<?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 Valentina PF on Medium]]></title>
        <description><![CDATA[Stories by Valentina PF on Medium]]></description>
        <link>https://medium.com/@valentina.perezalonso?source=rss-ace5298c36e8------2</link>
        <image>
            <url>https://cdn-images-1.medium.com/fit/c/150/150/0*SG4NgawLe0aX-z_p</url>
            <title>Stories by Valentina PF on Medium</title>
            <link>https://medium.com/@valentina.perezalonso?source=rss-ace5298c36e8------2</link>
        </image>
        <generator>Medium</generator>
        <lastBuildDate>Sun, 22 Nov 2026 23:22:44 GMT</lastBuildDate>
        <atom:link href="https://medium.com/@valentina.perezalonso/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[How to get started with ecommerce analytics in 7 easy steps]]></title>
            <link>https://medium.com/@valentina.perezalonso/how-to-get-started-with-ecommerce-analytics-in-7-easy-steps-cf2ab9a68bec?source=rss-ace5298c36e8------2</link>
            <guid isPermaLink="false">https://medium.com/p/cf2ab9a68bec</guid>
            <category><![CDATA[ecommerce-data-analytics]]></category>
            <category><![CDATA[data]]></category>
            <category><![CDATA[ecommerce-data]]></category>
            <category><![CDATA[data-analytics]]></category>
            <category><![CDATA[ecommerce-analytics]]></category>
            <dc:creator><![CDATA[Valentina PF]]></dc:creator>
            <pubDate>Thu, 04 Aug 2022 11:29:26 GMT</pubDate>
            <atom:updated>2022-08-04T11:29:26.493Z</atom:updated>
            <content:encoded><![CDATA[<figure><img alt="" src="https://cdn-images-1.medium.com/max/1024/1*H3Wh48VeALPEt9m2j5WtQA.jpeg" /></figure><p>When running an ecommerce shop, it’s crucial to track and understand online shopping trends and changes in customer behaviors in order to make better decisions for your business. You not only need to attract new customers to your website, but you also need to keep them there and make sure they come back to purchase your products again.</p><p>How do you learn how to optimize your website and marketing efforts, what kind of sales and campaigns to run, and which products to stock or ditch? (Essentially, what do you do to guarantee the sustainable growth and success of your business?) You kickstart your ecommerce analytics journey, of course.</p><p>Ecommerce businesses start accumulating data the moment they begin operating. You can follow the behavior of your customers in your online shop in Google Analytics; your ecommerce platform (Shopify or Magento, for instance) holds your order lifecycle information. Google Ads, Facebook Ads, TikTok Ads, etc., all have insights into the performance of your marketing campaigns.</p><p>The data is there; you just need to learn how to leverage it and put it to good use. Through ecommerce analytics, you can inform your sales, marketing, and company-wide strategies, operating in a data-driven way that allows you to make informed decisions instead of playing a guessing game.</p><p>This is how you get started with ecommerce analytics:</p><ul><li>Centralize your data sources</li><li>Define your ecommerce KPIs</li><li>Take the time to understand your customers</li><li>Set up proper website tracking</li><li>Start analyzing your ecommerce data</li><li>Summarize and present your findings</li><li>Make sure you are using the right tools and processes</li></ul><figure><img alt="" src="https://cdn-images-1.medium.com/max/1024/0*k4pOCm0sKnunOnLi.png" /></figure><h3>Step 1: Centralize your data sources</h3><p>This might sound obvious, but the first thing you need to do when you start your ecommerce analytics journey is gather all of your shop’s data into one place. Your data will be scattered around different platforms and channels, so to enable a smooth analysis process, it’s crucial to centralize all of your data sources.</p><p>Our best practice recommendation would be to ingest all your data into a cloud data warehouse (like BigQuery or Snowflake). This way, your raw data is safely stored in an infinitely scalable space that can be easily accessed and connected with your data modeling and visualization tools. These would be the steps you’d follow when working on your data after having gathered all of your sources.</p><p>Many businesses use <a href="https://www.y42.com/blog/when-to-break-up-with-spreadsheets?utm_source=medium&amp;utm_medium=organic&amp;utm_campaign=ecommerce_analytics_article">spreadsheets</a> to collect and transform their data, especially in the earlier stages of their venture. However, if your company continues to scale, there will come a time when spreadsheets are no longer viable. When this happens, you might want to look into a more powerful analytics tool — like <a href="https://www.y42.com/?utm_source=medium&amp;utm_medium=organic&amp;utm_campaign=ecommerce_analytics_article">Y42</a> — that allows you to take care of all the relevant steps in your data pipeline.</p><h3>Step 2: Define your ecommerce KPIs</h3><p>Across an organization, teams usually work with various platforms, and each platform has a different definition and naming convention for the metrics they keep track of.</p><p>For instance, let’s say you want to compare the cost of your ads per day on Google Ads and Facebook Ads. You will find that Google names the column showing how much you spent as <em>cost, </em>whereas Facebook names it <em>amount spent.</em> When building your data model, merging both columns would be easier if they both had the same name. In situations like this, you would need to consolidate a unified naming convention.</p><p>As for confusing metric definitions, the way Shopify defines <em>net sales</em> might not match the way your business defines net revenue. Our internal understanding of <em>net revenue</em> is revenue excluding taxes, discounts, cancellations, and returns. Shopify, meanwhile, refers to <em>net sales </em>as revenue after discounts and returns but including taxes. In our opinion, Shopify’s definition is misleading because it includes taxes in the net sales metric.</p><p>However, it’s not just the platforms themselves that have diverse names and definitions for the same metrics. Internally, different teams might be looking at the same KPIs and naming or defining them in their reports completely differently.</p><p>If you don’t align as an organization on company-wide naming conventions and metric definitions, you risk finding yourself with team-specific silo reportings.</p><p>Our best practice tip is to come up with a key metrics definition glossary that reflects your business logic. This glossary will help you put fundamentals in place and ensure that everyone in the organization builds reports on the same base, following the same key metric principles.</p><figure><img alt="" src="https://cdn-images-1.medium.com/max/1024/0*QS6xZk_wa6Qf3-0N.png" /></figure><h3>Step 3: Take the time to understand your customers</h3><p>As an ecommerce or DTC business, your customers are your most valuable source of information. The way customers behave on your website, the orders they place (and the ones they don’t), the reviews they write for your products — these are all golden nuggets of data that will allow you to:</p><ul><li>Optimize your website for customer retention and improve it to provide the best customer experience possible.</li><li>Change or upgrade the products that are getting less love from your customers.</li><li>Promote your best-sellers.</li><li>Evaluate which products you should discontinue, which ones you should keep restocking, and what kind of new products your customers are likely to buy.</li></ul><p>But in order to make the most of your customer data, you need to know your customers inside out:</p><ul><li>Who is your target audience? Who is your target persona?</li><li>What are your customer demographics? How old are they? Where are they located?</li><li>What are your customers interested in? What do they like? What don’t they like?</li></ul><p>Knowing the answers to these questions will show you, for instance, which kind of social media channels (if any) you should promote your products on. Let’s say your customers are between the ages of 15 and 25 — in this case, TikTok might be the best place for you to promote your business.</p><p>What’s more, if you have customers located in different time zones, you might want to adjust your email marketing to send your messages at different times so that you reach your clients when they’re most likely to check their inboxes.</p><p>The same goes for the type of campaigns you run. If your customers are socially aware and interested in topics surrounding sustainability and the environment, you might want to design campaigns around these subject matters. You could highlight your sustainability practices, what you do to guarantee fair employee wages, how you handle your carbon footprint, etc. Showing your customers that you care about the same things as them will allow you to resonate with them more.</p><figure><img alt="" src="https://cdn-images-1.medium.com/max/1024/0*CmuTXUFmkwlgdDAb.png" /></figure><h3>Step 4: Set up proper website tracking</h3><p>As stated in <strong>step 3</strong>, your customers are your biggest source of data, so you want to be able to access that information. In order to understand the behavior of your customers in your online shop, you need to first be tracking it.</p><p>If you haven’t done so already, go ahead and set up your Google Analytics ecommerce tracking. ASAP! By tracking your customers’ journey in your webshop you can learn:</p><ul><li>Which products people are buying</li><li>How much money they are spending on your store on average (AOV)</li><li>What your conversion rate is</li><li>How much time it takes them to purchase an item</li><li>Which products they add to their carts but don’t purchase</li><li>Which discount codes they’re using (if any)</li><li>How long they spend on your website, which tabs they visit, how quickly they exit your website, and when they leave the website</li></ul><p>You can gather all of these insights and more by tracking your ecommerce website in Google Analytics.</p><h3>Step 5: Start analyzing your ecommerce data</h3><p>You’ve gathered your data, you’ve defined your metrics, you understand your customers, and you’ve set up your tracking, so now it’s time to look at the numbers and extract actionable insights from them.</p><p>After the data integration process comes the data transformation step. Make sure you clean, normalize, and unify the data coming from your different sources. During the modeling step, you should also be able to merge relevant data sources with each other and perform calculations that will enrich your analysis.</p><p><a href="https://www.y42.com/blog/the-ultimate-data-transformation-guide?utm_source=medium&amp;utm_medium=organic&amp;utm_campaign=ecommerce_analytics_article">Click here if you want to learn more about the benefits of transforming your data and discover some data transformation best practices.</a></p><p>Once you have a clean dataset, you can start working on your visualizations and use the dashboards you build to discover trends, anomalies, patterns, and actionable opportunities. Some typical ecommerce dashboards include:</p><ul><li>Website analytics dashboards</li><li>Revenue dashboards</li><li>Supply chain dashboards</li><li>Order lifecycle dashboards</li><li>Sales performance dashboards</li><li>Marketing performance dashboards</li></ul><figure><img alt="" src="https://cdn-images-1.medium.com/max/1024/0*Lv3x0LB_H-kZFxVj.png" /></figure><h3>Step 6: Summarize and present your findings</h3><p>Once you have interpreted the data, it’s time to organize your main findings in a more structured way. After conducting your analysis, you should be able to craft a story that presents the essential facts and learnings derived from it.</p><p>When you’re sharing your analysis with the team, the magic will happen when you start connecting the data insights you have gathered with your customers and their behavior. Remember, though, not to draw conclusions that go beyond what the data shows. Also, make sure you use attractive and easy-to-digest visuals in your reports so that you can explain your results and convey your message more efficiently.</p><p><a href="https://www.y42.com/blog/top-tips-for-powerful-data-driven-storytelling?utm_source=medium&amp;utm_medium=organic&amp;utm_campaign=ecommerce_analytics_article">Click here to gain all the insights you need on data-driven storytelling best practices.</a></p><h3>Step 7: Make sure you are using the right tools and processes</h3><p>The best way to get the most valuable insights from your ecommerce analytics data is to use the right tools and processes. You’ll have so much data to work with that you’ll need a place where you can centralize it all. Also, because you’ll be building so many reports for different teams and everyone will want to stay on top of their data, you will need to implement an <a href="https://www.y42.com/blog/how-to-automate-your-data-workflow-an-intro-to-data-orchestration?utm_source=medium&amp;utm_medium=organic&amp;utm_campaign=ecommerce_analytics_article">automation process</a> to keep everyone’s reports up to date.</p><p>Considering that you also have to clean and transform your data before you can actually visualize it, we recommend getting yourself an all-in-one data analytics tool — a tool that will enable you to integrate, model, visualize, automate, and export your findings, all in one single platform. Imagine how much time you would save by not having to worry about managing different tools to achieve one single purpose: working on your ecommerce data and guiding the company toward data-driven decision-making through your insights.</p><p>If you’re interested in learning more about how an all-in-one data analytics tool could help you step up your ecommerce analytics game, how about you consider <a href="https://www.y42.com/?utm_source=medium&amp;utm_medium=organic&amp;utm_campaign=ecommerce_analytics_article">Y42</a>?</p><p><a href="https://www.y42.com/form/discovery/book-a-call?utm_source=medium&amp;utm_medium=organic&amp;utm_campaign=ecommerce_analytics_article">Get in touch with our team to chat about how to make the most of your ecommerce analytics.</a></p><p><em>Originally published at </em><a href="https://www.y42.com/blog/get-started-with-ecommerce-analytics"><em>https://www.y42.com</em></a><em>.</em></p><img src="https://medium.com/_/stat?event=post.clientViewed&referrerSource=full_rss&postId=cf2ab9a68bec" width="1" height="1" alt="">]]></content:encoded>
        </item>
        <item>
            <title><![CDATA[5 best ways to improve your marketing analytics]]></title>
            <link>https://medium.com/@valentina.perezalonso/5-best-ways-to-improve-your-marketing-analytics-c8b3e8c0d3ed?source=rss-ace5298c36e8------2</link>
            <guid isPermaLink="false">https://medium.com/p/c8b3e8c0d3ed</guid>
            <category><![CDATA[data-analytics]]></category>
            <category><![CDATA[data-marketing]]></category>
            <category><![CDATA[improve-marketing]]></category>
            <category><![CDATA[analytics-marketing-data]]></category>
            <category><![CDATA[marketing-analytics]]></category>
            <dc:creator><![CDATA[Valentina PF]]></dc:creator>
            <pubDate>Thu, 04 Aug 2022 11:16:44 GMT</pubDate>
            <atom:updated>2022-08-04T11:16:44.268Z</atom:updated>
            <content:encoded><![CDATA[<figure><img alt="" src="https://cdn-images-1.medium.com/max/1024/1*NYnTRNcQ_60fjg8dHcpC1Q.png" /></figure><p>With a <a href="https://www.cambridgenetwork.co.uk/news/bango-launches-worst-ad-campaign-world">third of search ads</a> never reaching the promised audience, more than $60 billion of ad spend per year is predicted to be wasted. Despite having access to copious amounts of data about audiences and their behaviors, misplaced expenditure is still happening.</p><p>The volume of marketing analytics available to us is ever-growing, and the challenge of juggling insights from increasingly more touch points while still trying to stay on top of multiple data sources often leads to data-overwhelm, or worse, apathy.</p><p>You might think it’s easy to stay in your comfort zone, doing the same old type of marketing and investing in the same channels. But if your performance is stagnant, that isn’t really helping to drive growth — is it?</p><p>Now that we have the ability to track, monitor, and analyze virtually every possible type of audience interaction, why do we still not utilize this information to its fullest potential?</p><h3>The role of data analytics in marketing</h3><p>Data analytics in marketing is essentially the process of using data to find patterns and identify opportunities for improvement so you can make the best use of your resources.You can use it to see who your audience is, how they perceive your brand, and what types of messaging resonate with them.</p><p>With a data-driven strategy powered by strong marketing analytics, you can understand how prospects interact with your company and which of your marketing strategies have the strongest impact on their behaviors. This enables you to choose where best to focus your future marketing efforts to drive more conversions.</p><p>You’re likely already engaged with and performing some form of marketing analytics, be it email or ad campaign analysis, or even tracking your social media performance. But if you start looking more closely at your data, you can dig out invaluable insights that will allow you to further optimize your campaigns and drive up the marketing team’s value for the company.</p><h3>How can you improve your marketing analytics?</h3><p>You need to master marketing analytics so that you develop marketing campaigns and strategies that are fine-tuned to what your target demographic is actually interested in.</p><p>Here are our top five suggestions for improving your marketing analytics game:</p><figure><img alt="" src="https://cdn-images-1.medium.com/max/1024/0*0yF7Mk2tSM3VWB1Z.png" /></figure><h3>Focus on the right metrics</h3><p>When was the last time you reviewed the metrics or KPIs you collate? Perhaps you inherited the role of collecting and analyzing the marketing data and never questioned whether your data sources are still the right ones, or whether there are different types of data that your marketing team would benefit from analyzing.</p><p>We recommend you challenge the status quo. Just because the KPIs you’re analyzing are the metrics your organization has always collated, that doesn’t mean they’re the right ones (or the only ones). Ask the stakeholders that consume your analysis if there’s more information they would like to know that they’re not currently getting with the metrics you’re tracking. By gathering feedback and doing your own research, you can suggest new KPIs that might be more beneficial for the team in the long run.</p><p>For example, when analyzing the performance of their social media activity, many businesses become fixated on metrics like follower growth or post impressions — but do these actually communicate how well your efforts have performed against your business goals? Instead, focus on follower quality rather than quantity. Five new followers in your target persona sweet spot are more valuable than 50 new followers in unrelated fields (or worse, bots). Focus instead on analyzing <em>who</em> rather than <em>how many </em>are interacting, and consider engagement metrics such as comments and clicks over impressions and views.</p><h3>Ask relevant and insightful questions of your data</h3><p>Companies that struggle to get meaningful insights from their data are often not asking the right questions. The higher the quality of your questions, the more you’ll excel as a data analyst. Take a step back and think critically about the areas you really need to focus on. And be specific. For instance, looking at revenue increase is vague; focusing on the quarterly revenue per email (RPE) sent for your three key product lines is clear and specific.</p><p>Once you have a focus, you can identify what kind of information you need to know and what kind of analysis would be most beneficial for your use case. When it comes to data analytics, there are four types of analysis: descriptive, diagnostic, predictive, and prescriptive. Below are some examples of each type with questions that will allow you to drill down further into your data:</p><figure><img alt="" src="https://cdn-images-1.medium.com/max/1024/0*x4LCarhszNjGKEAa.png" /></figure><p>If you are unclear about what you want to know, always start with “what has happened?”. This is the foundation that will springboard you toward the next areas to delve into. Once you understand what has happened with your Facebook advertising, for example, you can move on to comparing and contrasting with the metrics across your LinkedIn and Twitter channels to identify patterns and trends. This will naturally lead to looking ahead and asking what actions you could take based on the analysis your data has revealed.</p><h3>Forecast trends and performance</h3><p>Speaking of identifying patterns and trends, becoming skilled at interpreting your marketing data to successfully forecast, make predictions, and plan ahead is truly going to raise your marketing analytics game.</p><p>Instead of just collecting data about your ad or email campaign performance and letting it gather dust once you’ve understood the ins and outs of what happened during specific campaigns, why don’t you put your historical data archive to good use? You can analyze your historical marketing data to identify trends and patterns in performance or audience behavior that can help you predict how your next campaign might perform and identify actions you could take now to change or improve the predicted results.</p><p>For example, seeing what day of the week or time of day you tend to get the most click-throughs from certain channels could help you determine when to schedule a new piece of content marketing to go live or send that next email campaign. Find out how gaming coaching specialist, Titan Academy, did exactly that by implementing Y42 and Google BigQuery, <a href="https://www.y42.com/customers/customer-story-titan-academy?utm_source=medium&amp;utm_medium=organic&amp;utm_campaign=marketing_analytics_article"><strong>tripling its email CTR</strong></a> to re-engage existing users.</p><h3>Optimize your marketing with proper attribution modeling</h3><p>So, now that you’re focusing on the right metrics and asking the most relevant questions, you can really start optimizing your marketing data. Attribution modeling is a great way to do this; it helps businesses track the steps a customer takes before a conversion in order to attribute that conversion to the correct source.</p><p>To get started, you need to choose which type of marketing attribution is right for your business. Single-click touch points such as the last click, first click, or last non-direct click are simpler and more straightforward. For example, when you do last-click attribution, only your customer’s very last touch point prior to the conversion receives 100% credit for the sale. Multi-touch attribution modeling is more sophisticated and allows you to look at the different interactions each customer had with your business or product before they converted, so you can see which touch points contributed to the success of that conversion.</p><p>These insights start joining the dots when you look at which marketing activities are having the most impact on your customer journey, arming you with the power to confidently adjust budgets and allocate media spend accordingly in order to achieve the strongest ROI based on real learnings.</p><p>How you set up and design your attribution modeling is key. Our <a href="https://www.y42.com/form/attribution-modeling-guide-v2?utm_source=medium&amp;utm_medium=organic&amp;utm_campaign=marketing_analytics_article">downloadable guide on attribution modeling </a>has everything you need to know about why attribution modeling matters and how to choose the right one for you.</p><h3>Automating data collection across platforms</h3><p>One of the biggest pain points for marketing analysts is connecting all the disparate data sources that make up your marketing efforts. It is a scenario that will be familiar to many of you: exporting your ad campaign results as CSV files from the different channels you used, uploading them to your spreadsheet, and then manually connecting the tables to analyze the overall performance of your ads. From Facebook, Twitter, Instagram, and LinkedIn, to Google Ads and email marketing platforms — the volume of data you need to integrate and tie together to be able to report on it in a meaningful way is huge. Plus, this manual effort is clunky and can get messy — all of which can be avoided.</p><p>A data analytics platform can do the heavy engineering work for you, easily connecting your marketing data sources and automatically syncing data from all your channels so you can monitor marketing performance and create reports hassle-free. A scalable data platform like <a href="https://www.y42.com/?utm_source=medium&amp;utm_medium=organic&amp;utm_campaign=marketing_analytics_article">Y42 </a>allows you to track every number back to its source with one click instead of dissecting SQL code, and gives you one secure place for loading, transforming, automating, visualizing, and exporting data.</p><h3>Step up your marketing analytics game</h3><p>There’s a school of thought that in marketing, there is a tendency to base decisions on the data that’s easily available, without considering what information may be missing. By implementing the five steps above, your marketing analytics will start to reveal newer, more insightful learnings that you didn’t even know you needed to know.</p><p>A strong marketing analytics strategy leads to data-driven marketing decisions that can drive growth, transform performance, and add value to the business. Value such as improved brand awareness, higher customer loyalty, and increased sales can all stem from having the insights to invest in the right marketing efforts and move away from those that are not delivering results.</p><p>Now you’ve got the tools to be a data superhero, enjoy soaring to new heights with your marketing analytics and seeing the difference you can make.</p><p><a href="https://www.y42.com/form/discovery/book-a-call?utm_source=medium&amp;utm_medium=organic&amp;utm_campaign=marketing_analytics_article">Want to step up your marketing analytics game? Book a call with our experts to see how our scalable data platform can support you to generate the business value you need.</a></p><p><em>Originally published at </em><a href="https://www.y42.com/blog/top-tips-marketing-data-analytics"><em>https://www.y42.com</em></a><em>.</em></p><img src="https://medium.com/_/stat?event=post.clientViewed&referrerSource=full_rss&postId=c8b3e8c0d3ed" width="1" height="1" alt="">]]></content:encoded>
        </item>
        <item>
            <title><![CDATA[Improve your data analytics in 5 easy steps]]></title>
            <link>https://medium.com/@valentina.perezalonso/improve-your-data-analytics-in-5-easy-steps-3bc40b927971?source=rss-ace5298c36e8------2</link>
            <guid isPermaLink="false">https://medium.com/p/3bc40b927971</guid>
            <category><![CDATA[data-analyst]]></category>
            <category><![CDATA[career-advice]]></category>
            <category><![CDATA[improve-data-analytics]]></category>
            <category><![CDATA[become-data-analyst]]></category>
            <category><![CDATA[data-analytics]]></category>
            <dc:creator><![CDATA[Valentina PF]]></dc:creator>
            <pubDate>Fri, 17 Jun 2022 02:21:25 GMT</pubDate>
            <atom:updated>2022-07-28T10:56:00.722Z</atom:updated>
            <content:encoded><![CDATA[<figure><img alt="" src="https://cdn-images-1.medium.com/max/1024/1*fnbhgo29bj4aUI_lJ4Y1Dg.png" /></figure><p>So you know how to write SQL queries, can explain spreadsheets to a child, know how to bring numbers to life in engaging graphs, and can write a few lines of code in Python. But is that all there is to being a data analyst?</p><p>Being capable of doing what is expected of you and being great at what you do are two different things. And we want you to be the latter: <em>excel</em> at what you do, always. (No pun intended.)</p><p>That’s why, in this article, you’ll find five tips on how to become a kick-ass data analyst:</p><ul><li>Check your data diligently</li><li>Think about speed</li><li>Understand the business logic</li><li>Master the craft of data storytelling</li><li>Use the right tools</li></ul><p>Perhaps you’re not a data analyst just yet, but would love to become one. If that’s you, then we’re already one step ahead. <a href="https://medium.com/@valentina.perezalonso/should-you-become-a-data-analyst-find-out-if-a-career-in-data-is-for-you-92f834ca6e93">We prepared an on the how-tos and whys around becoming a data analyst.</a></p><p>But for those of you who have gotten the basic skills down and know there’s always room for improvement, this article’s for you. Let’s dive in.</p><h3>Step 1: Check your data diligently</h3><p>A good data analyst checks their data sources once and then gets down to business. A <em>great</em> data analyst is very attentive to detail and double, or even triple-checks their data.</p><p>This might sound simple, but the truth is not everyone thinks triple-checking their data is necessary. Your rule of thumb should always be “there’s something wrong with this dataset” because more often than not, that will be the case.</p><p>People can be messy and overlook mistakes, or they might forget to update their sources in time and expect you to build them a dashboard with the wrong data. You need to be on your toes, scanning those datasets for any inaccuracies or errors and fixing them before jumping into the analysis.</p><p>That’s why the data transformation step is crucial. This is when you actually take the time to explore your data, clean it, filter it, and get it into shape before moving on to visualization.</p><p>However, you will need to make a judgment call on whether or not the errors are worth fixing. If fixing them will take hours out of your day, should you take this on? Or should someone else take care of this issue? Will it affect the end results if you overlook the problem? Depending on the use case you’re working on, you’ll know best which path to follow.</p><h3>Step 2: Think about speed</h3><p>If you’re constantly working with large amounts of data, you need to learn how to navigate your datasets fast. The quicker you are, the more analysis you can produce, and the more business insights you can generate. Your speed moves the company forward, but this isn’t to say that you should overwork yourself. It’s all about working smarter.</p><p>How can you achieve this? By mastering the art of SQL to surf your databases and retrieve the data you need, and then using as little code as possible to model and visualize it. Learning how to write code as fast as possible is your best bet for finding the information you need. Bringing that information to life with just a few clicks, without having to double think what the code should look like for that specific graph, is the recipe for lightning-fast deliveries. What’s better than being The Flash for your data team?</p><h3>Step 3: Understand the business logic</h3><p>If the business team asks you to build them a dashboard and you have all the sources you need, you just have to apply the technical skills you’ve honed over the years to clean their data and visualize it.</p><p>That’s all well and good, but what if you don’t understand what the business team actually does in your company? What if you don’t know what the market segmentation looks like? Who their ICP is? Who the competitors are? What MQLs, SQLs, and Opportunities are? In other words, what if you don’t understand the business logic behind the analysis you’re conducting?</p><p>Well, if you don’t understand how the business operates, if you don’t have a commercially driven mindset, if you don’t understand who the customers are and who the company is targeting, then you won’t be able to grasp and communicate the impact and importance of your analysis and how it can affect business decision-making.</p><p>If you just build dashboards when asked to do so, then you’re delivering insight to the teams. But if you build the dashboard knowing what the data is about and understanding the purpose behind the numbers, then you can deliver actionable insights — pieces of information that feed back into the teams’ strategies and operating plans and turn into steps to be followed.</p><p>However, delivering actionable insights doesn’t just come from understanding how the business functions. You also need to know how to communicate your findings in an effective way.</p><h3>Step 4: Master the craft of data storytelling</h3><p>As a data analyst, you need to be well versed in communicating both with computers and people. But why is communicating with people so important? You need to be prepared to present the results of your analysis, not only in a comprehensive way, but in a compelling way that drives people to take action based on the learnings you share. You have to find the sweet spot to make sure your data discoveries don’t go unused.</p><p>So what do you do? You turn to data storytelling. You summarize and present the most important facts and learnings derived from your analysis. You’re concise and structured, prepare what you’re going to say based on the audience you’re talking to, and you aid your story with easy-to-digest visuals.</p><p>But you never (and we mean it — <strong>never) </strong>draw conclusions that go beyond the data. A great data analyst presents the facts and suggests a multitude of possible interpretations, but they don’t make up fairytales. You should simply learn to deliver those facts in a way that inspires people to come up with business hypotheses and test them:</p><ul><li>“This is what the data says, and I <em>suspect</em> the reason behind this is …”</li><li>“The data shows this … and some <em>possible reasons</em> might be …”</li><li>“The results could <em>serve as an inspiration</em> for tests A, B, or C moving forward …”</li></ul><h3>Step 5: Use the right tools</h3><p>Getting into the habit of thoroughly checking and questioning your data, learning how to surf datasets as fast as possible, nailing your understanding of your company’s business logic, and becoming an expert at delivering data-driven business insights are all incredible tips for becoming the best data analyst you can be. But there’s only so much you can do on your own.</p><p>Honing your skills will only bring the results you want to see if you actually implement the right data analytics tools. Without a tool that enables you to smoothly integrate, model, and visualize your data, how are you going to become the Speedy Gonzales of data analytics? And let’s not even get started on automating your data workflows. What’s more efficient than not having to think twice about the actuality of your datasets and always being able to create up-to-date reports?</p><p>Our best practice solution here is to search the market for a full-stack, end-to-end data platform that not only helps you fulfill all the necessary steps to deliver stellar data analysis, but also enables you to shine as the kick-ass data analyst you know you can be.</p><p>One such tool is . It’s a data platform that allows you to put your SQL knowledge into practice while taking advantage of the no-code interface that lets you integrate, orchestrate, and even visualize data in just a few minutes.</p><p><a href="https://www.y42.com/form/discovery?utm_source=medium&amp;utm_medium=organic&amp;utm_campaign=become_better_data_analyst_blog">Want to learn more about Y42? Don’t hesitate to get in touch with our team.</a> Remember, great data analysts always go after the solutions that will help them work smarter, not harder.</p><h3>Bonus: Keep up with the times</h3><p>Before we let you go, here’s a little bonus tip from our very own data experts: <strong>“Stay up to date with the latest happenings in the data space.”</strong> Meaning: consume as much data analytics-related content as you can to ensure you’re always one step ahead of the rest.</p><p>Some of our team’s recommendations include:</p><ul><li><strong>Podcasts: </strong>Analytics Anonymous, The Sequel Show, DataTalks.Club, The Analytics Power Hour, The Analytics Engineering Podcast, The Data Chief, The Marketing Analytics Podcast (by Supermetrics), and The Banana Data Podcast.</li><li><strong>Data Twitter accounts: </strong>@pedram_navid, @WeHateData, @mwong023, @_abhisivasailam, @sarahmk125, @Mike_Kaminsky, @tayloramurphy, @clairebcarroll, @sethrosen, and @rahulj51.</li></ul><p><em>Originally published at </em><a href="https://www.y42.com/blog/how-to-become-a-great-data-analyst-in-5-steps"><em>https://www.y42.com</em></a><em>.</em></p><img src="https://medium.com/_/stat?event=post.clientViewed&referrerSource=full_rss&postId=3bc40b927971" width="1" height="1" alt="">]]></content:encoded>
        </item>
        <item>
            <title><![CDATA[How to automate your data workflow: An intro to data orchestration]]></title>
            <link>https://medium.com/@valentina.perezalonso/how-to-automate-your-data-workflow-an-intro-to-data-orchestration-a4b033a95bef?source=rss-ace5298c36e8------2</link>
            <guid isPermaLink="false">https://medium.com/p/a4b033a95bef</guid>
            <category><![CDATA[data-orchestration]]></category>
            <category><![CDATA[data-pipeline]]></category>
            <category><![CDATA[data-automation]]></category>
            <category><![CDATA[data-workflow]]></category>
            <category><![CDATA[data-pipeline-automation]]></category>
            <dc:creator><![CDATA[Valentina PF]]></dc:creator>
            <pubDate>Wed, 15 Jun 2022 20:23:40 GMT</pubDate>
            <atom:updated>2022-07-28T10:48:48.960Z</atom:updated>
            <content:encoded><![CDATA[<figure><img alt="" src="https://cdn-images-1.medium.com/max/1024/1*7yzZLvJgqh62cPpBMV_WPQ.png" /></figure><p>With big data comes big results, but not without challenges. Sifting through huge volumes of information, checking if your integrations, models, and sources are up to date, and constantly retracing your steps can leave you thinking <em>so much data, so little time</em>. But what if there was another way? Drum roll, please . . . enter data orchestration.</p><p>Designed with the modern data analyst in mind, data orchestration simplifies your data journey. Imagine each of your data pipeline components running exactly when they should. It’s not even a problem when your integration takes longer to update, because the components that are dependent on that data will wait until the update is done before they run. Doesn’t that sound like an ideal scenario?</p><p>If you’re currently juggling multiple data workflows, stick with us. We’re here to tell you all about data pipeline orchestration, the pain points it alleviates, and how it will transform the way you use data technology.</p><h3>What is data orchestration?</h3><p>In a nutshell, data orchestration refers to the process of coordinating the execution of your data workflows. It allows you to automate your updates, use your data more effectively, and ultimately, streamline data-driven decision-making. Basically, data orchestration ensures the seamless integration, scheduling, and movement of your data workflows by creating dependencies between the different components of your pipeline.</p><p>Picture it like this: You have a model with an output table that showcases the daily ad spend on Tiktok, Facebook, and Google. So, the integrations that feed into this model are Tiktok Ads, Facebook Ads, and Google Ads. And because you want to know the amount of money you’re spending on ads on a daily basis, you need to update the entire pipeline every day. But it would be a pain to do that manually, wouldn’t it? So in situations like this, you orchestrate your data pipeline and schedule a daily integrations update which will then trigger an update of your model — this results in you always having the latest information in your output table.</p><p>How does a data orchestration system achieve this? Typically, it uses directed acyclic graphs (DAGs) to establish the workflows. For the non-data nerds among us, DAGs are a collection of all the tasks you’d like to run, organized by their relationships and dependencies, and customized to suit your data workflow. Using DAGs, data orchestration tools enable you to select the data points between which your data should automatically update.</p><h3>How can data orchestration help you?</h3><p>Now we’re getting to the good part: all the ways that data orchestration saves you time, gives you control over your data, and keeps you one step ahead of the competition. Trigger warning . . . let’s talk pain points!</p><ul><li><strong>Spotting errors</strong>: There’s nothing worse than producing a report or forecast only to discover the data it’s based on is outdated or incomplete. Backtracking and chasing minor mistakes eats into time and energy that could be better channeled elsewhere — like actually deriving insights from your data and influencing business decisions. With built-in debugging tools as well as monitoring and alerts, data orchestration tools — along with platforms that offer data orchestration — immediately inform you of any issues and offer tips on how to resolve them.</li><li><strong>Scheduling updates in the right sequence</strong>: 1am, 3am, 5am . . . you know the drill! Keeping your data pipeline moving is no mean feat — not only do you need to run workflows at certain time intervals or based on a certain logic, but you need to execute data tasks in the same order each time to maintain accuracy. Data orchestration handles this for you automatically, creating and connecting the necessary dependencies and updating the data between the different elements of your pipeline. Gone are the days spent setting 10 separate schedules! With data orchestration, you only need to set a single schedule.</li><li><strong>Strengthening data management</strong>: With fragmented data sources, models, and automations, it’s difficult to create and maintain your data workflows, as well as trace a calculation back to the data they used. And if you can’t easily spot and fix errors, or monitor how your data is being processed, how can your business use its data effectively? A robust data orchestration system gives you full oversight of your pipeline and its processes, ultimately strengthening your data management so you can unleash its true value.</li></ul><h3>Orchestrate your data pipeline with Y42</h3><p>You now know how data orchestration works and why it’s beneficial for you to start orchestrating your data pipelines. But in order to get there, you need to find a data analytics tool that will allow you to complete this function.</p><p>Have you heard about ? This full-stack data platform offers you the flexibility to set up your entire data pipeline, all in one place. First, you your data, then you it, and then you set up any <a href="https://www.y42.com/product/automate?utm_source=medium&amp;utm_medium=organic&amp;utm_campaign=data_orchestration_blog">data exports </a>you might need. Once all of these steps are complete, the <a href="https://www.y42.com/product/orchestrate?utm_source=medium&amp;utm_medium=organic&amp;utm_campaign=data_orchestration_blog">data orchestration</a> engine is ready to come in. It can actually be set up in just a few clicks:</p><p>1. Once you’re in the orchestration layer, you’ll be prompted to create a DAG made up of the data components you wish to analyze.</p><figure><img alt="" src="https://cdn-images-1.medium.com/max/600/0*qA6YfvEQm4if-g_y.gif" /></figure><p>2. The ‘Structure’ tab then allows you to pull in your desired components individually, or you can use the auto-generate function to automatically connect your data sources and models.</p><figure><img alt="" src="https://cdn-images-1.medium.com/max/600/0*Vn1nl_oDEEsjIYj5.gif" /></figure><p>3. Once you’ve committed your orchestration, you’ll be able to schedule its run time. This can be as frequently as every 30 minutes, every hour, every day, or even weekly — whenever you want.</p><figure><img alt="" src="https://cdn-images-1.medium.com/max/600/0*UoYZ3-DarL5dqzBW.gif" /></figure><figure><img alt="" src="https://cdn-images-1.medium.com/max/600/0*SbMVq93AM6YLqdKX.gif" /></figure><p>4. You can keep track of your data orchestrations in the ‘Overview’ tab, which indicates when each DAG ran and for how long.</p><figure><img alt="" src="https://cdn-images-1.medium.com/max/1024/0*aYctHvNG6GNtfwi2.gif" /></figure><p>5. For a more in-depth overview, Y42’s data lineage layer shows every aspect of your data pipeline all at once, so that you can trace every single step.</p><figure><img alt="" src="https://cdn-images-1.medium.com/max/600/0*ZBDju6dB8UpiE6sD.gif" /></figure><p>This automated nature of Y42’s orchestration layer is designed to ensure your data is as accurate as possible. Each stage of the data orchestration process must be completed for the tool to move on to the next one. Should the system run into an error along the way, you’ll be notified immediately — aiding data management by giving you the chance to tackle any issues before they have an impact.</p><p>No more frantically switching between programs to revise and refresh all your data! Once each of your dependencies is ready to go, your orchestration can move forward. All you need to do is decide which output table to use and when you want that data. It’s really that simple.</p><p>Got any questions? Want to know how orchestrating your data could benefit your specific use case? <a href="https://www.y42.com/form/discovery?utm_source=medium&amp;utm_medium=organic&amp;utm_campaign=data_orchestration_blog">Then get in touch with the Y42 team.</a> Our data experts would be more than happy to clarify any of your doubts.</p><p><em>Originally published at </em><a href="https://www.y42.com/blog/how-to-automate-your-data-workflow-an-intro-to-data-orchestration"><em>https://www.y42.com</em></a><em>.</em></p><img src="https://medium.com/_/stat?event=post.clientViewed&referrerSource=full_rss&postId=a4b033a95bef" width="1" height="1" alt="">]]></content:encoded>
        </item>
        <item>
            <title><![CDATA[Top tips for powerful data-driven storytelling]]></title>
            <link>https://medium.com/@valentina.perezalonso/top-tips-for-powerful-data-driven-storytelling-c772d0b44e33?source=rss-ace5298c36e8------2</link>
            <guid isPermaLink="false">https://medium.com/p/c772d0b44e33</guid>
            <category><![CDATA[data-analytics]]></category>
            <category><![CDATA[data-visualization]]></category>
            <category><![CDATA[data-insights]]></category>
            <category><![CDATA[data-driven-storytelling]]></category>
            <category><![CDATA[data-storytelling]]></category>
            <dc:creator><![CDATA[Valentina PF]]></dc:creator>
            <pubDate>Wed, 15 Jun 2022 19:23:45 GMT</pubDate>
            <atom:updated>2022-06-15T20:40:29.553Z</atom:updated>
            <content:encoded><![CDATA[<figure><img alt="" src="https://cdn-images-1.medium.com/max/1024/0*gscLrbPC2ulD91b9.jpg" /></figure><p>Do you struggle to powerfully communicate the insights behind your data? Is it a challenge getting the business to take action using the data at your disposal?</p><p>Data storytelling is the critical ‘last mile’ of communication that transforms your findings into actionable insights that will drive change and add value.</p><p>You might be great at collating and analyzing data, but if you don’t effectively channel this data into a compelling and actionable communication, your data insights will go unnoticed. Done well, data-driven storytelling can transform your organization and create lasting change, resulting in a more responsive, agile, and business-resilient operation.</p><h3>The power of data storytelling</h3><p>Stories beat statistics in two ways: One, they are more <em>memorable </em>than statistics, and two, they are more <em>persuasive</em>. The human brain can be influenced to make decisions through both logic and emotion, so data alone won’t cut it — we need the story behind the numbers to be the bridge that connects our emotional side with our logical side.</p><p><em>“No one ever made a decision because of a number, they need a story,” </em>Israeli-American psychologist and economist Daniel Kahneman famously said — and it’s true. Our brains expect and want a coherent story. It’s not enough to just provide insights; you need to craft an inspiring message that evokes action.</p><p>However, there is an art and science to data storytelling, and this starts with three pillars: narrative, data, and visuals. Combining data with a narrative helps you explain the insights in a way that’s easy to understand and remember. Data visualization brings the data to life and helps people see the trends and patterns that are often missed when just looking at rows and columns of figures.</p><p>But how do you actually ‘do’ data storytelling and where should you start? Read on for our top considerations for doing data storytelling well.</p><h3>1. Define your objectives and know your audience</h3><p>Before crafting your story, you need to understand who your audience is and what the goal of your data storytelling is. Who are you presenting the story to and what kind of metrics are important to them? Is it a specific team within your company or the board of directors? The kinds of things the senior management and other business influencers need to know will differ from what your own team or manager will want to hear. You may need to tell different stories from the same set of data, depending on your audience.</p><p>Think about what your different audiences will want to know and how they will want this presented to them. For example, if you’re talking to the finance team, ROI and cost-saving metrics will be more relevant, so shape your data insights around them. If you’re presenting to your sales and marketing colleagues, the most important KPIs for them might be MQLs, SQLs, opportunities, and cost per lead.</p><p>Tailoring your data sharing to your audience will help influence the message you convey and how you share it.</p><h3>2. Use the data storytelling arc</h3><p>A good story is always carefully structured, and data storytelling is no different. The narrative you want to communicate can be thought through using the following storytelling arc. Essentially, you start by setting the scene, creating a hook to capture your audience’s attention, contextualizing the business problem, and building up the tension to the big reveal — the key piece of data or insight you want to share and the action you want the audience to take:</p><ul><li><strong>Set-up &amp; hook</strong> — this is where you provide business context and background. For example, when presenting the marketing team’s quarterly performance, your context and background would be what happened in the previous quarter and what the team is looking to achieve, change, or discover in this quarter.</li><li><strong>Rising insights</strong> — these are the supporting details that reveal deeper insights into the problem or opportunity. It’s a bit like when you peel an onion — you reveal more layers. This is where you’d dive deeper into the marketing data to reveal any challenges or key discoveries, and provide the details of the channels you tested, the campaigns you ran, and how many leads they generated.</li><li><strong>‘Aha!’ moment</strong> — the major finding or central insight. This is essentially the key thing you want your audience to remember, and the action you recommend they take. Carrying on with the marketing example, this might be the fact you have realized Google search isn’t working for the team when it comes to generating leads, so you’re recommending not putting so much budget into Google campaigns and trying out Twitter ads instead due to the campaign results there.</li></ul><figure><img alt="" src="https://cdn-images-1.medium.com/max/1024/1*hSCrLbdjwXbr6Ikr22ADjA.png" /></figure><h3>3. Consider the right data visualization</h3><p>As the data storyteller, your job is to make it as easy as possible for people to consume the data you’re sharing without fatiguing them with too many stats. Illustrate your narrative with data visuals that can quickly provide the insights your audience wants, enhancing your storytelling and making your data more compelling.</p><p>Be sure to choose formats that are easy to digest and follow, that don’t need much explanation. Let the visuals do the talking. That way, your ‘aha!’ moments will be more impactful and memorable. You should also think about the right type of data visualizations to use to display different types of information. Bar charts are great for comparisons, while line graphs are better for showing trends. Likewise, scatter plots are ideal for displaying the relationship between two quantitative variables such as ad spend and revenue.</p><figure><img alt="" src="https://cdn-images-1.medium.com/max/1024/1*ujVtQ-XnAgtNUNVgOVmUUw.png" /></figure><h3>4. Be objective and transparent</h3><p>Your data storytelling should present the facts in an unbiased way, even if you’re using the data to support a specific point of view. Never draw conclusions that go beyond the data and always support your claims with the insights the data shows. Stretching the facts or adding interpretation — intentionally or unintentionally — can diminish your findings and reduce the audience’s trust in your information.</p><p>For example, let’s say the marketing team saw a 30% increase in leads from one month to the next, then a decline of 30%. You know the team tested a campaign on a new channel during the month of the increase, so a possible objective reason for the subsequent decline would be that they stopped running ads on that new channel. However, adding your own thoughts — for instance, that outside macroeconomic events such as war, inflation, or high unemployment levels have had an impact — would be stretching the facts and viewed as added interpretation.</p><p>Support your claims with the insights the data shows and avoid sweeping generalizations. Remember, the goal of your data storytelling is to inspire change, action, or transformation based on real data.</p><h3>5. Wrap up with a strong conclusion and call to action</h3><p>When sharing your data insights with the team, don’t forget to end your story by sharing potential actions and recommendations. Link back to the objectives you defined at the start of your data storytelling and share your suggestions for what actions could be taken as a result of your findings. If there isn’t a clear outcome and it requires further dialogue and discussion, make that the conclusion of your presentation. Your data storytelling could be the catalyst that sparks an important conversation between key decision-makers within your organization, inspiring them to take action or consider important and lasting change.</p><h3>Data storytelling done well</h3><p>Combining the right data with the right narrative and visuals creates a powerful story that could change attitudes and behaviors within your business, inspiring and motivating team members to make data-driven decisions and test new business hypotheses. What’s more, data storytelling is an essential skill in today’s data-driven economy, and one that can truly drive lasting business transformation.</p><p>Want to tell a better story with strong data visualization? <a href="https://www.y42.com/form/discovery?utm_source=medium&amp;utm_medium=organic&amp;utm_campaign=data_storytelling_blog">Book a call with our data experts at Y42</a> to find out how to build dashboards and charts that support your data storytelling with compelling visuals.</p><p><em>Originally published at </em><a href="https://www.y42.com/blog/top-tips-for-powerful-data-driven-storytelling"><em>https://www.y42.com</em></a><em>.</em></p><img src="https://medium.com/_/stat?event=post.clientViewed&referrerSource=full_rss&postId=c772d0b44e33" width="1" height="1" alt="">]]></content:encoded>
        </item>
    </channel>
</rss>