<?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 Sofiya Urumova on Medium]]></title>
        <description><![CDATA[Stories by Sofiya Urumova on Medium]]></description>
        <link>https://medium.com/@urumovasoph?source=rss-542c7df2297f------2</link>
        <image>
            <url>https://cdn-images-1.medium.com/fit/c/150/150/1*PfamHytxbkwwqI5U5oXevA.png</url>
            <title>Stories by Sofiya Urumova on Medium</title>
            <link>https://medium.com/@urumovasoph?source=rss-542c7df2297f------2</link>
        </image>
        <generator>Medium</generator>
        <lastBuildDate>Tue, 17 Nov 2026 03:54:03 GMT</lastBuildDate>
        <atom:link href="https://medium.com/@urumovasoph/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[Applying Service Design Thinking to Your Design Process]]></title>
            <link>https://medium.com/@urumovasoph/applying-service-design-thinking-to-your-design-process-c7457599f941?source=rss-542c7df2297f------2</link>
            <guid isPermaLink="false">https://medium.com/p/c7457599f941</guid>
            <category><![CDATA[design-strategy]]></category>
            <category><![CDATA[template]]></category>
            <category><![CDATA[service-design]]></category>
            <category><![CDATA[design-thinking]]></category>
            <category><![CDATA[product-design]]></category>
            <dc:creator><![CDATA[Sofiya Urumova]]></dc:creator>
            <pubDate>Wed, 04 Sep 2024 13:31:23 GMT</pubDate>
            <atom:updated>2025-10-08T15:40:51.130Z</atom:updated>
            <content:encoded><![CDATA[<p>I first heard of Service Design during my post graduate program at <a href="https://www.arts.ac.uk/subjects/communication-and-graphic-design/postgraduate/graduate-diploma-graphic-design-chelsea">UAL</a> back in 2011. One of my assignments was to represent the core principles of Service Design, which include a cohesive Customer, Company and Employee Experience.</p><figure><img alt="" src="https://cdn-images-1.medium.com/max/1024/1*H1K_mBWd5dsqrt6rthzjww.png" /><figcaption>Service Design Simplified</figcaption></figure><p>Although, I truly enjoyed learning about Service Design in University, it was not until a few years later that I was able to deeply explore it at the workplace.<br>Service Design taught me to always take a step back and to explore every problem with a futuristic mindset.</p><p>To paint the big picture, let’s start with the basics.<br>Many of you may be familiar with the term Service Design but for those of you, who aren’t, we‘ll review the fundamentals before we dive into its application in the overall design process.</p><h3>So What is Service Design?</h3><p>Service Design is a fairly new discipline as the term was coined by Lynn Shostack in 1982, <a href="https://www.nngroup.com/articles/service-design-101/"><strong>History of Service Design | Service Design 101 NN/g</strong></a><strong>.</strong><br>There are many definitions of Service Design and one of my favorite is <a href="https://medium.com/u/d61f63103c5f">Daniele Catalanotto</a>’s, from <a href="https://store.swissinnovation.academy/course-what-is-service-design"><strong>Swiss Innovation Academy | What is Service Design?</strong></a><strong> </strong>about the interactions between humans and organizations.</p><blockquote>Service Design means to plan or create interactions between humans and organizations. Service Designers aim to bring together people, infrastructure, communication, and material.</blockquote><p>An interaction designer by heart, I am biased in my preference for a definition. For the user experience designers in the room, consider the following</p><blockquote>User experience design revolves around the interactions between the customer and the organization, while service design focuses on the interactions between the people and processes, within the organization, responsible for delivering the customer experience of a product or a service.</blockquote><p>NN/g do a wonderful job comparing the two in <a href="https://www.nngroup.com/articles/ux-vs-service-design/"><strong>UX vs. Service Design</strong></a>.</p><figure><img alt="" src="https://cdn-images-1.medium.com/max/1024/1*IoPXe8439Avr6N6a86mLoQ.jpeg" /><figcaption><a href="https://unsplash.com/photos/c2Y16tC3yO8">Two cups of espresso by Nathan Dumlao</a></figcaption></figure><p>One of my favorite summaries of Service Design comes from the great <a href="https://medium.com/u/11f71c90f14c">Marc Fonteijn</a> in his article <a href="https://www.servicedesignshow.com/what-is-service-design/"><strong>What Is Service Design: The Final Answer</strong></a>:</p><blockquote>Imagine you have two coffee shops right next to each other selling the exact same coffee at the exact same price. Service design is what makes you walk into the one and not the other, come back often and tell your friends about it.</blockquote><p>If you want to learn more about how the coffee is made, go a step further and use the right <a href="https://servicedesigntools.org/"><strong>Service Design Tools</strong></a> for the job.</p><figure><img alt="" src="https://cdn-images-1.medium.com/max/1024/1*DSupWQ7CejtgZVlo84RRKg.png" /><figcaption><a href="https://servicedesigntools.org/tools">Service Design Tools Map</a></figcaption></figure><h3>How to Visualize the Different Levels of Service Design ?</h3><p>To better understand how Service Design is different than User Experience Design, let’s visualize it in a single diagram.</p><p>My personal favorite is <a href="https://medium.com/u/777fe7ae995f">Mariel Macia</a>’s from <a href="https://medium.com/die-krieger-des-lichts/service-design-excuse-me-what-b0efe1ef228a"><strong>Service Design. Excuse me… what?</strong></a></p><figure><img alt="" src="https://cdn-images-1.medium.com/max/1024/1*9LKQn-OcA9tk-m99a8MKsg.png" /><figcaption>Service Design in a nutshell: Customer Journey, Fontstage, Backstage and Processes</figcaption></figure><h4>What about the Customer Journey?</h4><p>One common denominator across Design disciplines, Social Science and Business is the customer journey. Everyone wants to get closer to their target customer and build experiences that enhance their journey.<br>Service Designers are experts in mapping customer journeys and they use different tools to speed up their process. Even if you start with posted notes on a board, sooner or later you are faced with digitalizing the process. To be more efficient in doing so, take a look at <a href="https://servicedesigntools.org/tools/journey-map"><strong>the Service Design Tools Journey Map</strong></a> and <a href="https://www.theydo.com/service-design"><strong>TheyDo</strong></a><strong>. They Do </strong>is a magic product for Customer Journey Management, which becomes an essential part of the process when multiple journeys are created.<br>If you want to learn more about the scale of customer journey management, head to <a href="https://www.nngroup.com/videos/what-is-journey-management/"><strong>What is Journey Management NN/g</strong></a>.</p><h4>What is a Service Blueprint?</h4><p>While the customer journey map focuses on the pain points and opportunities of the target customer, the Service Blueprint goes a little deeper and also analyses the organizational processes underneath.</p><p>Blueprints are invaluable in painting a high level picture of the business, assuring alignment across teams and processes and identifying future areas of improvement.</p><figure><img alt="" src="https://cdn-images-1.medium.com/max/1024/1*ngnpAWQOvNAH4P_swCN2CA.png" /><figcaption>Airbnb’s Blueprint as <a href="https://twitter.com/bchesky/status/1654137863007612930?s=20">tweeted by Brian Chesky</a></figcaption></figure><p>Figma is often the tool of choice of many Product Designers. If you’re like me and you enjoy dabbing into both Product and Service Design, I recommend the use of a <a href="https://servicedesigntools.org/tools/service-blueprint"><strong>Service Blueprint Template</strong></a>.</p><p>To streamline my blueprints, I designed an <a href="https://www.figma.com/community/file/1411041133022608043/service-blueprint-auto-layout-template-and-prototype"><strong>easy-to-tweak auto-layout template</strong></a> and added it to <a href="https://www.figma.com/@sofiyaurumova"><strong>Figma Community</strong></a>. Make sure to make a copy of it and leave your feedback.</p><p>To provide an actual example of the Blueprint template, I recorded the usual activities of a dog handler in such a manner so that experienced and first-time dog owners can get a grasp of the process in a glance. As illustrated, Blueprints can be applied to everyday tasks and processes as well as professional projects.</p><figure><img alt="" src="https://cdn-images-1.medium.com/max/1024/1*ajqDhO5dcEnwJD6nRq8RCQ.png" /></figure><p>Use blueprints as the main sources of truth between your executive stakeholders and your Research, Development and Delivery teams.</p><h4>Final thoughts</h4><p>If you’re passionate about problem solving and you’re curious to piece the big picture together, service design is here to help you. <br>If you find this article interesting and you want to learn more about Service Design join the <a href="https://www.service-design-network.org/"><strong>Service Design Network</strong></a> and read <a href="https://www.service-design-network.org/books-and-reports/service-design-from-insight-to-implementation"><strong>Service Design from Insight to Implementation</strong></a>. This book is an invaluable resource and it provides proven strategies for success such as examples of using service blueprints to make more informed investment decisions.</p><figure><img alt="" src="https://cdn-images-1.medium.com/max/662/1*RM9MfR6lm6txSfJK0Hus5g.png" /></figure><p>I hope that you find the inclosed resources helpful and that you share in the comments’ section how applying service design thinking to your day to day design process changed your perspective of the problems you solve.<br>Enjoy!</p><img src="https://medium.com/_/stat?event=post.clientViewed&referrerSource=full_rss&postId=c7457599f941" width="1" height="1" alt="">]]></content:encoded>
        </item>
        <item>
            <title><![CDATA[Machine learning, data annotation and text analytics for designers]]></title>
            <link>https://uxdesign.cc/artificial-intelligence-and-design-d449dfd527a4?source=rss-542c7df2297f------2</link>
            <guid isPermaLink="false">https://medium.com/p/d449dfd527a4</guid>
            <category><![CDATA[user-experience]]></category>
            <category><![CDATA[ai]]></category>
            <category><![CDATA[product-design]]></category>
            <category><![CDATA[design-systems]]></category>
            <category><![CDATA[artificial-intelligence]]></category>
            <dc:creator><![CDATA[Sofiya Urumova]]></dc:creator>
            <pubDate>Mon, 06 Apr 2020 12:31:00 GMT</pubDate>
            <atom:updated>2020-04-22T00:28:14.735Z</atom:updated>
            <content:encoded><![CDATA[<h4>About Artificial Intelligence and Design.</h4><p>Artificial Intelligence has been a trending topic for the last 5+ years. It had its peek for <a href="https://trends.google.com/trends/explore?date=2015-01-01%202020-01-18&amp;geo=US&amp;q=artificial%20intelligence">a most searched term on Google in January, 2018</a>. I had the chance to join the front line in April, 2018 and I have been honored to do my small part in the contemporary Machine Learning landscape.</p><p>Often when people refer to Artificial Intelligence, they mean Machine Learning. Machine Learning is a scientific study explored in detail by many an expert in the field. In simple terms, it is the process of teaching a machine to deduct information in order to complete given tasks. It may be a simple task as classifying a paragraph of text or a more complex one as <a href="https://en.wikipedia.org/wiki/Named-entity_recognition">named-entity recognition</a>.</p><p>For further readings on the subject, I recommend these helpful resources: <a href="https://mitpress.mit.edu/books/machine-learning">Machine Learning from The MIT Press Essential Knowledge series</a> and <a href="https://www.awwwards.com/AI-driven-design">the Artificial Intelligence Driven Design Series from Awwwards.</a></p><h3>The project</h3><p>In order to learn, the machine needs to be taught through training data.</p><p>For that project, the training data was written text that varied from online product reviews to survey responses to chat bot conversations and others.</p><blockquote>The process of transforming the original data into training data involves manual annotations of the text. This is a long, painstaking and error-prone approach.</blockquote><figure><img alt="" src="https://cdn-images-1.medium.com/max/1024/1*nppTFYTshoVM1TSVUZK4SQ.png" /></figure><h3>The problem</h3><p>How to annotate data fast and produce high quality text annotation on a large scale?</p><p>There are existing tools that offer rapid annotations such as <a href="https://brat.nlplab.org/">Brat</a> and <a href="https://prodi.gy/">Prodigy</a> but none of them provides the base of a scalable annotation process in a timely manner.</p><p><a href="https://twitter.com/des0ps"><strong>A very talented engineer</strong></a><strong> and </strong><a href="https://twitter.com/SophieGva"><strong>I, the product designer</strong></a><strong>, took the initiative to design an internal annotation platform, which will allow :</strong></p><ul><li><strong>text annotation,</strong></li><li><strong>curation,</strong></li><li><strong>correction and</strong></li><li><strong>review of the data.</strong></li></ul><blockquote>The final goal was to design a suite of products that will allow the NLP(Natural Language Processing) team to work together in a fast and collaborative manner on the same platform. That was how Sapiens was born.</blockquote><h3>The solution</h3><p>A multi-faceted platform for data annotation.</p><figure><img alt="" src="https://cdn-images-1.medium.com/max/1024/1*weUMKVkEADzqh9_GNd3zQw.png" /><figcaption>Sapiens blueprint</figcaption></figure><h3><strong>1. Data labeling</strong></h3><p>Data labeling required that we first craft the text annotation tool. Considering the entirety of the project, we sought a scalable solution and started defining <a href="https://projects.invisionapp.com/share/F5UEQPOCSJA#/screens"><strong>Sapiens Design System</strong></a>.</p><h4>Establishing a unified design language was a fundamental milestone</h4><figure><img alt="Adele logo" src="https://cdn-images-1.medium.com/max/1024/1*BgIbn-eYuqa2YukYKrAYUQ.png" /><figcaption><a href="https://adele.uxpin.com/">Adele by UXPin</a></figcaption></figure><p>We started our design system research at <a href="https://adele.uxpin.com/">Adele - the repository of publicly available design systems and pattern libraries.</a> After careful exploration, we studied two specific systems, which were abundant in useful components:</p><ul><li><a href="https://material.io/design/">Material Design</a> and</li><li><a href="https://ant.design/">Ant Design System</a></li></ul><p>As the product designer, I laid the grounds for <a href="https://projects.invisionapp.com/share/F5UEQPOCSJA#/screens/388985912"><strong>Sapiens Design System</strong></a><strong>.</strong></p><p>I began by following the <a href="https://atomicdesign.bradfrost.com/chapter-2/">Atomic Design Methodology</a>, focusing on the necessities first: grid and layout, modular typography scale, colors, buttons, molecules, organisms and pages. The aim was to craft templated flows, which can be used across the platform.</p><figure><img alt="" src="https://cdn-images-1.medium.com/max/1024/1*m0bhWUgnq5vKRfBC3CU-3A.png" /></figure><p>The head engineer was an essential contributor to <a href="https://sapiens-design.keatext.ai/?path=/story/general-colors--brand-colors">the Design System</a>. He kept it up to date and made sure it was consistently applied in the annotation tool. He also introduced one more point of reference for both of us <a href="https://tailwindcss.com/">Tailwind utility-first CSS framework for the frontend.</a></p><p>Frontend wise we relied on ReasonML with Tailwind and backend wiseon Elixir with a GraphQL API.</p><p>The design system is a work in progress and we continue to grow it in InVision and Storybook.</p><figure><img alt="" src="https://cdn-images-1.medium.com/max/1024/1*Fzms7BKxv93po7L4Oj28Mw.png" /><figcaption>A table component from Sapiens Design System.</figcaption></figure><p>Once we agreed on the design system foundations, we were able to move forward with the implementation process of the tool at a rapid pace. We were able to validate our assumptions fast thanks to the fact that the target users of the tool(the NLP team) sat next to us in the same office.</p><blockquote>Sapiens empower users to annotate large volumes of data simultaneously and in collaboration.</blockquote><p>The advantage of the tool is that multiple annotators can work on one dataset at the same time. Once each of them complete their batch of annotations, an automatic consensus is performed on all of the data. The annotations with the high enough agreement score are considered ready and the rest are available for the curators and correctors to work on them.</p><figure><img alt="" src="https://cdn-images-1.medium.com/max/1024/1*jctbymnbrSCOmXU499VC7Q.png" /><figcaption>Annotating a sentence in Sapiens</figcaption></figure><p>Due to budget and time constraints, we had to be very careful with the tasks, we chose to implement. We sought to build the essential user flows and validate them internally before we released them for external annotators in the platform. When the annotation tool was in a stable shape to support multiple users and to allow them to easily complete their tasks, we proceeded with the Data validation; hence, the creation of the Curation and Correction tools.</p><h3><strong>2. Data validation</strong></h3><p>When annotations for a given project were complete, they were classified based on the Annotator’s agreement. The odd ones, the cases when multiple people annotated the same sentence differently, had to be reviewed by a corrector. The corrector was the final decider, who settled the annotation agreement.</p><p>Eventually, the Golden Annotations, which had 100% agreement between the annotators, curators and correctors were the ones used to train the machine learning model. A curation task involved validating the annotations of other collaborators in a project.</p><figure><img alt="" src="https://cdn-images-1.medium.com/max/1024/1*ycbuaFjK_QoCseHm2ZYeNA.png" /><figcaption>Curating a sentence in Sapiens</figcaption></figure><h3>Outcome</h3><p>The numbers spoke for themselves. With the same number of annotators, the quality of the data remained high, and the number of annotations doubled.</p><figure><img alt="" src="https://cdn-images-1.medium.com/max/1024/1*WJxBLCsCptR8Csa1CP1o1w.png" /><figcaption>Annotations comparison</figcaption></figure><figure><img alt="" src="https://cdn-images-1.medium.com/max/1024/1*3AaXUC5uVg7_-FjqgSb0_A.png" /><figcaption>Annotations timeline in Sapiens</figcaption></figure><h3>🍏 Food for thought</h3><p>Although, we were happy with the initial outcome, we had to evaluate if we were going in the right direction. To measure the success of Sapiens, we took a few factors into consideration.</p><p><strong>We aimed to provide</strong></p><p>☑️ Faster annotations and curations by multiple collaborators</p><p>☑️ Easy to use platform and access to well defined guidelines</p><p>☑️ Quick and efficient development</p><p>We were especially proud of the efficient communication and good rapport we built in the team. It was a breeze to share the knowledge with new team members. Thanks to the well documented design system, the new team members were able to implement new screens in the platform in less than 3 days using only a white board sketch as a guideline.</p><figure><img alt="" src="https://cdn-images-1.medium.com/max/1024/1*Mj5YC--RFaYncvJMpIIjMw.png" /><figcaption>Task — whiteboard sketch</figcaption></figure><figure><img alt="" src="https://cdn-images-1.medium.com/max/1024/1*_YPfRVeKHzMaAzAL_VLsNg.png" /><figcaption>Task — implementation of the sketch</figcaption></figure><h3><strong>Learnings and room for improvement</strong></h3><p>Sapiens helps Natural Language Processing teams to work fast and in collaboration in order to train high quality text data. The platform is stable but imperfect. There is more to be desired but while building it we learned a lot. We managed to enable teams to train their data models easily and to find enjoyment in this painstaking process.</p><p>While doing so, we became experts in prioritizing goals, selecting the better design trade offs, exposing and including other team members to our process and most importantly, producing a tool that makes our users’ life a little bit better.</p><img src="https://medium.com/_/stat?event=post.clientViewed&referrerSource=full_rss&postId=d449dfd527a4" width="1" height="1" alt=""><hr><p><a href="https://uxdesign.cc/artificial-intelligence-and-design-d449dfd527a4">Machine learning, data annotation and text analytics for designers</a> was originally published in <a href="https://uxdesign.cc/">UX Collective</a> on Medium, where people are continuing the conversation by highlighting and responding to this story.</p>]]></content:encoded>
        </item>
    </channel>
</rss>