scale of operations, and your budget. Standard third-party tools are ideal for SMEs and enterprise-level businesses that want to understand large volumes of information. If you are looking for a tool that automates the process end to end and offers qualitative insights, an AI-powered analysis tool is your best choice.</p><p>Invest in a tool that offers all essential features and functionalities, such as text collection, cleaning, labeling, vocabulary customization, and sentiment analysis. Features like multi-language translation and transcription are ideal for companies that collect feedback in multiple languages and through speech-based and text-based channels.</p><p>Not all platforms analyze content the same way. While some analyze only Net Promoter Scores (NPS), others analyze text-based feedback, like surveys. Standard third-party solutions often don’t cover speech-based feedback like support conversations. Opting for tools that support all communication channels is always a great idea. That way, you can enjoy scalability and flexibility. It also makes spotting trends and identifying patterns across channels easier.</p><p>Some tools do broader-level analysis, while others do granular-level analytics. Both help you in decision-making, but granular analytics platforms don’t require you to do any manual tasks to understand the insights. AI-powered tools with sentiment analysis are also a good choice for gaining accurate and granular insights into customer mindsets.</p><p>It is a no-brainer, right? You are investing in an analytics software for CRM data enrichment, so what use is it if the tool does not offer CRM integration? Choose a tool that allows easy setup and integration with your CRM, text analytics tools, and other APIs and management software.</p><h3>Best 5 Customer Feedback Analysis Software for Your Business</h3><p>While choosing feedback analysis software for your company, the best approach is to do your research, shortlist a few, and book a demo to identify the one that suits your business. To help you save time, we have shortlisted five industry-leading customer analysis tools.</p><p><a href="https://www.lettria.com/">Lettria</a> is a no-code AI platform that allows companies to make the most of their textual data through Natural Language Processing (NLP) automation. It requires minimal configuration and offers an intuitive user experience and easy integration with CRMs. Users can use the platform to collect textual data of all kinds and formats from multiple channels and analyze it, like as an NPS for example. The platform runs a gamut of powerful features, ranging from text collection to cleaning to annotation to vocabulary management.</p><p><strong>Pricing and plans: </strong>Customizable plans are available at competitive pricing. <a href="https://www.lettria.com/pricing">Click here to get a quote</a>.</p><p><a href="https://www.glanceable.io/">Glanceable</a>, a B2B SaaS solution for customer review analysis, is ideal for SMEs and large businesses. It consolidates all your customer review information into a single platform and generates AI-powered insights that you can use to drive sales and revenue. They offer solutions and products for brand perception, thematic analysis, survey generation, review management, and internal benchmarking, to name a few.</p><p><strong>Pricing and plans: </strong>Plans start at €499. To know more about them, <a href="https://www.glanceable.io/plan-and-pricing">click here</a>.</p><p><a href="https://www.qualtrics.com/au/">Qualtrics</a> is a customer experience management (CXM) platform that lets you gather customer feedback information through the digital journey and surveys and consolidate it into a single platform. The tools offered by the company help you understand the collected information from surveys for usable insights and improve your customer journey. They also have product, employee, NPS, and brand experience solutions, which makes it a good choice for businesses that need a 360-degree experience management solution.</p><p><strong>Pricing and plans: </strong>Subscription-based customizable plans are available. <a href="https://www.qualtrics.com/au/contact/">Contact the sales team</a> for more info.</p><p><a href="https://www.ibm.com/in-en/watson">IBM Watson Studio</a> is a leading machine learning and data science platform for AI-powered businesses. It offers a suite of data science tools for users to import surveys and analyze data sets, including visual data, using several analytical techniques. From NLP services for advanced text analytics to speech-to-text and text-to-speech tools to language translators to natural language classifiers, they offer a range of services and products to get knowledge from user data.</p><p><strong>Pricing and plans</strong>: Multiple licensing options, pay-as-you-go, and all-in-one packages are available. <a href="https://www.ibm.com/cloud/watson-studio/pricing">Click here</a> for more details.</p><p><a href="https://www.chattermill.com/">Chattermill</a> is a Unified Customer Intelligence platform to unify and analyze customer feedback information, including support and conversation data, obtained from across multiple channels like surveys, etc. It uses deep learning AI for customer review analysis, detects sentiments, and offers unbiased and result-oriented insights. The salient features of the user platform include feedback collection, content aggregation, tagging, thematic analysis, data visualization, sentiment analysis, NPS, and more.</p><p><strong>Pricing and plans: </strong>Free and subscription-based plans are available. <a href="https://www.chattermill.com/pricing">Click here</a> for more details.</p><h3>Start Enriching Your CRM Data Today</h3><p>Your CRM database is a goldmine, but if not enriched using the proper tools, it can become a minefield that harms your sales, revenue, and reputation. Why suffer from data usability roadblocks when you can get insights with ease and accuracy using a customer feedback analysis tool?</p><p>Whether you are a start-up without a technical team or an established company with teams of experts, you can leverage the power of such AI-enabled analysis platforms. Invest in a customer feedback analysis tool like <a href="https://www.lettria.com/">Lettria</a> and harness the power of data to stay ahead of the game.</p><p>Learn more about <a href="https://www.lettria.com/">Lettria</a></p><p>‍</p><p><em>Originally published at </em><a href="https://www.lettria.com/articles/how-to-enrich-your-crm-with-customer-feedback-analysis"><em>https://www.lettria.com</em></a><em>.</em></p><img src="https://medium.com/_/stat?event=post.clientViewed&referrerSource=full_rss&postId=6dde4e2d7a56" width="1" height="1" alt="">]]></content:encoded>
        </item>
        <item>
            <title><![CDATA[What Can Social Media Sentiment Analysis Tell You About Your Customers? — Lettria]]></title>
            <link>https://lettria.medium.com/what-can-social-media-sentiment-analysis-tell-you-about-your-customers-lettria-3980801b1138?source=rss-4f5319276610------2</link>
            <guid isPermaLink="false">https://medium.com/p/3980801b1138</guid>
            <dc:creator><![CDATA[LETTRIA]]></dc:creator>
            <pubDate>Mon, 23 Jan 2023 18:20:19 GMT</pubDate>
            <atom:updated>2023-01-23T18:20:19.917Z</atom:updated>
            <content:encoded><![CDATA[<h3>What Can Social Media Sentiment Analysis Tell You About Your Customers? — Lettria</h3><p>They say your brand is what your customers say it is. However, with social media becoming one of the vital marketing platforms for businesses of all types and sizes, this mantra has slightly changed — your brand is what your customers say it is on social media platforms.</p><p>All businesses that want to improve their social media presence and brand image must analyze and decode what customers say and understand their opinions, emotions, and attitudes. Guess what the best way to do it is? Social media sentiment analysis.</p><p>In this article, we will take a detailed look at social media sentiment analysis — what it is, what it says about your customers and your brand, how to extract sentiment from social media conversations, how to use the data you obtain from it, and more.</p><h3>What Is Social Media Sentiment Analysis?</h3><p>Social media sentiment analysis, also called opinion mining, is a type of sentiment analysis in which you collect and analyze the information available on various social platforms to learn how people perceive your brand, products, or services. It examines social conversations and user-generated content, such as brand mentions, comments, and posts, to reveal the feeling within them.</p><p>The process is much more than tracking quantitative metrics, which merely measure the number of comments, reactions, and reviews. Social media sentiment analysis uses Natural Language Processing (NLP) and machine learning to decipher a text and interpret its context, thus revealing whether a post or comment is positive, negative, or neutral.</p><h3>Why Should Businesses Need Social Media Sentiment Analysis?</h3><p>According to feedback <a href="https://www.renolon.com/social-media-for-business-statistics/">surveys</a>, about 71% of SMEs use social media for sales and marketing purposes. While businesses interact with users through posts, users respond to them through likes, shares, brand mentions, reviews, and the like. <a href="https://statusbrew.com/insights/social-media-statistics/">Statistics</a> reveal that more than 200 million users visit at least one business profile per pay, and about 80% of users have tweeted something about a brand at least once.</p><p>Almost all social platforms let you measure and track crucial metrics like reach, impressions, audience growth rate, video view rate, virality rate, click-through rate, etc. These metrics help you assess how strong your social presence is and how good your marketing strategies are.</p><p>That said, such metrics make brands obsessed with quantitative data, such as the number of mentions. They overlook the emotions, opinions, attitudes, and feelings comments and brand mentions carry. It is one thing to count the likes, brand mentions, and posts, but understanding the sentiment behind them is a different game altogether.</p><p>That is why businesses need sentiment analysis. It tells you what quantitative metrics do not reveal — how social media users perceive your business and your products/services.</p><h3>What Exactly Does Social Media Sentiment Analysis Do?</h3><p>All social conversations wear one sentiment or the other — either there is a positive sentiment, a negative sentiment, or an absence of both. Sentiment analysis examines a social conversation, such as a comment or a mention, and identifies its polarity — whether it contains a positive, negative, or neutral sentiment.</p><p>Depending on how advanced and feature-rich the tool you use, it might also identify the context, the exact emotion expressed by the customer, and the specific aspect to which the sentiment is attached.</p><p>Confusing? <a href="https://twitter.com/on_perspectives/status/1596187322248216576">Here</a> is an example to help you understand the process better:</p><figure><img alt="" src="https://cdn-images-1.medium.com/max/929/0*NpXM8iQovTE17Fsk.jpeg" /></figure><p>A sentiment analysis platform will use machine learning and NLP techniques to examine the text, its context, and terms and modifiers like wrong, mess, etc., and identify it as a negative comment. It will further analyze it to detect the underlying emotion — disappointment or frustration, to be used for feedback.</p><p>That is not all. Here, the customer seems to have problems with product pictures and descriptions. Advanced analytics platforms will also identify those specific aspects with which the customer has a negative sentiment.</p><p>Want to know more about how it works? Check out our blog — <a href="https://www.lettria.com/articles/what-is-sentiment-analysis-and-how-does-it-work#how-does-lettria-handle-sentiment-analysis">What Is Sentiment Analysis and How It Works.</a></p><h3>How To Do Social Media Sentiment Analysis?</h3><p>Though you can do manual analysis, it is a cumbersome task that consumes your time, energy, and resources. The good news is that you can automate the whole process with a sentiment analysis platform that uses Natural Language Processing (NLP) and Artificial Intelligence (AI) to tap into customer emotions.</p><p>The process of sentiment analysis may vary slightly from one tool to another. However, it typically involves a few steps.</p><h3>Step 1: Data gathering</h3><p>The first step is to gather data from various social platforms using APIs, scraping tools, and other methods. Most tools look for posts of your brand or business over a specific timeframe and collate them for analyzing.</p><h3>Step 2: Text cleaning</h3><p>The collected data is often noisy, unstructured, and cluttered. To make it suitable for analyzing, you must clean it first by removing punctuations, stop words, repetitive texts, etc. Text cleaning generally comes as a feature of many sentiment analytics platforms.</p><h3>Step 3: Data analysis</h3><p>The third and most crucial stage involves analyzing the collected and cleaned data using sentiment analysis algorithms. The focus of the analysis may vary depending on your business requirements and the tools you use.</p><ul><li><strong>Standard sentiment analysis</strong> involves classifying the data based on polarity — whether a text is good, negative, or neutral.</li><li><strong>Fine-grained or graded sentiment analysis tools</strong> will allow you to categorize the content more precisely, such as:</li><li><strong>Sentiment detection</strong> is also a feature in many advanced analytics tools. It helps detect emotions — anger, frustration, happiness, sadness, anticipation, etc.- within texts.</li><li><strong>Aspect-based sentiment analysis</strong> goes one step further. It assigns the identified feeling to certain pre-defined aspects, features, and topics. For instance, if a customer says, <em>I loved the features of the phone, but it is very pricey</em>, the tool will assign a good sentiment to phone features and a negative one to the price.</li></ul><h3>Step 4: Data visualization</h3><p>Once the tool analyzes data and detects sentiments/polarity, it turns the results into graphs and charts that are easy to understand. The reports will show you the surges in good sentiments (called peaks) and surges in negative sentiments (called valleys) within the timeframe you have initially chosen. Most tools will also show the results based on overall sentiment, sentiment over time, sentiment by rating, and sentiment by topic.</p><h3>Benefits Of Social Media Sentiment Analysis</h3><p>More often than not, brand mentions and comments are forms of honest feedback. It offers valuable insights into how people interact with your products/services and creates opportunities to manage your reputation. Analyzing social conversations and the feelings within them helps your business in more ways than one. It helps you:</p><h3>Engage with customers the right way</h3><p><a href="https://www.brightlocal.com/research/local-consumer-review-survey/">Surveys</a> show that 89% of users choose a business that responds to all their reviews and comments. However, many brands make the mistake of only responding to negative posts. They pay little or no attention to happy or neutral posts.</p><ul><li>Thank and appreciate a customer who left a positive comment</li><li>Address a negative comment using the right support team</li><li>Respond to a neutral comment in such a way that you can further engage the customer</li></ul><p>Social media sentiment analysis is a great tool to help you understand the nature of the comment and create an engagement strategy based on that. For example, you can:</p><h3>Understand customer trends</h3><p>Customer trends change over time, especially during specific periods, such as holiday seasons or crises like the COVID-19 pandemic, climate change, or recession. For instance, <a href="https://www.simon-kucher.com/sites/default/files/studies/Simon-Kucher_Global_Sustainability_Study_2021.pdf">reports</a> indicate that about 85% of clients have become greener in their purchasing choices. Likewise, <a href="https://www.agilitypr.com/pr-news/public-relations/coronavirus-crisis-consumer-impact-new-data-on-purchase-decisions-and-behaviors/">surveys</a> also show that the coronavirus pandemic has led people to spend more on fitness equipment and health supplements and reduce their spending on big-ticket items.</p><p>To offer your products or services according to the latest customer trend, you must learn and keep up with these changes. Social media sentiment analysis shows what customers consider a priority during a particular time and helps you change your marketing and content strategies accordingly.</p><h3>Personalize customer experiences</h3><p><a href="https://www.mckinsey.com/capabilities/growth-marketing-and-sales/our-insights/the-value-of-getting-personalization-right-or-wrong-is-multiplying">According to the latest statistics</a>, personalization can remarkably increase the likelihood of purchases, repurchases, and recommendations. About 76% of customers are more likely to purchase products or services from businesses that personalize the experience, while 78% are likely to recommend or repurchase.</p><p>Tracking social media data using sentiment analysis gives you insights that you can use to personalize your social media marketing strategy, content curation strategy, marketing campaigns, and retargeting ads.</p><h3>Offer better customer support</h3><p>Leaving comments or brand mentions to draw the attention of businesses has become the new normal. <a href="https://optinmonster.com/social-selling-statistics/">Statistics</a> reveal that about 33% of users prefer social media over phones to contact brands. <a href="https://www.convinceandconvert.com/social-media/42-percent-of-consumers-complaining-in-social-media-expect-60-minute-response-time/">Surveys</a> further show that 32% of people who reach out to businesses through social media expect a response within 30 minutes, while 42% expect a response within 60 minutes.</p><p>Social media sentiment analyzing is a great way to increase the quality of your customer service functions by capturing the pain points. For starters, it alerts your customer support teams regarding the issues clients face and helps react to those issues. Your teams can also use insights from sentiment analysis to proactively reach out to users, which enhances your brand reputation by several notches.</p><h3>Detect and manage a crisis</h3><p>Often, an influx of negative posts may indicate a brand or PR crisis. Monitoring your comments using a sentiment analysis platform is a great way to detect them early and manage them before it gets too late. The more time you take to respond, the more media coverage the issue will get and the more it will escalate.</p><p>Experts often cite the United Airlines crisis as a case in point. In 2017, videos surfaced on social media showing security officers removing a passenger violently from one of the planes of United Airlines due to overbooking. It had a snowball effect — the airline was mentioned 2.9 million times on social media, while their positive mentions declined from 90% to 30%. Due to their initial response, which seemed to justify the act, the company continued to receive backlash for the next 48 hours until they issued a proper apology.</p><p>Sentiment analysis can help brands in such situations to quickly assess the feeling of the public and come up with responses that ease the backlash.</p><h3>Tweak your content and products/services</h3><p>Comments are a treasure box that reveals whether your social media content is engagement-worthy. A sentiment analysis tool would show the spikes in your positive, negative, and neutral sentiments. It tells you what makes a customer tick, and you can use this insight to curate content and social posts that elicit more happy responses.</p><p>Likewise, the polarity of the sentiment may also reveal interesting facts about how user-friendly or functional your product or service is. Based on this, you can improve your services and products and change the game. More often than not, such data would also give ideas for developing new products or services.</p><h3>Understand the competition</h3><p>You can use social media sentiment analysis platforms to monitor your brand; you can also use them to monitor your competitors. It will give you valuable insights into what customers like or dislike about the products and services offered by your competitors and shape your marketing and content strategies accordingly.</p><h3>Use positive user-generated content for ads</h3><p>User-Generated Content (UGC), such as brand mentions, reviews, and posts, is also a great way to advertise your products and services. Once you have identified posts with positive sentiment, you can reshare them or incorporate them into your advertising strategy.</p><p><a href="https://www.nosto.com/blog/report-consumer-marketing-perspectives-on-content-in-the-digital-age/">Statistics</a> reveal that consumers consider such UGC to be 2.4 times more authentic than brand-created content. Further, if you feature UGC in your ads, they will likely fetch 73% more positive engagement on social platforms than traditional ads.</p><h3>Attract more customers</h3><p>According to the <a href="https://www.hootsuite.com/resources/digital-trends-q3-update">Global State of Digital 2020 report</a>, about 75% of internet users search for brands using social media. It means that a mention or comment you receive is not just a mention or comment — it is a de facto review that might help another customer to make a purchase decision. You would not want negative mentions to come in the way of your brand and clients, would you?</p><p>Social media sentiment analysis helps you identify negative comments and address them before issues escalate. Likewise, you can use happy comments and mentions to endorse your brand image. That way, you will come across as a proactive brand for a new customer who skims through your social media handles. It gives them the confidence to purchase your products and services.</p><h3>Social Media Sentiment Analysis Examples — Texts and Analysis</h3><figure><img alt="" src="https://cdn-images-1.medium.com/max/890/0*Igt-XkZxFI6znre4.jpeg" /></figure><p>Here are a few real-life examples that will help you see how social media sentiment analysis works:</p><figure><img alt="" src="https://cdn-images-1.medium.com/max/881/0*SLMWJNPaJ5RmzUTA.jpeg" /></figure><p>The comment and mention above show that the customer is happy with the product. A sentiment analysis tool will extract adjectives and modifiers like excellent and loved to indicate that the feeling is positive.</p><figure><img alt="" src="https://cdn-images-1.medium.com/max/748/0*9HdM937ecickOULL.jpeg" /></figure><p>Here, as you can see, the customer expresses two feelings — happiness about the quality of the product and disappointment about the delivery time taken. Doing an aspect-based sentiment analysis will assign two sentiments to these two aspects. It will enable the customer support person to address the issue, which is late delivery.</p><figure><img alt="" src="https://cdn-images-1.medium.com/max/853/0*ltmdC_v3WJeBJ4u7.jpeg" /></figure><p>This post also contains positive and negative feelings. It tells the brand that the multiple playlists option was one of their best features (positive), but the customer is sad that it is no longer available (negative). By performing sentiment analysis of such mentions, brands can understand the pulse of customers and tweak their product/service offerings.</p><figure><img alt="" src="https://cdn-images-1.medium.com/max/820/0*yMn2OLVY3DrQqzyo.jpeg" /></figure><p>This comment/mention shows how the language used on social media can be complex — though the customer uses the word congratulate (positive sentiment), it is sarcasm. The text also uses strong negative words such as worst, badly, incompetency, and ignorance. An AI-enabled sentiment analysis platform will place such texts into context and extract underlying feelings.</p><p>Here, the customer expresses neither positive nor negative sentiments. Instead, they raise a question regarding the real-life, practical usability of the product. In other words, the feeling of the post is somewhat neutral.</p><p>However, it still gives the brand an opportunity to step in and engage the customer by explaining more about the product or showing them how the product is a good choice due to XYZ features.</p><h3>Start Analyzing Social Media Sentiments Today with Lettria</h3><p>All brands want to know what customers think and feel about them — if there were a magic wand that would open a direct doorway to customers, all organizations would go for it. Well, there is no such wand, but there are social media feedback and sentiment analysis platforms like Lettria.</p><p>The sentiment and customer feedback analysis platform we offer enables you to enhance your social media presence and engage with your clients more meaningfully. With a no-code AI platform that uses NLP automation, we analyze your textual data and feedback with the utmost accuracy. Our platform is ideal for businesses of all types and sizes, even organizations without a dedicated technical team, data scientists, or developers.</p><p>If you are a business that wants to make the most of your social media comments and feedback, now is the time, and <a href="https://www.lettria.com/">Lettria</a> is the right choice.</p><p>‍</p><p><em>Originally published at </em><a href="https://www.lettria.com/articles/social-media-sentiment-analysis"><em>https://www.lettria.com</em></a><em>.</em></p><img src="https://medium.com/_/stat?event=post.clientViewed&referrerSource=full_rss&postId=3980801b1138" width="1" height="1" alt="">]]></content:encoded>
        </item>
        <item>
            <title><![CDATA[Perform Market Research And Competitive Analysis Using Sentiment Analysis — Lettria]]></title>
            <link>https://lettria.medium.com/perform-market-research-and-competitive-analysis-using-sentiment-analysis-lettria-b32bcb49cd28?source=rss-4f5319276610------2</link>
            <guid isPermaLink="false">https://medium.com/p/b32bcb49cd28</guid>
            <dc:creator><![CDATA[LETTRIA]]></dc:creator>
            <pubDate>Mon, 23 Jan 2023 18:18:42 GMT</pubDate>
            <atom:updated>2023-01-23T18:18:42.961Z</atom:updated>
            <content:encoded><![CDATA[<h3>Perform Market Research And Competitive Analysis Using Sentiment Analysis — Lettria</h3><figure><img alt="" src="https://cdn-images-1.medium.com/max/1024/0*0ohA92ZaXDfCnSmJ.png" /></figure><p>Running a business and finding success is all about making informed decisions. For this, you need to research the market, your customer’s expectations, your competitors, their strengths and weaknesses, and how your brand compares to them. The average business spends more than <a href="https://www.referenceforbusiness.com/small/Mail-Op/Market-Research.html">25%</a> of its annual marketing budget on research — that’s how important this is.</p><p>To know your audience and your competitors, you need to answer questions like — what’s available in the market? What does the customer expect? Who are your competitors? And so on. Answering these will help you create the best strategies for growth and take the competitive advantage.</p><p>The answers aren’t very hard to find. The truth is, customers aren’t shy about sharing their opinions. If they like or dislike your product, they’ll say so on online platforms. A <a href="https://www.notifyvisitors.com/blog/customer-feedback-survey-statistics/">survey</a> found that 85% of customers share feedback for good experiences and 81% give feedback when they are upset with the experience.</p><p>If you ask, they’ll even give you direct feedback as a review. Analyzing the sentiments of these reviews and interactions or Sentiment Analysis as it is better known plays an instrumental role in researching the market and the existing competition.</p><p>Let’s take two statements for example;</p><p>“The dress fits excellently”</p><p>“I found a cheaper option”</p><p>The first tells you that your customer is happy. It also tells you that ‘fit’ is an important purchasing consideration. The second not only says that the customer is unhappy with your service but also that your competitors are offering better prices.</p><p>Let’s take a closer look at what Sentiment Analysis is and how it can help with your marketing and competitor analysis.</p><h3>What Is Sentiment Analysis?</h3><p>As the name suggests, <a href="https://www.lettria.com/articles/social-media-sentiment-analysis">Sentiment Analysis</a> refers to understanding and analyzing your customer’s sentiments. Relying only on traditional metrics like the number of views, shares, clicks and comments is no longer enough. They give you an incomplete picture. You need to read each comment and understand what your customers are saying and why to improve your business strategies.</p><p>While you can read 20 reviews a day and analyze them manually, you cannot track brand mentions in real-time or make sure you know everything your customers are saying by relying on your efforts alone. Sentiment Analysis automates this analysis with Artificial Intelligence (AI) like that provided by Lettria which uses Natural Language Processing (NLP) and Machine Learning (ML).</p><p>It starts by analyzing text to understand whether the basic emotion expressed is positive, negative or neutral by giving it a score from -1 to 1. Here -1 to 0 represents negative emotions, 0 reflects a neutral opinion and 0 to +1 represents a positive emotion. Depending on the tool you use, it can further break down the expression into other emotions such as surprise, frustration, confidence, excitement, embarrassment, love, etc. It also grades the intensity of these emotions to help you prioritize actions.</p><p>Sentiment Analysis not only makes you aware of what more customers are saying about your brand, but it also gives you a real-time view of your relationships with them.</p><h3>How To Perform Sentiment Analysis For Market And Competitor Research?</h3><p>AI platforms for Sentiment Analysis automate information extraction and analysis for market and competitor research. The overall process followed is fairly straightforward.</p><h3>Data Collection</h3><p>The first step is to scan the world wide web for instances where you and your competitor companies are mentioned. This could be on social media pages, blogs, discussion boards, online forums, review sites, e-commerce websites, etc.</p><h3>Extraction</h3><p>The phrases giving the customer’s comments and opinions about you and your competitors and data must then be cleaned and structured. All irrelevant information, noise and non-contextual content are identified and removed.</p><h3>Initial Sentiment Detection</h3><p>The algorithm then scores the relevant adjectives and adverbs as positive, neutral or negative according to a graded sentiment analysis. For example, “I ordered Chinese food” will get a neutral score. “The food is good” and “the food is excellent” will both be graded as positive sentiments with the latter getting a higher score. Similarly, “the food is not worth the price” and “my meal had a fly in it” will both be evaluated as negative with the latter getting a lower score.</p><p>The grading of scores allows brands to prioritize responses. In the above example, a customer’s viewpoint about pricing is secondary when someone else talks of food not meeting basic quality standards.</p><p>Having these scores for both you, and your competitors’ companies helps you identify in which areas you are stronger or weaker than each competitor. You can then easily perform a SWOT analysis, where you identify your strengths, weaknesses, opportunities, and threats. You can identify how to improve your services, how to adjust your pricing, how to improve features and strategy, to ultimately boost your sales. <a href="https://blog.hubspot.com/marketing/swot-analysis">Here</a> ‘s more information on SWOT analysis along with template examples.</p><h3>Contextual Analysis</h3><p>Customers don’t always make clear, short statements. They may use slang or be sarcastic. They may have had a good experience overall but some parts of it may be unsatisfactory. Hence, the emotions identified must be put in context. Let’s take a few examples.</p><p>“This was my second visit but I’m never coming back”</p><p>“We had a smashing time!”</p><p>“Paid for dinner, got a stomach ache for free”</p><p>The first statement starts on a positive note but the second half contradicts the statement to leave it with an overall negative grade. In the second case, ‘smashing’ which seems like a negative word is actually slang for the opposite.</p><p>The third statement is probably the most complex. It looks like the customer had a positive experience but, the sarcastic tone gives the statement a negative grade.</p><p>Thus, contextual analysis is an important part of identifying the right sentiment. Imagine the beating your reputation would take, if the company responded by saying, “Thank you. We hope to see you again” to the first or third statement!</p><h3>Sentiment Analysis And Market Research</h3><p>When you talk of market research, quantitative metrics such as web traffic, number of visitors, social media statistics, market share, etc. are the usual focal points. These numbers are direct indicators of growth and are relatively easy to understand.</p><p>However, relying only on these metrics may leave you with gaps in customer understanding and your business strategies. You know that a product was bought by 500 people but what made them choose your brand over others — was it the price? The ingredients? the latest marketing campaign?</p><p>You can understand the ‘Why’ when you have qualitative metrics. These metrics are typically not listed in the form of clear-cut numbers and they’re harder to measure. It doesn’t mean the information is inaccessible. You can find raw data for qualitative metrics in the form of customer-generated content on online forums, social media pages and even on your own website.</p><p>With sentiment analysis, you can search for these opinions and derive insights that help you understand your customers and the market.</p><h3>Identify Competitors</h3><p>Whether you’re an entrepreneur launching a new product or a company with an established product, you need to know who your competition is. Business as you know it is constantly evolving along multiple verticals and there’s no telling when or where a new competitor could emerge from.</p><p>Sentiment Analysis gives you inputs into the social buzzwords that have the most value to you. Let’s say your brand markets soaps. Sentiment Analysis can help you find brands associated with keywords like ‘cleansing’, ‘organic’, ‘hand-poured’, ‘rich lathering’, etc.</p><p>Along with the known labels, you could discover smaller, local brands that can grow and challenge your products. And new competitors crop up every day. According to the US Business Formation Statistics (BFS), 2021 alone saw <a href="https://www.oberlo.in/statistics/how-many-new-businesses-start-each-year">5.4 million</a> new business applications.</p><p>Now, you can track small brands too and the effect their campaigns may have on your audience. Ignoring their presence may cost you in the long run.</p><h3>Get Inspired By What Your Competitors Are Doing Right</h3><p>Sentiment Analysis shows you how customers feel about your competitors by doing an analysis over their social media and reviews. A spike in positive sentiment toward a competitor indicates that they have done something the audience approves of.</p><p>You should know what this is (or what they have stopped doing). The information could trigger a change in your own processes. In turn, this could help maintain brand loyalty and attract more customers.</p><p>For example, let’s say a restaurant saw a sudden spike in positive sentiments after extending its service hours. With this change, they can now cater to customers up to 11 pm while your restaurant serves dinner only till 10 pm. The surge in positive sentiment indicates that this action has made the restaurant accessible to a wider audience. These are customers your restaurant is missing out on. Taking other factors into consideration, you may want to look at changing your restaurant timings too.</p><h3>Know What Not To Do</h3><p>Every change you make to your product, your customer service, your branding, etc. costs money. So, before making a major change, you would probably conduct a series of tests. Let’s say you’re introducing a new flavor of ice cream. You can see how your competitors have fared with similar flavors. If there’s a negative customer sentiment associated with the flavor, maybe, your money is better spent exploring something else. With Sentiment Analysis, you lower your risk and increase your ROI.</p><h3>Discover What Matters To Your Market</h3><p>Whether it’s shoes or flight tickets, your customer will take many factors into consideration before swiping their card. Your brand’s success depends on how well you know your customers and the way they search and compare their options. Is price the most important consideration? Would your customer be willing to pay more for better quality? How much is their purchase influenced by the people endorsing your product?</p><p>Your customers have already given you the answers in their reviews and posts. With Sentiment Analysis, you can extract the relevant information and find your answers. Let’s say a customer said, “the dress fits well but it’s too expensive” — price is definitely a key consideration here and offering discounts may help you attract more customers.</p><h3>Setting Your Brand Apart From The Rest</h3><p>Today, maintaining the uniqueness of your product is a challenge all brands face. A product you spent years researching and developing can be copied in weeks. What matters more today is the way you treat your customers and how you position your brand.</p><p>Let’s take women’s empowerment as an example. Sentiment Analysis will show that this is a keyword all brands need to pay attention to. Just saying that you are an equal-opportunities brand doesn’t do much. You need to prove it by being at the forefront of conversations around it.</p><p>Anyone can market “products made by women” but can they all claim to market “products made by differently-abled women”? This could gain more traction by sharing part of the revenue earned with NGOs working towards women’s empowerment, for example. Sentiment Analysis can help you monitor conversations about buzzwords like women’s empowerment and tell you the direction they’re headed in. You can then look at incorporating it into future marketing campaigns and positioning your brand accordingly.</p><h3>Summing It Up</h3><p>Thorough market research and competitor analysis are crucial to giving your brand a shot at winning over the audience. Along with quantitative metrics, you can start understanding sentiments around your product, brand, competition, etc. and get a more complete view of what matters to your customers. By doing this and prioritizing customer experiences, company revenue can grow by <a href="https://www.bain.com/insights/are-you-experienced-infographic/">4–8%.</a></p><p>A top Sentiment Analysis platform like Lettria makes this information accessible and puts it in a form that’s easy to understand. Rather than stop at positive and negative grading, it exacts a value for sentiments and helps you prioritize them. On top of that, it can be used in any industry.</p><p>Lettria follows a 2-step analysis process that takes into account 8 primary emotions and the context in which they are expressed. It offers all the benefits of a standard Natural Language Processing model as well as the ability to customize multilingual models according to your business.</p><p>Meeting customer expectations and setting yourself apart from your competitors is the only way you will be able to get a competitive advantage and meet your sales target. Explore how<a href="https://lettria.com/"> Lettria</a> can help your market research team start competitive analysis right away.</p><p>‍</p><p><em>Originally published at </em><a href="https://www.lettria.com/articles/perform-market-research-and-competitive-analysis-using-sentiment-analysis"><em>https://www.lettria.com</em></a><em>.</em></p><img src="https://medium.com/_/stat?event=post.clientViewed&referrerSource=full_rss&postId=b32bcb49cd28" width="1" height="1" alt="">]]></content:encoded>
        </item>
        <item>
            <title><![CDATA[How to Analyze Survey Data with Emotion Analysis — Lettria]]></title>
            <link>https://lettria.medium.com/how-to-analyze-survey-data-with-emotion-analysis-lettria-a90b0b1b6f5f?source=rss-4f5319276610------2</link>
            <guid isPermaLink="false">https://medium.com/p/a90b0b1b6f5f</guid>
            <category><![CDATA[sentiment-analysis]]></category>
            <category><![CDATA[emotion-analysis]]></category>
            <category><![CDATA[surveys]]></category>
            <category><![CDATA[nlp]]></category>
            <dc:creator><![CDATA[LETTRIA]]></dc:creator>
            <pubDate>Mon, 23 Jan 2023 18:15:15 GMT</pubDate>
            <atom:updated>2023-01-23T18:16:14.656Z</atom:updated>
            <content:encoded><![CDATA[<h3>How to Analyze Survey Data with Emotion Analysis — Lettria</h3><figure><img alt="" src="https://cdn-images-1.medium.com/max/1024/0*xOK276lwDHlsxTih.png" /></figure><p>Making your business a success is all about meeting customer needs. It’s as simple as this — would you give someone a knife when they want a spoon? Hence, you need to know what your customers want from you. Surveys are a great tool to understand their requirements and collect feedback on your brand experience. From designing a new product line to logistics and brand image, surveys play an important role in uncovering consumer insights.</p><h3>Setting up your survey questions</h3><p>Surveys should be designed to collect maximum information with minimal effort by the consumer. They typically contain close-ended and open-ended questions. Close-ended questions like multiple choice, ranking, checkbox or drop-down questions have anticipated results that are easy to quantify.</p><p>Let’s say you’re introducing new colors to a product and ask customers to pick between red and black. The results can be easily analyzed and you get a clear picture of your customer’s choices from the options presented to them.</p><p>But, what if you simply ask respondents to type their preferences? The results are not limited to 2 choices, you could also get variables like pink, green, blue, yellow, and so on as options. This is an example of an open-ended question.</p><p>Responses to open-ended questions may be a word, a phrase or even a paragraph. This type of data is categorized as qualitative data. Qualitative data analysis involves identifying themes, interpreting patterns and determining how these themes and patterns answer your questions.</p><p>With sentiment analysis, you can further transform qualitative results into quantitative results. For example, you could learn that 60% of your survey respondents prefer red while 20% prefer black and 10% prefer green. Let’s dive deeper into emotional sentiment analysis and survey data.</p><h3>The importance of open-ended survey questions and qualitative data</h3><p>Before delving into sentiment analysis for qualitative questions, it’s important to first assess the need for them. Having a survey with only quantitative results would definitely be easier to analyze and interpret. Give respondents the option to rank their preferences for products.</p><p>Create a choice between ‘Yes, I like it’ and ‘no, I don’t like it’. Let them rate their level of satisfaction on a scale of 1 to 10. The responses are statistical and can be easily analyzed to tell you what customers like and what they don’t like.</p><p>However, you might miss out on important details. A simple ranking system would tell you if customers like or dislike your products but it won’t tell you ‘why’ they feel this way. It is only by including questions that require qualitative responses and analyzing the sentiment behind these responses that you can get the answer to ‘why’.</p><p>You might also accidentally omit an option and hence get a skewed response. Take this survey question for example;</p><p>“How did you hear about us?</p><p>By limiting the variables to make this a closed-ended question, other communication channels like word of mouth, print and television advertisements, etc. have been omitted. This can be easily rectified by making it an open-ended question.</p><p>Open-ended questions also put you in a position to learn more from a small sample of respondents. For example, let’s say you introduced a new product and want to survey beta users on how they feel about it. Along with close-ended questions that tell you whether they like it or don’t, including open-ended questions can give them a way to tell you what specific features they think need improvement.</p><h3>What is emotion analysis?</h3><p>Emotion analysis or sentiment analysis is all about understanding the emotions behind a customer’s answers to a survey by using Artificial Intelligence (AI) types of software that do text analytics, Machine Learning (ML) and Natural Language Processing (NLP).</p><p>At its core, emotion analysis categorizes answers as positive, neutral and negative. It also analyzes the themes and subjects respondents are writing about. For each subject, it scores a response from -1 to 0 in the case of negative responses and numbers, 0 to 1, for positive responses. Neutral responses are scored as the number 0. These statistical responses make the interpreted results easier to visualize.</p><p>Sentiment analysis for surveys can help answer questions like:</p><ul><li>Which product has the most positive feedback?</li><li>What do customers like and dislike about a product?</li><li>Is the feedback trending in an overall positive or negative direction?</li><li>Has there been a shift in the nature of responses since last year or last interval?</li></ul><p>The variable of potential answers to these questions are vast.</p><h3>Why is emotion analysis important for survey analysis?</h3><p>Manually reading through every survey response, cross analyzing, and understanding the sentiment behind it is next to impossible. It’s tedious and there are a million better ways you can spend your time. Hence, it is best to automate the task with an emotion analysis tool. Let’s look at some of the benefits of doing so.</p><h3>Accurate customer insights</h3><p>Analyzing statements to understand the emotion behind them manually can quickly become subjective. Two people can look at the same statement and score it differently based on their own experiences. Hence, it isn’t easy to be consistent with manual emotion assessment. However, by automating the task, the same criteria are used to evaluate each statement. Hence, there are no biases.</p><p>Secondly, if you were to complete the task manually, you would probably pick a small sample from the survey answers to base your analysis. But, automating the task allows you to consider all the survey responses. As a result, the customer insights gained are more accurate and reliable.</p><h3>Real-time survey analysis</h3><p>A survey can have as little as 1 question. For example, you may want to include a survey at the end of every order confirmation to find out how the customer rated the experience and what they would like improved.</p><p>Using a sentiment analysis AI tool allows you to assess each survey response in real time. This helps you keep pace with customer expectations and spot issues and opportunities as they arise.</p><p>For example, if you see a spike in disgruntled clients, you know there is something wrong and can look into the sentiment expressed by them to understand where the problem lies.</p><h3>Ability to tackle issues quicker</h3><p>In continuation of the above point, emotion analysis helps you detect negative emotions at an early stage and give you the information required to tackle the issue before it grows bigger. For example, if you notice more than a few survey answers talking negatively about the quality of stitching, you can take a quick look at the product for yourself and make changes if required before you start losing customers.</p><h3>Step-by-Step Approach to Emotion Analysis of Survey Data</h3><p>Today, conducting a survey is easy. You can include survey questions as a final step in the order placement process, send surveys by email, add surveys to landing pages, etc. The first step in emotion analysis of a survey is to collect all the responses in one place. Once you have all the information required, here’s what happens.</p><h3>Thematic analysis</h3><p>Thematic analysis is based on deducing the meaning of the words used by people in submitting their answers. This can be done by identifying repetitive types of themes and patterns that emerge by grouping words and themes. The feedback is then quantified according to the sentiment expressed. Thus, there are 3 main steps involved:</p><h3>Identify what respondents are talking about</h3><p>The first step of analyzing qualitative responses in a survey is to identify the themes being talked about. For example, if you send out a survey to find out how clients feel about a product range, they may talk about price points, colors, material quality, availability, delivery timelines, etc. This will help you identify what is important to your clients.</p><p>Since you can’t manually read through thousands of survey forms, you will need to train the AI model to recognize a list of qualifier words. For example, a theme like price points may be identified by keywords like expensive, value for money, cheap, budget-friendly, etc. Responses can then be categorized into different groups based on this analysis.</p><p>The efficacy and accuracy with which AI software tools can categorize feedback depends on how many keywords are included in the list. Most text-processing platforms maintain a list of keywords and synonyms. In addition to a database of 750,000 words, <a href="https://lettria.com/">Lettria</a> allows you to modify the knowledge base to suit your needs. You can add product or industry-specific keywords as well as technical elements.</p><p>For example, you can add slang terms like ‘spendy’ or phrases like ‘costs an arm and a leg’ to your database. The more keywords you include in your database, the less likely you are to miss out on responses.</p><h3>Analyze how much they are talking about each theme</h3><p>There are numerous themes that may be identified in a survey. You need to identify which themes are being talked about the most to prioritize actions. For example, if people are talking about pricing as well as the quality of material, which aspect do you address first? Of course, you will want to address the point that is being talked about the most. For this, you need to know how many people are talking about each theme.</p><p>Let’s say you find 200 people talking about delivery timelines, 350 talking about prices and 150 talking about the material. Thus, it is safe to say that prices are the most important criterion for the survey takers. This is what you need to tackle first.</p><h3>Identify how they feel about each theme</h3><p>Now that the responses have been grouped into different categories, you need to identify the underlying sentiment correlation with each theme. This step is akin to the first step but with a smaller group of responses.</p><p>Let’s take price as a theme. In this case, broadly speaking, you need to understand whether customers think the product is expensive, cheap or value for money. In this instance, expensive may be considered a negative emotion while cheap and value for money are considered positive. Platforms like <a href="https://lettria.com/">Lettria</a> go beyond the three main emotions to include more complex emotions like anger, curiosity, disapproval, fear, pride, sadness, etc.</p><p>To understand how customers feel about a theme, you must first categorize answers into various emotion groups. Simple statements, like “The product was perfect for my budget” or “too expensive for me” are easy to analyze and quantify. The first falls into the positive category while the second would be categorized as negative.</p><p>However, not all feedback is as clear-cut. Let’s say a customer says, “It’s expensive but worth it.” The word expensive would give the statement a negative emotion score but in the overall context, this would be wrong. This brings us to the next step.</p><h3>Put emotional responses in context</h3><p>To analyze the emotions behind complex statements, the AI model must be able to understand the context in which the sentiment is being expressed. This is one of the advantages experienced by Lettria users. Lettria not only determines a statement’s polarity but also the emotional context within it.</p><p>In the above example, “It’s expensive but worth it”, Lettria would assess expensive as a keyword but still give the statement a positive score. Similarly, some responses may be sarcastic or include slang terms. Your AI model’s ability to recognize them as such and handle textual data accordingly will determine its efficacy. Getting a concise report of all this data is essential.</p><h3>Start analyzing your survey data today</h3><p>Surveys are undoubtedly important for businesses in all industries. By implementing emotion analysis for your survey feedback, you can process all the textual data received and get a more comprehensive, accurate view of your customers needs.</p><p>Emotion analysis platforms like <a href="https://lettria.com/">Lettria</a> ‘s software help quantify the qualitative responses and present them in a form that is easy to understand and derive value from. Though the NLP model is advanced with an expansive dataset, you can add keywords and synonyms even if you don’t have any technical knowledge.</p><p>By allowing technical and non-technical profiles to contribute and collaborate on emotion analysis projects, Lettria gives you more meaningful results through a report that can be immediately put to use. Remember, the more you listen to your customers, the better you will be able to serve them.</p><p>‍</p><p><em>Originally published at </em><a href="https://www.lettria.com/articles/how-to-analyze-survey-data-with-emotion-analysis"><em>https://www.lettria.com</em></a><em>.</em></p><img src="https://medium.com/_/stat?event=post.clientViewed&referrerSource=full_rss&postId=a90b0b1b6f5f" width="1" height="1" alt="">]]></content:encoded>
        </item>
        <item>
            <title><![CDATA[Speech Analytics Call Center Solutions: Why Your Company Needs it — Lettria]]></title>
            <link>https://lettria.medium.com/speech-analytics-call-center-solutions-why-your-company-needs-it-lettria-6af8b1bd2eda?source=rss-4f5319276610------2</link>
            <guid isPermaLink="false">https://medium.com/p/6af8b1bd2eda</guid>
            <category><![CDATA[ai]]></category>
            <category><![CDATA[nlp]]></category>
            <category><![CDATA[speech-analytics]]></category>
            <dc:creator><![CDATA[LETTRIA]]></dc:creator>
            <pubDate>Mon, 23 Jan 2023 18:10:12 GMT</pubDate>
            <atom:updated>2023-01-23T18:17:20.592Z</atom:updated>
            <content:encoded><![CDATA[<h3>Speech Analytics Call Center Solutions: Why Your Company Needs it — Lettria</h3><figure><img alt="" src="https://cdn-images-1.medium.com/max/1024/0*VmTWakF5NEYR9dod.png" /></figure><p>When you’re setting up a call center phone system, ‘call recording’ and ‘live call listening’ are two important prerequisites. These features allow managers and trainers to listen to calls and voice and suggest improvements based on their analysis of the conversation. But, you can manually analyze only so many calls.</p><p>Manual analysis is slow and it may be biased. And, you get a partial view of what your audience wants. It is also difficult to perform when managers are overseeing remote teams.</p><p>That raises the question — why limit yourself to a small sample when you can get a full perspective with speech analytics software?</p><p>A <strong>speech analytics call center</strong> gives you near real-time insights on improving customer interactions across channels. It automates the process to make it faster, able to handle larger call volumes, and more reliable. Let’s dive deeper into speech analytics and how it benefits call centers.</p><h3>What is call center speech analytics?</h3><p>Speech analytics refers to the automated analysis of conversations between customer service representatives and customers using Machine Learning (ML) and Natural Language Processing (NLP). The results of this analysis can be used in many ways such as for quality control and training or to optimize operations.</p><p>It works by identifying words and patterns from recorded and transcribed conversations to analyze whether the customer sentiment expressed is positive, negative or neutral. Further, it assesses the keywords identified against the context of the conversation to evaluate call quality and monitor agent performance.</p><p>The automated process runs on every call made and received by your call center. It can also be used to draw similar inferences from communication via other channels such as text messaging, emails or interaction with chatbots. When you’re comparing speech analytics tools, this feature becomes important, especially in today’s age of omnichannel business communication.</p><h3>What Speech Analytics can do for your Call Center</h3><p>Empowered with speech analytics, call centers can make data-driven decisions to help stand out for their customer experience. Some of the key ways speech analytics can help your call center are:</p><h3>Identifying conversation themes</h3><p>Firstly, speech analytics helps understand what customers want by identifying themes that recur across conversations. Let’s take the example of a tea retailer. There are so many varieties, flavors, and packaging options to explore — how do you choose what to focus on?</p><p>Ideally, this must be based on what your customers want. Speech analytics helps identify the focal points that matter by highlighting common themes. If a majority of your customers talk about tea bags, this may be a vertical worth exploring.</p><p>Similarly, it helps you identify areas that need improvement. For example, if you notice a growing number of customers complaining about late deliveries, you may need a new logistics partner or find a better way to approximate delivery timelines when orders are being placed.</p><h3>Assessing how customers feel about your brand/product/service</h3><p>Speech analytics gives you a clear perspective on how customers feel about your brand and the products/services offered. By analyzing keywords against the context of the conversation, each conversation is scored from -1 to 1 in terms of the sentiment expressed.</p><p>Conversations, where a negative emotion is expressed, are scored from -1 to 0 while conversations with a positive emotion are scored from 0 to 1 with 0 being the score for a neutral emotion. Here are a few examples:</p><ul><li>Positive sentiment: I loved the rose-flavored tea</li><li>Neutral sentiment: The product matched the online description</li><li>Negative sentiment: I opened the package to find 2 cups broken</li></ul><p>Speech analytics tools like <a href="https://lettria.com/">Lettria</a> that use advanced AI can go beyond simple statements to extract sentiments from complex statements with slang or sarcastic notes. Let’s say a customer said, “It’s so good to speak to you, I’ve been on hold for only 20 minutes.” Though the word ‘good’ has a positive inference, the context it has been used is sarcastic and hence the statement will be graded as a negative sentiment.</p><p>Understanding your customer’s sentiments can enable your customer service agents to respond appropriately and take measures to improve the experience. Imagine the customer’s frustration in the above example if the agent responded by saying, “Nice to speak to you too”!</p><h3>Measure call center agent performance</h3><p>Just as speech analysis tools evaluate the choice of words being used by customers, they do the same to assess the sentiments being expressed by your call center agents. Do they sound happy, sad, angry or worried?</p><p><a href="https://lettria.com/">Lettria</a> does the monitoring and uses deep learning to categorize sentiments into 28 such emotional values. This data can prove critical for understanding how your customers are being treated and the impression being formed about your brand. You may have a great product but if your agents sound irritated on call, your customers aren’t likely to stay loyal to the brand.</p><h3>Mitigate fraud and compliance</h3><p>Using speech analytics also helps identify instances where standard operating procedures are bypassed leading to fraud and compliance risks. For example, having an agent ask the customer for an OTP could put your company at risk of financial fraud. The faster you can identify such rogue agents, the earlier you can address the issue.</p><p>Similarly, sending customers emails without their consent can put your brand at risk of having to pay fines and being blacklisted as spam. If a customer calls to unsubscribe from your mailing list, speech analytics can highlight the conversation so that appropriate action is taken to mitigate such risks.</p><h3>Benefits of Speech Analytics for Call Centers</h3><p>Call centers across industries can experience a range of benefits by using speech analytics. This includes:</p><h3>Understand customer needs better</h3><p>Rather than relying on a small sample of calls that can be manually analyzed, speech analytics assesses each and every call to give you a more accurate perspective on how your customers feel about the brand and what they want.</p><p>For example, your managers may have assessed 10 calls and found that all these conversations were centered around opening a new store in India. But, with automated speech analytics you may realize that the remaining 90 calls were from customers wanting a new store in Germany. Similarly, it helps assess what customers like and dislike the most about your service and increase sales.</p><h3>Improve conversion rates</h3><p>By helping you understand customer sentiments, speech analysis empowers agents to personalize interactions and build a better rapport with the customers. They can then leverage this rapport to cross-sell or upsell products and services.</p><p>It helps increase the chances of conversion as well as the overall customer experience. The software can also pinpoint keywords and responses that work well with specific customer demographics.</p><h3>Improve resolution rate</h3><p>If your agent misreads a situation and responds to a sarcastic statement with a joke rather than empathy, the customer is likely to get more irritated. Understanding the sentiment behind a customer’s call can put agents on the same page as them and helps avoid such instances.</p><p>The customer service agent is better equipped to respond to situations correctly and resolve their issue on the first call. Ultimately, a higher resolution rate reduces the overall call volume and gives your agents more time to spend on each call.</p><h3>Reduce customer churn</h3><p>There will always be customers who switch to competitors and unsubscribe from your mailing lists. While this is not completely avoidable, understanding the reasons behind this can help reduce customer churn. Speech analytics helps highlight keywords and themes that are causes of concern with respect to customer engagement and satisfaction. For example, if you see a pattern emerge where the words ‘cancel’ and ‘late delivery’ are used together, improving your supply chain may help keep customers from leaving and retain their loyalty.</p><h3>Spot industry trends</h3><p>Just as the products offered by your company need to keep up with market trends, so does your customer service team. Speech analytics helps identify trends in customer expectations to keep your call center in tune with the industry.</p><p>For instance, a speech analysis tool may highlight a growing number of conversations where customers of a flat-pack furniture brand ask for video tutorials of product assembly. Further study may identify an increase in the number of customers seeking self-service tools. Armed with this data you can work on developing such tools and keep your customers happy.</p><h3>Boost productivity</h3><p>We’ve already talked about how speech analysis helps improve call resolution and, in turn, makes call center agents more productive. In addition, it also boosts efficiency and productivity for managers.</p><p>Managers can now tell which agents require more hand-holding and which agents are doing well on their own without having to listen in to calls being made by everyone. This also helps avoid micromanagement and lets managers use their time more productively.</p><h3>Track agent performance without bias</h3><p>One of the biggest issues with manually assessing agent service is having a biased opinion. For example, a manager may assume that an agent is taking longer to resolve an issue simply because he is speaking to an elderly customer. However, in truth, it may be because of a language barrier.</p><p>Automated speech analytics bypasses such biases completely and helps track agent performance according to standardized KPIs. It lets managers know at a glance how well each agent is doing and helps spot deviations from call scripts that work as well as those that don’t. This insight can be used to alter scripts and make them more effective.</p><h3>Ease quality management</h3><p>Every call center is constantly working towards improving the level of customer service offered. This is one of the areas where speech analysis plays a pivotal role. It helps assess how agents are doing and what scripts can be followed as is and what scripts are being frequently changed.</p><p>Here’s a very simple example, let’s say the city changed the name of a street but the call center script still lists addresses with the old street name. With speech analysis, managers will be able to identify company agents having to make this deviation from the script. They can then update the script to maintain uniformity across all communication.</p><h3>Increase ROI</h3><p>Rather than take a hit-and-try approach towards introducing new customer service features or base decisions on a partial view of the customer’s needs, speech analytics gives you access to information that can help make wiser, more strategic decisions.</p><p>You are able to invest in developing services that are more likely to be profitable. You can also spot industry trends and shifts in consumer behavior in advance. This keeps you prepared and lowers the risk of losses.</p><h3>Speech Analytics Call Center Solutions</h3><p>Given the important contributions speech analysis can make to the everyday functioning and overall quality standards of a call center, it isn’t surprising to note that there are many tools available for speech analytics. They vary in terms of not only pricing but also the scope of analytics provided, processing speed, accuracy and ease of operations. If you’re looking for speech analytics tools for your call center, here are the top 3 best options you should consider.</p><h3>1. Lettria</h3><p><a href="http://lettria.com/">Lettria</a> is the leading speech analytics call center solution provider. It uses a platform-based approach with 15 pre-trained and customizable multilingual models. Lettria’s low-code customizable nature makes it capable of delivering advanced sentiment analysis with high levels of accuracy.</p><p>This is one of the reasons for its popularity. The second is the way Lettria analyzes speech. It goes beyond the basic positive, neutral and negative emotions to grade sentiments according to 28 emotional values. Lettria breaks the analysis into 2 steps.</p><p><strong>Step 1</strong>: It identifies the adjectives and adverbs within a statement to detect a base sentiment. It is then graded from -1 to 1 according to the emotional elements or sentence-level comparisons.</p><p><strong>Step 2</strong>: This brings us to step 2 where the AI processes the context of the statement. Thus, it not only evaluates the statement’s polarity but also determines the emotional value of the statement.</p><h3>2. Calabrio</h3><p>Calabrio is a workforce performance software that uses Artificial Intelligence to understand the needs of a customer and anticipate demands based on interactions with your contact center. It brings all contact center data to one interface so that everyone is connected. Further, it provides analytical data that aids in taking measures required to meet customer expectations and increase agent engagement.</p><h3>3. CallMiner</h3><p>The CallMiner platform for speech analytics helps capture, analyze and understand the emotions behind what the customer is saying. It uses AI-powered speech analytics technology to extract data from telephone conversations as well as emails, chats and social media interactions, classify it as positive or negative and deliver insightful reports.</p><h3>Boost Your Call Center Performance with Speech Analytics Today</h3><p>According to a <a href="https://www.mckinsey.com/capabilities/operations/our-insights/from-speech-to-insights-the-value-of-the-human-voice">McKinsey</a> paper, speech analytics can help call centers save up to 30% in terms of their costs and improve customer satisfaction scores by at least 10%. Speech analysis benefits apply to call centers across all industries and are especially effective for cloud contact centers.</p><p>With AI tools like Lettria, you can gain insights into customer needs that are unavailable elsewhere and identify opportunities to improve agent performance, operational efficiency and more. Explore the platform today and see how it can benefit your organization’s call center.</p><p>‍</p><p><em>Originally published at </em><a href="https://www.lettria.com/articles/speech-analytics-call-center-solutions-why-your-company-needs-it"><em>https://www.lettria.com</em></a><em>.</em></p><img src="https://medium.com/_/stat?event=post.clientViewed&referrerSource=full_rss&postId=6af8b1bd2eda" width="1" height="1" alt="">]]></content:encoded>
        </item>
        <item>
            <title><![CDATA[Lettria’s semantic categorization project as part of the development of a NLP solution — Lettria]]></title>
            <link>https://lettria.medium.com/lettrias-semantic-categorization-project-as-part-of-the-development-of-a-nlp-solution-lettria-81ffb1ae129?source=rss-4f5319276610------2</link>
            <guid isPermaLink="false">https://medium.com/p/81ffb1ae129</guid>
            <category><![CDATA[semantics]]></category>
            <category><![CDATA[text-processing]]></category>
            <category><![CDATA[ai]]></category>
            <category><![CDATA[nlp]]></category>
            <dc:creator><![CDATA[LETTRIA]]></dc:creator>
            <pubDate>Mon, 23 Jan 2023 18:02:22 GMT</pubDate>
            <atom:updated>2023-01-23T18:17:50.143Z</atom:updated>
            <content:encoded><![CDATA[<h3>Lettria’s semantic categorization project as part of the development of a NLP solution — Lettria</h3><p>Lettria’s textual data management platform is a natural language processing (NLP) tool based on several years of research and development involving algorithms, data science, software engineering and linguistics. Linguistics is at the heart of the work, as natural languages are the raw material for information. An essential step in the formalization of textual data, semantic categorization aims at identifying the type of meaning of each word in a given language production, the word being in theory the smallest autonomous signifying unit.</p><p>The creation of semantic categories thus participates in the structuring of data, but also helps to face a major problem in the processing of natural languages, that of ambiguity. This work is not without its obstacles. Phrasemes are one of them, as well as the syntactic and semantic particularities of verbs, especially pronominal verbs in French. All these pitfalls have led Lettria to make decisive choices, always with the objective of obtaining the most relevant results.</p><h3>Purpose</h3><h3>Information structuration</h3><p>Information structuration requires the identification of entities and events evoked in a text, as well as their properties and their relationships. To do this, Lettria has created several hundred semantic categories organized within graphs. The categories group words of the same part of speech with common semantic features, regardless of the language. Each of them has a definition and each word in a text must correspond to a single category. For example, the definition of the category <strong>aircraft</strong> is “air transport vehicle” and includes the English words <em>plane</em> and <em>helicopter</em> as well as the French words<em> avion</em> and <em>hélicoptère</em>. The definition of the category <strong>danger</strong> is “dangerous thing; that which threatens something or somebody” and includes the English words <em>danger, peril </em>and <em>risk </em>and the French words <em>danger, péril </em>and <em>risque.</em></p><p>In general, nominal categories have the same relationships to each other as nouns have to each other in terms of hypernymy and hyponymy. Howard Jackson defines hyponymy as follows:</p><blockquote><em>“Hyponymy refers to the semantic relation of inclusion: the meaning of a (more specific) lexeme is included in that of another (more general) lexeme. Thus </em>nova<em> is a hyponym of </em>star: star<em> is the superordinate term, and </em>nova<em> is the subordinate term in this semantic relation. Hyponymy is often important in displaying the semantic relations between items within a lexical field.” (Howard JACKSON, </em>Words and their Meaning, <em>p. 92, Routledge, 1988)</em></blockquote><p>Hyperonymy is therefore the inverse relationship of hyponymy. These relations can be found in the structure of our graphs: the category <strong>mammal</strong> has a more general meaning than the category <strong>carnivore</strong>, whose meaning is itself more general than the category <strong>dog</strong>.</p><figure><img alt="" src="https://cdn-images-1.medium.com/max/684/0*wPvdvuWyj1gkJ9iG.png" /></figure><p>The category <strong>skin_illness</strong> has a more precise meaning than the category <strong>illness</strong>.</p><figure><img alt="" src="https://cdn-images-1.medium.com/max/1002/0*SXRfewdhiMicphCl.png" /></figure><p>It is therefore logical that the nominal categories designating concrete entities are hierarchically arranged in a taxonomic type of graphical representation going from the most general to the most particular.</p><p>This organization facilitates the creation and the use of categories. The concept of taxonomy, typically associated with attempts to describe the living world, is explained by Jacques Ruffié:</p><blockquote><em>“Classification, or taxonomy, is therefore the process of gathering, through successive levels, living organisms that have more and more common characteristics. In doing so, we go from higher classification levels (phyla, classes) to lower classification levels (orders, families, species).” (Jacques RUFFIÉ, </em>De la biologie à la culture, <em>p. 18, Flammarion, 1976)</em></blockquote><p>Thus, the category <strong>plant</strong>, which contains the words <em>plant </em>and<em> vegetal,</em> has several subcategories including <strong>tree</strong> which contains the words <em>tree</em> and <em>shrub</em>. The <strong>tree</strong> category itself has several subcategories including <strong>conifer</strong>, which contains words like <em>cedar, larch</em> and <em>fir</em>.</p><figure><img alt="" src="https://cdn-images-1.medium.com/max/904/0*pQDRprBe2pcccpfK.png" /></figure><p>This type of organization is valid for other realities than living beings, as the linguist and lexicographer Alain Rey has analyzed:</p><blockquote><em>“These logical classes, whose constitution is studied by taxonomy, manifest themselves well in the field of hierarchical and systematic classifications of the natural sciences (nomenclatures), but also in the field of hierarchies of objects spontaneously observed and named within a culture (folk taxonomies).” (Alain REY, </em>La Terminologie, <em>p. 35, P. U. F., coll. “Que sais-je”, 1979)</em></blockquote><p>For example, in our graph, the category <strong>drink</strong> is superior to the category <strong>alcoholic_drink</strong>, which is superior to several categories including <strong>wine</strong> and <strong>beer</strong>. Such an organization facilitates both the creation of graphs and their consultation. As we shall see, for certain nouns designating abstract entities and for verbal categories, the hierarchy is based on another type of organization.</p><figure><img alt="" src="https://cdn-images-1.medium.com/max/920/0*SCU4azusdiwLedBI.png" /></figure><p>This raises the question of the granularity of the categories, i.e. their level of precision. Lettria has chosen to propose a general representation intended to cover all knowledge domains without privileging any of them, while keeping the possibility of proposing complementary graphs that are more detailed in specialized domains such as medicine or agronomy. This is why some categories group the hyponyms of a word with this word itself. For example, the category <strong>dog</strong> is a final category — it has no subcategories — which contains the word <em>dog</em> and its synonyms <em>mutt</em> and <em>bow-wow,</em> but also <em>poodle, greyhound </em>and <em>corgi,</em> although these are hyponyms of <em>dog.</em></p><p>Moreover, such an organization is necessarily subjective, as a word can have several hypernyms. The classification of living beings itself does not have a consensus and is constantly evolving at the rhythm of scientific discoveries. It is therefore the most salient semantic features and the most commonly accepted representations by non-specialist speakers that have been retained. <em>Mother</em> is generally considered as a hyponym of <em>woman,</em> but we preferred to include it in the category <strong>family_member</strong> (“person designated by his family or marital relationship”) with<em> uncle</em> and <em>cousin.</em> <em>Bell pepper</em> is included in the category <strong>vegetable</strong> although from the scientific point of view it is a fruit. We are in a so-called <em>naive </em>or<em> popular</em> and non-scientific representation of the world, and this is what we are looking for by formalizing the information conveyed by everyday language, outside of any scientific field.</p><p>Lettria’s database, including thousands of words associated to their possible categories, is therefore both a linguistic database and an encyclopedic database that offers a structured representation of the realities and concepts designated by the words. Eventually, the categories will be enriched with ontological properties that will allow, among other things, to assign them a certain number of attributes (for example, a person has a date of birth, a city has geographical coordinates) but also to link them by relations other than the taxonomic hierarchy (a person having a city of birth, the corresponding categories will be linked by a given relation). This enrichment will not only allow us to refine our structuration work, but also to offer Lettria’s clients a firm base for the creation of their own ontologies.</p><h3>Disambiguation</h3><p>The creation of categories is also of great use in Lettria’s disambiguation work. <em>Ambiguity </em>receives the following definition:</p><blockquote><em>“Ambiguity is a property of natural languages due to the absence of a one-to-one correspondence between form and meaning. It can be defined semantically, close to immediate perception: any statement that can be interpreted in several ways is ambiguous.” (Michel ARRIVÉ, Françoise GADET &amp; Michel GALMICHE, </em>La Grammaire d’aujourd’hui,<em> p. 60, Flammarion, 1986)</em></blockquote><p>Building its NLP model, Lettria chose to tackle this problem with machine learning, a field of artificial intelligence that aims to create models that improve their performance according to the data they process. Thus, annotation campaigns were set up. Lettria’s teams annotated textual data of various origins by assigning each word the appropriate category. After several months of training, the correct category is most of the time automatically recognized by the machine thanks to context analysis.</p><p>Note that the semantic ambiguity also concerns the name of the categories. A category is often named after one of the most representative words of the category in question. Since the aim is to create a univocal model, it is necessary that a name be used for a single category, but also that each name be as unambiguous as possible, because even if each category is accompanied by a definition, an ambiguous name can be a source of error during annotation. In this case, a monosemic word or a word with a clearly delimited meaning is preferred. In some cases, a paraphrase is also used. The category with the definition “character of a person who lacks moral values” is thus named <strong>moral_lowness</strong>.</p><h3>The case of phrases</h3><h3>Lexical fixation</h3><p>A major obstacle in semantic categorization is the impossibility of assigning meaning to words that appear in specific sequences : the meaning of a sequence of words is not always the combination of the meanings of the units that compose them. When it’s raining cats and dogs, you don’t see animals falling. Pulling someone’s leg doesn’t require touching them. And when you say “break a leg” to someone, you are not enjoining them to hurt anyone. To this is added the existence of fossil words, which can never or no longer be used freely. <em>Fro,</em> in <em>to and fro, nother, </em>in <em>a whole nother,</em> or the French<em> fur </em>in au <em>fur et à mesure </em>(“as one goes along”) are typical examples.</p><p>This is the lexical fixation phenomenon: the elements of discourse lose their semantic autonomy within certain combinations. These combinations are grouped under different names, among which we find <em>phrases</em> ( <em>unités phraséologiques</em> in French), and their definition varies according to the authors, just like their delimitation:</p><blockquote><em>“The set of fixed expressions is heterogeneous, and several distinct types may be observed.” (Rosamund MOON, </em>‘There is reason in the roasting of eggs’: a consideration of fixed expressions in native-speaker dictionaries,<em> in </em>EURALEX ’92: Proceedings I-II : papers submitted to the 5th EURALEX International Congress on lexicography in Tampere, Finland (Studia translatologica),<em> p. 494, Oxford University Press, 1992)</em></blockquote><h3>Tokenization and phraseological categorization</h3><p>Lettria attempts to solve this problem in two stages of data processing. The first is tokenization, which consists of segmenting each sentence into tokens, i.e. manipulatable linguistic units such as words and punctuation elements, but also entities such as proper nouns<em> (Statue of Liberty)</em> or dates. At this stage, a certain number of polylexical sequences are identified as minimal linguistic units semantically indecomposable. <em>Bell pepper </em>and <em>de facto, </em>although composed of several sequences of signs separated by spaces, will be lemmatized and treated as minimal elements and categorized as such. But these are only short monosemic sequences, identified as words in the commonly accepted sense of the term.</p><p>A later stage allows the recognition of phrases at a higher level in the discourse and previously listed by Lettria as phraseological sequences that can be decomposed but whose combination has a non-predictable meaning. It is at this stage that locutions, expressions and phrasal verbs are identified, to which one or more possible categories are associated, as these sequences may be polysemous. During disambiguation, the units listed are identified and categorized by the machine in the same way as monolexical units. The sequences <em>lose life </em>and<em> cry up</em> (and all their conjugated forms) will therefore be respectively assigned the categories <strong>die</strong> and <strong>praise</strong>. When several categories are possible, disambiguation works in the same way as for monolexical units, thanks to contextual analysis. For example, the verb <em>deliver</em> will receive the category <strong>give_(sthg)</strong> or <strong>express_(sthg)</strong> depending on whether the object designates something material or an information.</p><h3>Literal meaning vs figurative meaning</h3><p>But if some phrases can receive several categories, there are also sequences which can be phrases or not: like monolexical units, polylexical units can have a literal and a figurative meaning. <em>Piece of cake</em> can thus be a phrase with a figurative meaning, “something easily done” (Merriam-Webster.com, 2022), and thus correspond to the <strong>simple_thing</strong> category: <em>Operating this device is a piece of cake. </em>But, of course, it can also be used literally:<em> I’ll have a piece of cake for dessert.</em> It is therefore not possible here to assign a category to the whole sequence insofar as the semantic unity is at the level of each word.</p><p>The preferred solution is based on the frequency. When the literal meaning is possible but a priori much less frequent, the figurative meaning is chosen. This is why <em>to give the cold shoulder</em> will be considered as a phrase receiving the category <strong>neglect_(sthg)</strong>. When the figurative meaning is less frequent or in the case of a more or less equal frequency, it is the literal meaning that is privileged, as for <em>piece of cake</em>. The sequence is then considered as a free combination of autonomous elements which will each receive their own category.</p><h3>Categorizing verbs</h3><h3>A non-taxonomic organization</h3><p>The semantic categorization of verbs raises several problems, starting with their organization within a graph. Some verb meanings can be more precise than others: <strong>get_old</strong> could be a subcategory of <strong>get_changed</strong> for example. But, in the vast majority of cases, verbs do not have hypernymy and hyponymy relationships between them like those of nouns. Verbs with a meaning close to that of <em>reflect,</em> such as <em>think</em> or <em>brainstorm,</em> have nuances that do not require a different level of precision. Therefore, in the graph devoted to verbal categories, only the final categories, those without subcategories, can receive verbal meanings.</p><p>In this graph, the intermediate categories between the initial node and the different final categories have the function of ensuring the overall coherence and of facilitating the location of the right category while annotating, which would be less possible with a simple list. The main branches correspond to the different types of actions or states expressed by the verbs: movement, possession, change, communication, etc. For example, the POSSESSING branch contains the HAVE and HAVE_NOT sub-branches. HAVE contains the categories <strong>have_(sthg)</strong> and<strong> obtain_(sthg)</strong>, while HAVE_NOT contains <strong>have_not_(sthg)</strong> and<strong> lose_(sthg)</strong>.</p><figure><img alt="" src="https://cdn-images-1.medium.com/max/922/0*DyPH-zN2_KuKiCmg.png" /></figure><h3>The argument structure</h3><p>In addition to the thematic organization, the verbal categories graph and its categories themselves are created according to the argument structure of the predicates. In order for the discourse to be intelligible, words impose certain syntactic and/or semantic constraints. Verbs are the most typical representatives of this. A conjugated verb must be accompanied by other linguistic units (usually noun phrases) that indicate which entities are participating in the process. These units are called <em>arguments </em>( <em>actants</em> in French).</p><p>Verbs are therefore at the heart of textual data structuration: they provide information not only about the events and states they designate, but also about the entities designated by other words in the sentence. For example, in <em>Paul sold his car to Sarah,</em> we know from the verb <em>sell</em> that Sarah owns the car and that Paul no longer owns it. In order for the model to report this information, it is necessary to transmit to the machine through categories taking into account two properties related to the argument structure of verbs: verbal valency and semantic roles.</p><h4>Verbal valency</h4><p>However, many verbs alternate between a one-argument valency and a two-argument valency. <em>To move,</em> for example, can be used as an intransitive verb and as a transitive verb:<em> I move slowly / I move the chair. </em>In the first case, it is the subject that undergoes the action while in the second it is the object that is moved. Although it is the same verb, the information concerning the arguments is totally different according to the valency and this must be taken into account in the representation of the information.</p><p>This is why this graph has a first division between an ACT branch and an INTERACT branch. Generally speaking, we find on the ACT side verbs with one argument and on the INTERACT side verbs with two arguments. This is the case for verbs of movement and verbs of change <em>(The pages have turned yellow / The sun has turned the pages yellow), </em>which creates parallel categories in both parts of the graph. For some types of verbs, the difference is between two and three arguments, respectively on the ACT side and on the INTERACT side. This is the case for verbs of possession:<em> own (Paul owns a car)</em> is on the ACT side while<em> give</em> <em>(Paul gives his car to Sarah)</em> is on the INTERACT side.</p><h4>Semantic roles</h4><p>In addition to verbal valency, there is the question of the semantic roles assigned to the arguments. The semantic role is the role played by each argument in the process expressed by the verb. A noun with a certain grammatical function (subject or object) can have different semantic roles depending on the verb that governs it: agent, patient, beneficiary, goal, place, etc. The list is not exhaustive and varies according to the approach.</p><p>However, in the same semantic field, and therefore in the same branch of the graph, two verbs of identical valency may not assign the same roles to the arguments. Among the verbs of possession, let’s take the case of the verbs <em>offer</em> in <em>They offered a protective helmet to the team</em> and <em>equip</em> in <em>They equipped the team with a protective helmet.</em> In the first sentence, the beneficiary is the indirect object, whereas in the second sentence the beneficiary is the direct object. It is therefore necessary to assign different categories to these verbs which generate different semantic structures: it has consequences on the structuring of the information. Thus, <em>offer</em> is found in the category <strong>give_(sthg)</strong> while <em>equip</em> is found in <strong>provide_(sbody)</strong>. We note that the names of these categories emphasize this difference, the direct object being more often a thing in the first case and a human being in the second one.</p><p>This phenomenon can also exist for the same verb. This is the case of <em>supply: This producer supplies local restaurants with wine / This producer supplies wine to local restaurants.</em> In the first example the beneficiary is the direct object and in the second case it is the indirect one. The verb<em> supply</em> is therefore found in both categories, <strong>provide_(sbody)</strong> and <strong>give_(sthg)</strong>.</p><h3>French pronominals</h3><h4>Transitive form vs pronominal form</h4><p>French presents a particular difficulty in the categorization of pronominal verbs, which are traditionally divided into three groups whose properties will lead to differences in annotation. Essentially pronominal uses (sometimes called in French<em> pronominaux lexicalisés,</em> “lexicalized pronominals”), that accept only a pronominal object in the same person as the subject, are generally not problematic. This is the case for verbs that exist only in the pronominal form, such as <em>s’abstenir</em> (“to abstain”),<em> s’évanouir</em> (“to faint”) or <em>se repentir </em>(“to repent”), or for certain verb meanings that can also be used in a non-pronominal way. Thus, in the sentence <em>Je m’attends à ce qu’il pleuve demain</em> (“I expect it to rain tomorrow”), <em>s’attendre</em> does not correspond to any transitive use of <em>attendre.</em> The pronoun<em> se</em> is considered semantically empty and has no semantic role. This last use therefore belongs in the expect category with non-pronominal verbs such as <em>pressentir </em>and <em>pronostiquer.</em></p><p>When a transitive verb is used in the pronominal form without change of meaning, it belongs to the same category. In the examples <em>Elle maquille sa sœur</em> and <em>Elle se maquille, </em>the patient is always the object: <em>sa sœur</em> in the first sentence and the pronoun<em> se</em> in the second one. But some cases are more problematic. In the sentence <em>Je me déplace lentement,</em> should we consider that<em> déplacer </em>belongs to the category <strong>move_in_space</strong> (“to go to another place”) with intransitive verbs such as <em>marcher </em>(“to walk”) or should we integrate it to the category <strong>move_(sthg)_in_space</strong> like the transitive use of <em>move</em> <em>(I move the chair slowly)</em>? The second option has been chosen, considering that the subject <em>(Je) </em>acts on itself <em>(me)</em> as in the previous example.</p><p>Two criteria are therefore essential to judge that a pronominal form is similar to a transitive form. First, we must make sure that the verb exists with the same meaning in the transitive form. This was not the case with <em>s’attendre: </em>one cannot say <em>*J’attends mon frère à ce qu’il pleuve demain.</em> On the other hand, it is the case of <em>déplacer</em> in our last example:<em> Je déplace le fauteuil lentement.</em> It is necessary that the subject is the agent of the action. This is the case in <em>Je me déplace lentement,</em> but it is not the case in a sentence like <em>La pièce s’est réchauffée rapidement</em> (“The room warmed up quickly”). <em>Se réchauffer </em>will therefore find its place in the <strong>get_hot</strong> category and not in <strong>heat_(sthg)</strong>. We note that the usual lexicographical analysis corroborates our choice, as <em>se réchauffer</em> has its own definition in dictionaries: “Devenir plus chaud. <em>La température se réchauffe.</em>” (“To become warmer. <em>The temperature warms up.</em>”, <em>Le Petit Robert de la langue française,</em> édition numérique, Éditions Le Robert, 2022).</p><h4>Accidental passive pronominals</h4><p>This last case, sometimes considered as a passive pronominal, should be distinguished from what we will call accidental passive pronominals. In the sentences <em>Les robes vertes se vendent beaucoup cette année</em> (“Green dresses are selling a lot this year”) and <em>Cette machine se pilote facilement </em>(“This machine is easily controlled”),the verbs do not meet the criteria necessary for their inclusion in the same category as their transitive form. However, they are semantically different from the previous example. In<em> La pièce s’est réchauffée rapidement,</em> it is difficult to modify the utterance to give<em> la pièce</em> the object function without slightly modifying the meaning of the sentence: <em>On a réchauffé la pièce rapidement</em> implies an agent performing a voluntary action. This was not the case with the pronominal form. However, it does work with the other two examples. <em>On vend beaucoup de robes vertes cette année </em>and <em>On pilote facilement cette machine</em> do not change the meaning of the original sentences which already implied the existence of an agent acting voluntarily: people or organizations that sell green dresses sell a lot of them this year and anyone who drives this machine drives it easily.</p><p>This is similar to the passive voice with omission of the agent: <em>Les robes vertes sont beaucoup vendues cette année ; Cette machine est facilement pilotée.</em> Such a transformation is again not possible with the first example:<em> La pièce a été rapidement réchauffée </em>implies an agent that the initial sentence does not imply. Moreover, in the case of accidental passive pronominals, one can easily express the same idea with an impersonal turn of phrase: <em>Il se vend beaucoup de robes vertes cette année; Il est facile de piloter cette machine.</em> Again, the same attempt with the first example necessarily implies an agent and changes the meaning of the sentence:<em> Il a été rapide de chauffer la pièce.</em></p><p>In the sentences with <em>vendre </em>and<em> piloter</em>, we will therefore classify the verbs in the same categories as their transitive uses, contrary to the example given with <em>réchauffer.</em> The attempts at transformations shown above are all criteria for correctly categorizing a pronominal verb with passive value. And the result corresponds to our structuration needs: in the case of the accidental passive pronominals, we can formalize the action of one entity on another, while in the other case there is only one entity involved and it undergoes a change of state.</p><p>‍</p><p>The problems peculiar to the creation of semantic categories and, more generally, by the structuring of textual data, therefore require in-depth knowledge and analysis of natural languages and the specific characteristics of each one. This is a necessity that Lettria takes fully into account in the building of its models, with a constant concern for efficiency and with a single goal in mind: to offer its clients the most successful solution possible in order to extract from their data the relevant information for the smooth operation and development of their business.</p><p>‍</p><p><em>Originally published at </em><a href="https://www.lettria.com/articles/lettrias-semantic-categorization-project-as-part-of-the-development-of-a-nlp-solution"><em>https://www.lettria.com</em></a><em>.</em></p><img src="https://medium.com/_/stat?event=post.clientViewed&referrerSource=full_rss&postId=81ffb1ae129" width="1" height="1" alt="">]]></content:encoded>
        </item>
        <item>
            <title><![CDATA[Enriching your CRM with the unstructured data from call transcripts and voice-to-text — Lettria]]></title>
            <link>https://lettria.medium.com/enriching-your-crm-with-the-unstructured-data-from-call-transcripts-and-voice-to-text-lettria-3d89cd6a3e8e?source=rss-4f5319276610------2</link>
            <guid isPermaLink="false">https://medium.com/p/3d89cd6a3e8e</guid>
            <category><![CDATA[sales]]></category>
            <category><![CDATA[crm]]></category>
            <category><![CDATA[ai]]></category>
            <category><![CDATA[nlp]]></category>
            <dc:creator><![CDATA[LETTRIA]]></dc:creator>
            <pubDate>Tue, 08 Nov 2022 15:51:54 GMT</pubDate>
            <atom:updated>2022-11-08T15:51:54.659Z</atom:updated>
            <content:encoded><![CDATA[<h3>Enriching your CRM with the unstructured data from call transcripts and voice-to-text — Lettria</h3><figure><img alt="" src="https://cdn-images-1.medium.com/max/1024/0*TXu68ZsUXamdPle_.png" /></figure><p>We all want more accurate and more complete data. But sometimes it is difficult to turn unstructured text into processed, analyzed, and structured data. This is a particular problem for companies that handle a large number of customer-facing phone calls from either their sales or customer support teams.</p><p>How do you turn all of that information into actionable insights? If that sounds like a familiar problem then you’re in the right place to find the answer. Lettria’s natural language processor (NLP) allows you to import transcripts of your phone calls, other recorded spoken interactions, and voice-to-text directly into the platform and then automatically enrich your Customer Relationship Management (CRM) software with invaluable information.</p><p>It is the most efficient way to identify key trends and important pieces of information from thousands (if not hundreds of thousands) of hours of calls. We know that you don’t have the time and resources necessary to manually review each one of these calls or other voice recordings and analyze them for important details, so we’ve made the process as easy as possible.</p><p>You get to automate every step. From transcription right up through to when the data is imported into your CRM (or other software). This helps our users to improve their understanding of their customers, get a better sense of their overall business performance, and use the information to improve their sales and customer support processes.</p><p>Your phone calls with customers and leads are one of the most important aspects of your business. It is often the best, and occasionally the only, way to communicate directly with a customer or lead. So how can an organization turn those one-to-one interactions into large data sets where they can identify trends, search for key information, and monitor business performance? The answer is through natural language processing.</p><p>Interested in finding out more? Over the course of this article we will explain to you how natural language processing can help you to identify key business insights from your call transcripts and other voice-to-text use cases, allow you to enrich your CRM with key data from that unstructured data, and help you to get your NLP project started.</p><p>But let’s start with the basics. Just what is natural language processing?</p><h3>A (quick) beginner’s guide to natural language processing</h3><p>The first example of natural language processing was demonstrated in 1954 when IBM’s 701 mainframe was used to translate Russian sentences into English. In the nearly 70 years that have followed, natural language processing has become one of the fastest developing and most advanced areas of artificial intelligence.</p><p>Natural language processing allows computers to understand the human language. We typically refer to the language that computers use as machine code or machine language and it would appear as total nonsense to anyone unfamiliar with software engineering. You can think of natural language processing as the bridge between this machine code and human languages. The meaning of the language is extracted by breaking text into words, establishing a contextual relationship between different words, and structuring that information into data.</p><p>Its applications can be seen in a wide variety of different areas and most people will come into contact with some form of natural language processing on a virtually daily basis. Whether that is still in the form of a language translator as IBM first demonstrated, or through chatbots or voice assistants, NLPs form the foundations of all interactive systems.</p><p>Natural language processing has advanced since those early days at IBM and is now capable of analyzing huge sets of language-based data in a way that humans could not manage themselves. It’s totally consistent. Unbiased. Unrelenting.</p><p>So how exactly does natural language processing even work? If you’re a data scientist or developer this will all come across as pretty basic to you and you can skip ahead to the next section, but if you’re new to the world of NLPs then it is certainly worth understanding a little bit about the technology itself.</p><p>Natural language processing includes several different techniques for interpreting human language. These techniques include everything from a statistical approach to machine algorithmic methods, but the end goal is always the same.</p><p>By running a series of examples through an Artificial Intelligence (AI) model and gradually teaching it the relationships between different words and phrases it is eventually able to break down unstructured text into smaller elements and identify those associations and understand their meaning.</p><p>As NLPs have advanced, deep learning techniques mean that computers can now understand the full complexity of the human language — including processing grammatical errors, misspellings, abbreviated words, or incorrect punctuation. This is tremendously important for most modern NLP applications as it turns complicated unstructured text into a numeric structure that can be used by downstream applications.</p><p>It is this very process that allows your email provider to identify spam or your home assistant to play the song that you’re desperately yelling at it for. Who would want to live in a world where your smart speaker plays Ryan Adams when you’re really looking to relive your glory days through the nostalgic appreciation of Canadian icon Bryan Adams. I shudder at the thought.</p><p>These NLP fundamentals then help more advanced AI models and platforms to perform more complicated tasks, including:</p><ul><li>Content categorization: this includes search, indexing, and duplication detection</li><li>Topic discovery and modelling: Identifying the meaning and themes within a text</li><li>Corpus analysis: Understanding document structure and preparing it for further models</li><li>Contextual extraction: Automatically pulling structured information from text-based sources</li><li>Machine translation: translating text or speech from one language to another</li><li>Document summarisation: Generating synopses of larger text</li><li>Speech-to-text and text-to-speech: We probably don’t need to explain that one</li><li>Sentiment analysis: Identifying the mood or emotion expressed within a text</li></ul><p>And look, we know what you’re thinking. Online translations aren’t that accurate, spam filters miss a lot of bad emails, chatbots can be pretty infuriating to talk to, and your voice assistant often fails to understand you. But that isn’t a problem with the technology, it’s just a case of how advanced is the specialization and customization of the NLP that you’re interacting with.</p><p>That’s why choosing the right NLP and training your AI model for your specific requirements is essential if you’re going to get the results that you need. The functions and examples that we’ve just provided are simply very common use cases and some of the times when you’re likely to encounter a real-world interaction with natural language processing.</p><p>Each of those more specific functions allow NLPs to carry out the tasks that have made them an integral feature of everyday life. There’s a good chance that your organization is already using natural language processing in some way, shape, or form, but one new way in which companies are benefitting from the technology is by using it to enrich their CRMs with information from call transcripts and voice-to-text.</p><h3>‍Using an NLP to enrich your CRM with data from call transcripts</h3><p>That quick guide to natural language processing mentioned many of the applications for the technology, but also revealed how useful it could be for anyone looking to extract data from their call transcripts.</p><p>The issue is, not all off-the-shelf NLPs specialize in textual data processing and building your own requires a great deal of resources — in terms of skills, time, and money. On top of that, automating the workflow can be a challenge and manually inputting the information at any one of the steps in the process is a major hindrance for the project itself.</p><p>And we know what you might be thinking, you already use software to automatically transcribe your calls and these are monitored and reviewed by managers. On top of that, your sales team might be pretty good at logging most of the information from their calls in your CRM.</p><p>But, we hate to break it to you, but information is definitely going missing from those calls. Not through incompetence or mismanagement, but simply because it’s impossible to manually record all of that information — especially when we are talking about customer support or care interactions.</p><p>Roughly 80% of the average CRM’s database is unstructured and a large part of this will be comprised of information from call transcripts. That means that you’re making a lot of sales or customer success decisions based on only 20% of the information that is available to you. And that means that you’re definitely missing insights or trends that could have a huge impact on performance.</p><p>But the impact of enriching your CRM with the value from within that unstructured data goes far beyond your ability to make smarter business decisions. In a world where we are constantly looking to automate processes, it also improves your ability to accurately program workflows.</p><p>Say you want to automatically send an email to every customer that has expressed a negative experience or opinion in the past month. If you rely on manual call logging then you will miss the vast majority of customers that would fall into this category.</p><p>But with the right NLP, you will be able to automate this process by creating a tag for a negative experience, have customers automatically assigned to this tag via the natural language processing results, and include each of these profiles included in your next email campaign.</p><p>Let’s even say you want to ensure that customers expressing an interest in a certain product or with a specific type of complaint automatically receive a callback. That information will be highlighted, included in your CRM, and your teams will be alerted to the fact that a call needs to be made.</p><p>Enriching your CRM with unstructured data will improve your decision making process, improve your processes, and increase efficiency. It allows your sales team and customer success teams to focus on the person on the other end of the phone and not on notes that they might need to take. It ensures that every call receives exactly the same standard of care, analysis, and attention.</p><h3>The Lettria solution</h3><p>That’s why Lettria is uniquely positioned to address this particular use case. Not only does the platform offer the benefits of an off-the-shelf solution combined with the accuracies and strengths of a fully customized NLP, but we’ve also developed specific integrations and functionalities that make enriching your CRM with data from your call transcripts and other recordings…well…sort of easy.</p><p>We know. Easy is never a term that you want to hear applied to fairly complex technology or software and it might even be a red flag for some of our readers. But trust us. We’ve really worked hard on this and tried to make this process as simple as possible. We’ve built a platform that allows you to turn a foundation into a customized NLP and found every possible way to save you time without reducing your capabilities.</p><p>Don’t believe us just yet? We understand, but keep reading and you’ll find out just how Lettria solves this particular problem and some examples from some use cases that we’ve worked on with clients.</p><p>So, let’s get to the bit about how Lettria handles the problem of enriching your CRM with your unstructured data.</p><p>By now you should at least have some understanding of how natural language processing works. What you might not yet be aware of are the two major challenges that make enriching a CRM with unstructured data difficult.</p><p>The first is creating a specialized NLP that is capable of processing unstructured textual data to your requirements and using that to populate fields within your CRM. The second is having an NLP that automates the two other key steps within this workflow: those being automatically processing the call transcripts and then transmitting that data to your CRM once the processing has been completed.</p><p>Lettria handles both of these challenges by making the process as simple as possible. For the first issue, our platform approach means that you get to start with a sort of NLP foundation that you can fine-tune for your specifications.</p><p>As we’ve already discussed with any technical software decision, you always have to ask yourself whether it is more effective to build or buy what you need. The major advantage associated with building your own NLP solution would be the fact that it would be specialized for your industry, the terms used by your leads and employees, and any specifications that might be unique to your company, project, or sector. The issue is that most companies don’t have the resources necessary to successfully build their own NLP.</p><p>Those lack of resources automatically forces most companies to look at off-the-shelf solutions. These require fewer in-house specialists and can be much quicker to get up and running, but that sometimes means that you sacrifice some level of specialization.</p><p>Lettria’s platform approach means that you get the best of both worlds. With 15 pre-trained multilingual models to start from, you get access to an AI model that you only need to focus on fine tuning in order to get the best results. This makes starting a project with Lettria 4 times faster than building one your own, but much more accurate than other off-the-shelf solutions.</p><p>The Lettria platform also allows you to connect to third-party software via an API and solve any integration problems. This means that you can automate the importation of call transcripts into your NLP platform and automatically enrich your CRM with the data that it processes.</p><p>In short. You get a specialized NLP platform that requires less time to set up and less time to manage on a daily basis. It’s a win all round.</p><p>But the advantages don’t stop there. This might be sounding a bit like an infomercial, but this is the section where I show you that bonus feature that you might not yet have realized that you needed. What is it you ask? Collaboration.</p><p>Most NLP projects are managed exclusively by data scientists and developers and for some projects that might be OK, but it isn’t the case for a lot of them. In fact, 85% of NLP projects fail because of the challenge of moving from development into production.</p><p>The reality is that, whilst your data scientists and developers might be more interested and comfortable with training an AI model and improving accuracy, your sales or customer success teams will be the ones more interested in the actual results.</p><p>That is clearly the case when you look at a problem like getting more value out of the unstructured data from within your CRM. Sure, your data scientists will fall in love with how accurate the Lettria platform is, but it’s your sales and customer support teams that will benefit from the trends that are identified, the processes that are automated, and the time that is saved.</p><p>Where other NLPs fail to actively involve non-technical profiles in the project, the Lettria platform encourages participation and makes it simple enough for anyone to use. The no-code approach means that you don’t need to have advanced technical skills in order to use the platform and your non-technical teams will be able to be involved in the project from the very start.</p><p>This has a number of benefits:</p><p>Collaboration really is the final step in ensuring that you can enrich your CRM using the unstructured data from your transcripts. If you’re looking at NLP solutions that don’t encourage and allow non-technical profiles to take an active role in a project then you are significantly reducing your chances of success.</p><p>At Lettria, we’ve not tried to develop an NLP for every use case, but we have tried to develop the best NLP platform for the use cases that we want to address. And we think that we’re pretty good when it comes to handling unstructured textual data and enriching your CRM with the information contained within your call transcripts.</p><p>But don’t just take our word for it. Here are a couple of real-world examples from clients that have used Lettria for some very different projects. What do they have in common? The huge challenge of taking large sets of unstructured textual data and extracting key insights from it that can improve processes, increase resource efficiency, and allow them to manage by exception.</p><p>Now, before we get there, we’ll address the elephant in the room in the fact that both use cases are from French organizations. Lettria is a French company and, whilst we take pride in our advanced multilingual models and our international footprint, some of our most advanced projects are from companies that started using us early on.</p><p>Let’s take a closer look.</p><h3>Solving the problem of unprocessed data in CRMs</h3><p>Created in 1991, La Poste started as the French national postal service but is now part of Groupe La Poste, which has expanded to included a bank and insurance company (La Banque Postale), a logistics service provider (Geopost), and a mobile network operator (La Poste Mobile).</p><p>With annual revenues of over €34 billion, the group now employs around 250,000 people across its various businesses. La Poste has more than 17,000 branches and post offices in France and relies on delivering a consistent level of customer service and experience across its business.</p><p>The nature of their business means that they communicate with customers and leads via phone calls on a daily basis. Their call centers and sales teams form the backbone of their business. Like many other organizations, they’ve implemented policies and introduced software that allow them to ensure consistency across those calls, but extracting key data from the calls themselves was a problem that they were struggling to solve.</p><p>La Poste’s problem is in no way unique. Any company handling thousands of calls will know how big of a challenge it is to extract key insights from each exchange and incorporate them into their CRM, knowledge base, and business strategy. Only 3% of their calls were being processed and reported to their CRM, so there was little doubt that valuable data was being missed.</p><p>When they started to look for a way to solve this problem, they not only found that natural language processing was the only viable option, but actually the perfect solution. After benchmarking various solutions and consulting with Illuin Technology, La Poste decided that they needed the customization that might be typically associated with an open-source solution combined with the advantages offered by a platform approach.</p><p>The Lettria platform was exactly what they were looking for and they started their NLP project in September of 2021. Whilst most natural language processing projects take a lot of time to get off the ground, Lettria’s platform helps users to reduce their time-to-market by 75% and La Poste’s proof of concept was released after just 4 months.</p><p>Training your own AI model can be incredibly time consuming and resource heavy, but Lettria has focused on optimizing this process by allowing you to customize the pre-trained models and this meant that the La Poste project only required an average of 2 hours of workshops per week during the development phase and under 100 hours of work to develop a fully customized AI model.</p><p>To make things even simpler for La Poste, Lettria developed connectors for both their speech-to-text provider (Allo-Media) and their CRM (Microsoft Dynamics) to simplify integration. This allows Lettria to automatically extract data once it has been processed by Allo-Media and push structured data directly to Microsoft Dynamics.</p><p>Although many NLP projects are handled exclusively by data scientists and development teams, the Lettria platform encourages collaboration and the inclusion of business profiles that helps users to get the most out of their projects.</p><p>In the case of La Poste, this meant that project managers, call center supervisors, and salespeople were all included in the process to draw on their knowledge and ensure that the platform was used by every stakeholder that would benefit from the insights that it identified.</p><p>In just a matter of weeks, 32 labels had been created and identified within conversations. Those labels refer directly to CRM fields allowing the AI to train itself in very precise classification patterns. This classification language model combined with an API means that raw text can be analyzed and key elements (like topics and entities) can be extracted.</p><p>Even with the reductions in time-to-market that Lettria offers, it can be easy to focus on the heavy-lifting that can be required at the start of any NLP project, but the advantages are clear to see once a project is up-and-running.</p><p>Whereas La Poste were previously only processing 3% of their calls, 75% of calls now have at least one label, and, going forward, Lettria will allow them to automatically process every phone call and input that data directly into their CRM.</p><p>This invaluable information has allowed La Poste to develop a more complete and sophisticated understanding of its customers, improve consistency and standardization across phone calls, and adapt its business offerings as a result of business insights that were identified through the analysis.</p><p>The La Poste use case is an excellent example of how a business can benefit from enriching their CRM with data from call transcripts. But, more than that, it is also an example of how important it is to choose an NLP option that allows both technical and non-technical profiles to take an active role in the project.</p><p>By choosing Lettria, La Poste was able to ensure that sales and customer success teams could contribute to the project, understand its value, and benefit from the insights that were being identified.</p><p>After all, as we discussed in the section above, although NLP projects are often thought of as the exclusive domain of data scientists and developers, the end-users of the information that they produce are often those with business profiles.</p><p>So that’s one example of how Lettria has been used to extract key information from the unstructured data of call transcripts, but not every company uses our platform for exactly the same use case or challenge.</p><h3>Improving processes and structuring patient data</h3><p>The applications for using natural language processing to analyze transcripts of spoken interactions actually goes far beyond calls from sales teams and customer support. There are instances where processing the unstructured data from other recordings and voice-to-text can also benefit your organization.</p><p>Similar to La Poste, Assistance Publique Hôpitaux de Paris (AP-HP) found themselves in a situation where they had more unstructured data than they could handle. With nearly 40 hospitals it is the largest university hospital in Europe and with 100,000 employees, it is also the largest employer in the Paris region.</p><p>All of those hospitals add up to a lot of patients. And a lot of patients means a lot of data. All of those patient reports contain crucial information that, in this instance, could very much mean life or death. But you don’t want doctors and nurses to be tied down dealing with generating reports when they could be focusing on the patients themselves.</p><p>Since 2020, AP-HP has been running the BoPA innovation chair that is focusing on identifying problems from the operating room in order to find human and technological solutions. This innovative project is based on 5 major blocks: the <em>Human Factor Block</em>, the <em>Viz Block, </em>the <em>Bot Block</em>, the <em>Light Block,</em> the <em>Touch Block,</em> and the <em>Box Block </em>(a name derived from the black box that is used for flight recordings on airplanes).</p><p>Their focus include the fields of surgeon-patient communication, surgical image capture, natural language analysis in the operating room, augmented reality using digital twice or fluorescent light, collaborative robotics or cobotics (design of collaborative robots), and the protection of operating room and patient data.</p><p>The project includes collaboration with a number of other organizations and institutions, including INRIA, Institut Mines-Telecom, and Université Paris-Sacaly and brings together leading researchers from the fields of virtual reality, digital twins, and artificial intelligence.</p><ul><li>The dictation had to be transcribed by hand by an assistant</li><li>The text then had to be put into a report on the hospital’s letterhead by a separate department</li><li>and then the report could be analyzed by a medical secretary to detect key variables about the patient which could then be added to their Electronic Health Record (EHR).</li></ul><p>One of the problems that this initiative identified was the fact that patient reports were very difficult to manage. They are generated at every stage of a patient’s journey (pre, peri, and post) and constantly changed in format and in the usage of technical terms.</p><p>Even something simple as a voice dictation required 3 steps:</p><p>This process required 25 minutes of human time per report and this was an area where the BoPA innovation chair felt that technology could speed up the process and reduce the amount of resources involved in the analysis and data input.</p><p>So, in February 2020, right before the world was about to be plunged into a global pandemic that would put an unprecedented strain on healthcare infrastructure and medical resources, AP-HP decided to work with Lettria to create an AI model specific for the healthcare sector.</p><p>This process took only three months, which was an incredible achievement given the circumstance in which it was being developed. With tens of thousands of reports generated each month, natural language processing and automatic language processing (ALP) stands to save AP-HP an incredible amount of time by automating the processing and analysis of each report.</p><h3>The problem isn’t unique, but the use cases can be</h3><p>This only goes to show how the application for natural language processing to enrich CRMs and other software with the information from unstructured data can be found in virtually every industry. This article has very much focused on the call transcript use case as it is the most prevalent, but there are many situations in which it can save an organization resources and help them to improve their processes.</p><p>It’s actually part of the reason why we create content like this. We know how powerful our solution is, but we don’t always know the problems that it can help our users to solve. By putting some examples out there and letting you think about your own business, there’s a good chance that new users will come to us with new use cases that we can help them with.</p><p>As with any good software, we’ve tried to create the best tool possible and we’re always happy to see the creative ways in which our users apply it to help them address their needs. But one thing is for certain, by specialiazing in the processing of unstructured data and simplifying the integration with all of the third-party tools that you already use, we’ve made the Lettria platform an incredibly powerful solution for finding key information that you’re currently missing.</p><h3>Start enriching your CRM with information from unstructured data</h3><p>We’ve explained the problem and shown you the solution, so what are you waiting for? If your company handles a large number of phone calls with leads or customers then it is absolutely certain that they can benefit from enriching their CRM with the information from unstructured data.</p><p>All too often we fall into the trap of thinking that processes and software solutions are good enough. They’ve got you this far, so why look for something better? But in this instance this is about turning the data that you already have into data that you can actually use.</p><p>All of those call transcripts contain business insights and trends that could make the difference for your next campaign, your next quarter, or your next product. And whilst you might already be using some form of natural language processing for some of your business needs, it’s unlikely that it has the capabilities required to make this process simple and automated.</p><p>Lettria’s decision to specialize in textual data processing makes it different to many of the off-the-shelf NLPs on the market. Its platform approach and functionalities also make it uniquely suited to use cases like this one.</p><p>So if you think that you could benefit from this type of project, then you should be clicking on one of those helpful links that we will have put into this article and getting in touch with our team. We can help you to develop a fully customized NLP platform in a fraction of the time that it would take to develop your own solution or even get a different NLP up and running.</p><p>We know, this is a long article with a lot of information in it, but that’s just the kind of thing we specialize in. You’ve handled the analysis on this occasion, let’s automate that process for your call transcripts.</p><p>‍</p><p><em>Originally published at </em><a href="https://www.lettria.com/articles/enriching-your-crm-with-unstructured-data?"><em>https://www.lettria.com</em></a><em>.</em></p><img src="https://medium.com/_/stat?event=post.clientViewed&referrerSource=full_rss&postId=3d89cd6a3e8e" width="1" height="1" alt="">]]></content:encoded>
        </item>
        <item>
            <title><![CDATA[Le scandale RGPD de H&M en Allemagne]]></title>
            <link>https://lettria.medium.com/le-scandale-rgpd-de-h-m-en-allemagne-77e875063307?source=rss-4f5319276610------2</link>
            <guid isPermaLink="false">https://medium.com/p/77e875063307</guid>
            <category><![CDATA[intelligence-artificielle]]></category>
            <category><![CDATA[crm]]></category>
            <category><![CDATA[nlp]]></category>
            <category><![CDATA[conformity]]></category>
            <category><![CDATA[rgpd]]></category>
            <dc:creator><![CDATA[LETTRIA]]></dc:creator>
            <pubDate>Tue, 16 Mar 2021 16:08:11 GMT</pubDate>
            <atom:updated>2021-03-16T16:08:11.541Z</atom:updated>
            <content:encoded><![CDATA[<figure><img alt="" src="https://cdn-images-1.medium.com/max/1024/1*4-Xpp-bopsje2181AWYIOw.jpeg" /></figure><blockquote>Le 1er octobre 2020, l’autorité de protection des données de Hambourg inflige au groupe suédois H&amp;M <strong>une amende de 35,25 millions d’euros</strong>. En cause ? Son centre basé à Nuremberg, qui constituait une importante base de profilage de ses salariés, à leur insu. Ce scandale illustre l’utilisation abusive de la donnée personnelle des collaborateurs, et nous rappelle l’importance de mettre en place des outils de modération des zones de commentaires libres.</blockquote><h3>Quels sont les faits reprochés à H&amp;M ?</h3><p>La décision rendue par l’autorité de protection des données de Hambourg à l’encontre du groupe H&amp;M concerne un site basé à Nuremberg.</p><p>Entre 2014 et 2019, <strong>la direction du magasin a opéré une collecte massive et intrusive de données de ses employés</strong>, à leur insu bien entendu :</p><ul><li>Des notes de correspondance étaient stockées en permanence sur le réseau ;</li><li>Les absences faisaient l’objet de « Welcome Back Talks ». Or à l’occasion de ces entretiens personnalisés, les managers pouvaient aller jusqu’à demander des informations liés aux symptômes et aux diagnostics de santé du salarié ;</li><li>Les managers profitaient aussi de discussions informelles pour obtenir encore plus d’informations sur la vie privée des salariés, comme leurs croyances religieuses ou leurs problèmes familiaux.</li></ul><p>Ces feedbacks d’entretiens et autres informations étaient ensuite stockées dans un fichier utilisé pour:</p><ul><li><strong>évaluer méticuleusement</strong> la performance des salariés ;</li><li><strong>profiler les salariés</strong> et pouvoir prendre des décisions à partir de ces profils.</li></ul><p>De surcroît, <strong>une cinquantaine de managers avaient accès à ces données exposant la vie privée des collaborateurs</strong>.</p><p>Cette grave violation des obligations applicables à la protection des données a eu de lourdes conséquences :</p><ul><li>la CNIL allemande a <strong>infligé une amende de 35,25 millions d’euros</strong> ;</li><li>le groupe a dû formuler <strong>des excuses publiques</strong> auprès de ses salariés et revoir entièrement la gouvernance des données.</li></ul><h3>Pourquoi le scandale n’a-t-il pu être évité ?</h3><p>Plusieurs raisons expliquent ce dérapage :</p><ul><li>Le dispositif était <strong>opaque</strong>, car les salariés n’étaient pas au courant de ces pratiques. Pour rappel, en France, un dispositif de surveillance et de profilage des salariés doit être soumis pour avis et consultation aux représentants du personnel.</li><li>La gouvernance était <strong>insuffisante</strong>. Les équipes existantes en charge de la conformité des traitements de données personnels n’ont sans doute pas pu jouer leur rôle de garde-fous . Dans ces conditions, cette situation aurait pu perdurer longtemps. Mais en 2019, une erreur de configuration de l’outil est à l’origine d’une fuite de données. Cette faille de sécurité a elle-même fuité dans la presse, attirant l’attention du régulateur allemand.</li></ul><h3>Que retenir de l’affaire H&amp;M pour votre conformité RGPD ?</h3><p>Cette expérience malheureuse n’est pas un cas isolé. Elle révèle les risques réels inhérents aux activités de votre entreprise.</p><ul><li>L’enrichissement de la connaissance du salarié est aujourd’hui une <strong>préoccupation croissante des Directions des Ressources Humaines</strong> qui veulent attirer de nouveaux talents et fidéliser les existants.</li><li>Cet objectif s’appuie de plus en plus sur la collecte de données non-structurées, en témoigne la **présence de champs de commentaires libres dans les SIRH **ou dans les outils de gestion de cv-thèques.</li><li>La saisie de données passe souvent sous les radars de surveillance. Souvent, <strong>aucun audit des bases de commentaires libres n’est effectué</strong>. Sans compter que les projets intrusifs tels que celui du site de Nuremberg ne sont pas toujours présentés aux équipes en charge de la conformité. La présence de données sensibles, de commentaires insultants ou subjectifs dans ces bases est un vrai enjeu pour votre conformité RGPD. Pour de nombreuses entreprises, cela a entraîné un véritable scandale médiatique auquel s’est ajouté la sanction du régulateur.</li></ul><h3>Comment maîtriser votre conformité RGPD grâce à la modération des commentaires libres ?</h3><p>Les technologies de reconnaissance de langage naturel liées au <em>Natural Language Processing</em> (NLP) sont un outil indispensable à intégrer dans votre kit de conformité RGPD. Elles prennent la forme de modules faciles à intégrer à vos outils (SIRH, CRM, ERP…). La conformité RGPD est plus acceptable par les directions métiers quand elle implique peu de contraintes ;</p><h3>Conclusion : Et vous, modérez-vous vos commentaires libres ?</h3><p>Etes-vous certain de la conformité RGPD de vos zones de commentaires libres ? N’attendez pas la fuite de données ou le scandale médiatique pour prendre en charge ce sujet épineux. D’autant que grâce à la reconnaissance des langages naturels, la modération de commentaires est aujourd’hui une tâche parfaitement automatisée.</p><img src="https://medium.com/_/stat?event=post.clientViewed&referrerSource=full_rss&postId=77e875063307" width="1" height="1" alt="">]]></content:encoded>
        </item>
    </channel>
</rss>