<?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 ENGIE Digital on Medium]]></title>
        <description><![CDATA[Stories by ENGIE Digital on Medium]]></description>
        <link>https://medium.com/@engiedigital?source=rss-f6d35b915fa5------2</link>
        <image>
            <url>https://cdn-images-1.medium.com/fit/c/150/150/1*tXen_ScnBZGluF0V5n1GFw.jpeg</url>
            <title>Stories by ENGIE Digital on Medium</title>
            <link>https://medium.com/@engiedigital?source=rss-f6d35b915fa5------2</link>
        </image>
        <generator>Medium</generator>
        <lastBuildDate>Wed, 18 Nov 2026 01:48:32 GMT</lastBuildDate>
        <atom:link href="https://medium.com/@engiedigital/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 deploy disruptive digital O&M solutions for Energy & Utilities companies?]]></title>
            <link>https://engiedigital.medium.com/how-to-deploy-disruptive-digital-o-m-solutions-for-energy-utilities-companies-751d9775ef1?source=rss-f6d35b915fa5------2</link>
            <guid isPermaLink="false">https://medium.com/p/751d9775ef1</guid>
            <dc:creator><![CDATA[ENGIE Digital]]></dc:creator>
            <pubDate>Thu, 11 May 2023 14:54:49 GMT</pubDate>
            <atom:updated>2023-05-11T14:54:49.468Z</atom:updated>
            <content:encoded><![CDATA[<h3><strong>How to deploy disruptive digital O&amp;M solutions for Energy &amp; Utilities companies? It’s all about trust.</strong></h3><p>By <a href="https://www.linkedin.com/in/louismariedanet/">Louis-Marie Danet</a></p><p>As Head of Customer Success for Smart O&amp;M — ENGIE Group’s new, standard, digital O&amp;M (Operations &amp; Maintenance) solution — I oversee all deployment projects to maximize value for ENGIE’s activities. My role is to ensure that Smart O&amp;M delivers concrete, measurable business benefits. This implies successfully guiding customers through the entire implementation journey, from initial discussions with key stakeholders to mass usage by O&amp;M teams in the field. After four years, Smart O&amp;M is now being extensively rolled out across regions like the Middle East and Southeast Asia, as well as globally for specific installations such as EV charge stations. I’d like to share a few practical insights on how to deploy in-house disruptive digital solutions for O&amp;M and show why success mostly depends on creating and maintaining trust over time.</p><h3><strong>ENGIE Digital is an internal software editor — why is an in-house Customer Success team necessary?</strong></h3><p>Digital transformation is an obvious mandatory step for large long-standing companies, and most of them face the same challenges in the transformation process, especially in the <a href="https://www.engie.com/en/activities/energy-solutions">Energy &amp; Utilities industry</a>.</p><p>Firstly, digital solutions require significant efforts in setting them up correctly: good quality data needs to be provided, working processes standardized; and use cases specified, etc. Secondly, robust change management must be applied to smoothly adjust a company’s existing operating model. Customer Success is needed to provide continuous support to adapt, adopt, and actually leverage technologies to unlock business value.</p><p>In addition, there are common preconceptions which will inevitably pop up, and will need to be overcome to successfully deploy a digital transformation:</p><ul><li><em>“Our current way of working is good enough, we’ve always managed to reach our targets like this”</em><br>→ make stakeholders aware that improving does not mean that they should be ashamed of the previous situation, but that they will be able to address limitations, which are often well-known internally.</li><li><em>“New technology is plug-and-play and will quickly solve all our current issues”</em><br> → manage expectations by providing ahead a clear vision of the path to follow and the conditions required to benefit from the software implementation.</li></ul><p>The Customer Success team is geared towards initiating trusting relationships with managers involved in the transformation and make it last by focusing on the targeted business benefits.</p><h3><strong>How to implement a new digital O&amp;M solution: be ready for a 4-step journey</strong></h3><p>There are several successive steps in projects for adopting new digital solutions. What makes this journey within the Energy &amp; Utilities industry quite specific is the pressure on O&amp;M costs. Making these activities efficient and profitable requires sharing resources (technical workforce, expertise, tools, machines, devices, and software, etc.) across assets and installations. And the business case for deploying a digital solution, <strong><em>especially in-house ones</em></strong><em>,</em> is only positive at scale. It is rare that the ROI of an implementation on a limited scope of activities will finance the full investment needed (solution set-up, administrator upskilling, user training, and change management material).</p><p>So, you don’t just need to launch the project, you also need to make sure you will steer it until the deployment at scale stage.</p><p>For the Customer Success team, the starting point is showing company entities that:</p><ul><li>The overall strategy and related guidelines make the cost of <em>not</em> changing the current way of working higher than initiating an O&amp;M transformation.</li><li>The digital team is reliable and trustworthy, it understands operational constraints and is able to support the design of a desirable target way of working.</li></ul><figure><img alt="Schematic representation of the 4-step journey" src="https://cdn-images-1.medium.com/max/811/1*7iFU0yiCfUvZcWgWnqZQcg.jpeg" /><figcaption>Diagram of the 4-step journey</figcaption></figure><h4><strong>Challenges and key success factors for each step</strong></h4><p><strong>1. Raise interest and convince business and operations teams to digitize activities with your solution.</strong></p><p>It is not easy to find the right sponsors within an organization: top management is often far from understanding O&amp;M teams’ daily challenges. In the field, managers have very limited time to discuss structural change in the ways of working and have restricted influence on general strategy; and cross-business functions may not be empowered to push change effectively. And, in large companies, you must also face general inertia.</p><p>Promoting an internal solution involves additional difficulties, especially at the beginning because you start from scratch, and you get compared to competitors, which can show return of experience in other organizations. As a matter of fact, in the digital field, market solutions always have a better image than internal ones because their marketing material is fancier, and they get less challenged by the target users.</p><p>To overcome these hurdles, you should:</p><ul><li>Identify key sponsors among the managers whose position requires them to challenge the <em>status quo</em> and convince them that your solution will help adjust to new market trends.</li><li>Provide them with a clear value proposition and the relevant angles for launching a project. Basically, suggest the right use cases for the right scope of activities to simplify decision-making processes.</li></ul><p>Obviously, drumming up enough backing to get a green light mostly relies on your capacity to be seen as trustworthy: show contextual business expertise and prove that the solution is appropriate.</p><p>With <a href="https://digital.engie.com/en/solutions/smart-om">Smart O&amp;M</a>, we managed to generate an initial flicker of interest thanks to strong presentation skills, but it took us two years of experience to finalize a first simple, comprehensive and convincing pitch of the solution’s business value.</p><p><strong>2. Don’t lose track of the project at the specification and configuration stage.</strong></p><p>Once the project is launched you can start defining how the solution must be used, and then configure it accordingly. This is the part where you will have to support your customer when they hit their first wall and realize that digital technology is not magic after all, and that significant work is needed to standardize processes and collect data.</p><p>We call this phase the “Valley of tears”! The gap between the initial target and what a company can actually adopt, based on process maturity, team flexibility and a legacy ecosystem that needs to be dealt with, is generally quite big in Energy &amp; Utilities.</p><p>This will often lead to revising the project’s goal from aiming for major improvements to making the solution work with an acceptable set-up, despite relinquishing some initial ambitions. For example, you will see customers deciding to postpone feature rollouts because the effort for the teams to prepare or use them is too high.</p><p>Here are two success factors to help ease this unavoidable pain:</p><ul><li>A strong, dedicated project team combining local and solution resources with complementary skills and knowledge (technical expertise, field experience, and O&amp;M know-how) and an appropriate governance body to sustain pace and strategic vision.</li><li>A knowledge management process to capitalize on lessons learnt from each project and collect best practices (including standard configuration examples) to make the next implementation more efficient and reduce total lead time before reaching the go-live phase.</li></ul><p>During this phase, trust is greatly at risk and even more so for internal software as managers will be tempted to blame a solution which is promoted by the Group rather than an external one, which they would have chosen themselves. This is where most projects stop or become irremediably damaged. We unfortunately experienced this with a few entities and managed to learn and improve from these failures. Only solid transformation project methodology, and experience, will remove inevitable doubts.</p><p><strong>3. Closely monitor adoption post go-live to boost added value and business transformation.</strong></p><p>A common mistake is thinking that the solution’s validation and go-live is the end of the journey… Handing it over to end users is an important project milestone but this is just the beginning of the actual business transformation. After a demanding implementation phase, if only minimum efforts are made to ensure that the solution is well adopted by Operations, this will likely lead to failure. Beyond functionality training programs, each category of teams involved in O&amp;M activities should be guided and supported in getting as much value as possible from the new technology in real life. Limited capacity to handle feedback, either through configuration adjustments or relevant usage suggestions, will end up with suboptimal optimization leverage.</p><p>Two top support tips:</p><ul><li>Track solution adoption with dedicated activity KPIs to detect potential solution misuse.</li><li>Use lean management methodologies, mobilize key users in short feedback loops to understand challenges, and suggest adjustments to improve and measure successful output.</li></ul><p>This step is key to fostering trust in the value of digital change within the company. Because it is challenging to transform ways of working in Energy &amp; Utilities, the project adoption team needs to be able to talk to each type of stakeholder, they need to understand and adjust to each specific point of view. So, as surprising and expensive as it may initially seem, having a skilled Customer Success team to support rolling out an internal solution is a rational and legitimate investment. This is the reason behind my role within Smart O&amp;M.</p><p><strong>4. Promote operational successes, while rolling out the solution at scale, to activate a virtuous circle.</strong></p><p>The last step of the journey is obviously the solution’s general rollout. It is challenging to move from punctual implementation on a limited scope to deployment at scale. Firstly, you need to be able to handle additional use cases related to non-standard processes and contexts, as each entity claims to have specific needs (contract, customer, installation, and local organization, etc.). Secondly, you will have to perform migrations when existing solutions are already being used in the field. Finally, you must industrialize configurations and training activities, which require significant investment in a transformation program and associated change management.</p><p>To address these challenges, you should focus on two topics:</p><ul><li>Introduce professional digital transformation program methodologies and resources to steer and perform rollout activities at scale and foster standard or existing use cases, as much as possible (refer for example to ERP deployment project: fit/gaps analysis matrix, migration tools, and standard training material, etc.)</li><li>Create powerful Marketing/Communications content to promote business value provided by the change, based on initial experiences and users feedback, to demonstrate the solution’s accessibility and global profitability to management and all O&amp;M teams.</li></ul><p>This is the most important phase for Customer Success as it will determine whether you can reach the critical mass of users required to make the solution cost effective. The more users you gain, the easier it will be to convince and guide additional teams through the journey and the stronger the business case will be. Smart O&amp;M is now entering this phase within <a href="https://www.engie.com/en">ENGIE</a>.</p><h3><strong>Customer Success: how can you tell that the journey is over?</strong></h3><p>In my opinion, a digital transformation journey is a never-ending story. Nevertheless, the Customer Success team’s job is done when a local entity fully appropriates the new digital solutions. This means that the local teams can imagine by themselves new use cases to optimize operations, cover new types of activities, and even use the solution to create new business opportunities. At this stage, you return to your role as software editor and your value will be mostly to follow up on business innovation to facilitate experience sharing between customer entities and cross fertilization. Basically, when we reach this stage, Smart O&amp;M will generate trust by itself and won’t need Customer Success to back it up.</p><p>Find out more about how Smart O&amp;M supports its clients: <a href="https://digital.engie.com/en/smart-om-manages-your-technical-installations-real-time-make-maintenance-efficient-and-transparent">https://digital.engie.com/en/smart-om-manages-your-technical-installations-real-time-make-maintenance-efficient-and-transparent</a></p><img src="https://medium.com/_/stat?event=post.clientViewed&referrerSource=full_rss&postId=751d9775ef1" width="1" height="1" alt="">]]></content:encoded>
        </item>
        <item>
            <title><![CDATA[The Experience Index puts WattsOn for Giants to the test]]></title>
            <link>https://engiedigital.medium.com/the-experience-index-puts-wattson-for-giants-to-the-test-90c934778d94?source=rss-f6d35b915fa5------2</link>
            <guid isPermaLink="false">https://medium.com/p/90c934778d94</guid>
            <category><![CDATA[design]]></category>
            <category><![CDATA[case-study]]></category>
            <category><![CDATA[ux-design]]></category>
            <category><![CDATA[ux]]></category>
            <dc:creator><![CDATA[ENGIE Digital]]></dc:creator>
            <pubDate>Fri, 28 Apr 2023 07:10:10 GMT</pubDate>
            <atom:updated>2023-04-28T07:10:10.055Z</atom:updated>
            <content:encoded><![CDATA[<p>By <a href="https://www.linkedin.com/in/elenaivankina/">Elena Ivankina</a></p><p>The team at WattsOn for Giants used the Experience Index to optimize its customer-centric approach, redefine the designer’s role, and better leverage its expertise to increase the product’s performance and user uptake. The Experience Index is a self-evaluation survey developed by ENGIE Digital’s Design Studio to gauge the level of user experience for the ENGIE Group’s digital products and services. Feedback.</p><h3><strong>Creating value through user experience</strong></h3><p>Why spend time and money on a digital solution if it does not resolve user issues or generate value?</p><p>These types of pitfalls can be avoided with user experience (UX). By building and pressure-testing a digital solution through this lens, you can ensure that it is useful, effective and viable for the company. This scope also includes product and service design. More than anything, the approach is a lever for addressing the business performance of digital projects. The Experience Index is an excellent way to make these objectives top-of-mind.</p><h3><strong>The Experience Index is a ‘made in ENGIE’ self-evaluation tool</strong></h3><p>Developed by the Design Studio at <a href="https://digital.engie.com/en">ENGIE Digital</a>, the <a href="https://www.engie.design/experience-index/">Experience Index</a> is a self-evaluation survey that measures the maturity level of user experience for <a href="https://www.engie.com/en">ENGIE</a>’s digital projects. It also suggests ways to improve product performance.</p><p>After answering a list of questions, you are given a score indicating your current stage in the design process. There are four levels of maturity: <strong>Emerging</strong>, <strong>Tactical</strong>, <strong>Mastered</strong> and <strong>Vital</strong>.</p><p>Based on your score, the Experience Index customizes recommendations for you on four key components: <strong>user experience</strong>, the <strong>team</strong>, <strong>methodology</strong> and <strong>leadership</strong>.</p><p><em>Learn more about the Experience Index </em><a href="https://engiedigital.medium.com/first-experience-index-annual-report-the-key-in-transforming-our-user-experience-culture-e95bda5c6de3"><em>here</em></a><em>.</em></p><p><em>Get to know the Design Studio </em><a href="https://medium.com/@engiedigital/engies-design-studio-is-all-about-collaboration-888bd2af1598"><em>here</em></a><em>.</em></p><h3><strong>Case study: WattsOn for Giants</strong></h3><h4><strong>1. About WattsOn</strong></h4><p><a href="https://wattson.engie.com/login">WattsOn</a> is the customer-facing digital platform that Global Energy Management (GEM) uses to manage and monitor its end-to-end power supply accounts, from advising, risk management and forecasts to transactions, consumption, and billing.</p><p>WattsOn has two main objectives: manage accounts and personalize the user experience.</p><p>The team at WattsOn for Giants handles the platform’s key customer accounts. It excels at always being mindful of what customers need, interacting with them to determine their pain points, and providing solutions.</p><p>It turned to the Experience Index to be able to prioritize and select the upgrades that would eliminate these pain points.</p><h4><strong>2. The Experience Index x WattsOn for Giants: working smarter not harder</strong></h4><p>Around half of the WattsOn for Giants team that took the survey obtained a maturity level of Tactical. It is a pretty typical score for a product from an ecosystem that is not digital-native. In this particular case, UX was not actually a key component of the platform’s decisions and priorities. It is more seen as a function that can be turned on now and then instead of an actual growth driver.</p><p>But the WattsOn for Giants team is very close to its users and knows them inside and out. They interface with them all the time and gather feedback through a number of channels like Jira, Confluence and live chats. The product team even holds sessions called <em>Living My Life</em> where they observe users in their environment.</p><p><strong>The UX paradox</strong></p><ul><li>For WattsOn for Giants, the vast majority of the team said they track large amounts of qualitative and quantitative data to measure the experience users receive, and that they take customer feedback very seriously.</li></ul><p>BUT</p><ul><li>Half the team acknowledged that they go with their gut feeling when it comes to this experience and virtually the same proportion did not know how the feedback was being used.</li></ul><p>So how did they get such a relatively low maturity score? And how can they improve it? It is a paradox that usually indicates the team is spending too many resources on priorities for the tech, functional and business aspects, and not enough on the shared product vision.</p><p>In this case, it usually turned out that few team members had a full grasp of the experience or the value they give customers. This results in dissociating the indicators from the product, which renders them meaningless.</p><p>Then user satisfaction statistics are often used as basic reporting data and vanity metrics that are basically pointless. <strong>So, all the work done measuring and leveraging user behavior turned out not to deliver on expectations.</strong></p><p>A fresh approach for the WattsOn for Giants team could be to improve how they manage critical stages of the user pathway, associate them with different user types, and share them with the team, who can make it into an actual routine decision-making tool.</p><h4><strong>3. Making indicators meaningful</strong></h4><p>It may seem like an ambitious plan, but kickstarting it could only be a matter of a designer investing a few days, and mere hours for the team and the users.</p><p>The five-step method:</p><ol><li>Define (or validate for WattsOn for Giants) the user pathway with the product team</li><li>Approve and supplement the pathway with a few users</li><li>Map out the pathway’s most critical stages, which are usually when the product is interacting with other software</li><li>Determine what you want to measure and why: this is the step where product vision and objectives count the most</li><li>Choose and configure a tracking tool or figure out your own way of tracking these KPIs.</li></ol><figure><img alt="" src="https://cdn-images-1.medium.com/max/1024/1*A2l8q6BwJS7RWD4gVLjU4Q.jpeg" /></figure><h3><strong>How can we ensure the process is a success?</strong></h3><p>Just because the process starts does not mean all the issues are instantly resolved. It is better to begin with a very narrow scope, and then expand it to solve the problems at hand.</p><h4><strong>Step 1</strong></h4><p>Choose the user pathway where you control the most elements:</p><ul><li>Available and engaged users and product team</li><li>A pathway where you decide the objectives and issues</li><li>An experienced designer who is part of the process and has in-depth knowledge of your product’s issues and environment so they can help you map the pathway and define the KPIs</li><li>A management team involved in the project that can tell you which business interests to focus on when deciding on relevant indicators</li><li>All the stakeholders who are part of the process and know how it is progressing</li></ul><h4><strong>Step 2</strong></h4><p>Centralize the feedback and other user-related data, and contextualize it by referencing the pathway and pre-set KPIs.</p><p>Assign someone the role and responsibility of managing that feedback. They will be tasked with sharing, analyzing, and leveraging it backed by the entire team’s expertise.</p><h3><strong>Bonus tips!</strong></h3><ul><li>Loop in the UX Designer at every stage with the project leads.</li><li>Invite the developers to participate in meetings with users or at least share the user feedback with them.</li><li>Make sure all the product team members take turns testing out the process. For example, ask one or two different team members each month to track the pathways or scenarios of some well-defined profiles and give feedback like real users. This will get everyone familiarized with the pathways as well as provide bug reports and feedback.</li></ul><blockquote><em>“UX is a vital aid in generating fast product acceptance and it provides countless opportunities like a competitive edge, customer engagement and loyalty, employee performance and convenience, plus it optimizes organizations and processes. The Experience Index is an excellent way to make these top priorities.”</em></blockquote><blockquote>Elena Ivankina, Lead Design Strategist at ENGIE Digital</blockquote><figure><img alt="" src="https://cdn-images-1.medium.com/max/1024/1*ZifEErGOJ-2V48Y0CyKalg.jpeg" /></figure><img src="https://medium.com/_/stat?event=post.clientViewed&referrerSource=full_rss&postId=90c934778d94" width="1" height="1" alt="">]]></content:encoded>
        </item>
        <item>
            <title><![CDATA[Designing a disruptive digital solution for the Energy & Utilities sector]]></title>
            <link>https://engiedigital.medium.com/designing-a-disruptive-digital-solution-for-the-energy-utilities-sector-595bf304720?source=rss-f6d35b915fa5------2</link>
            <guid isPermaLink="false">https://medium.com/p/595bf304720</guid>
            <category><![CDATA[maintenance]]></category>
            <category><![CDATA[operations]]></category>
            <category><![CDATA[software-development]]></category>
            <category><![CDATA[digital]]></category>
            <dc:creator><![CDATA[ENGIE Digital]]></dc:creator>
            <pubDate>Fri, 09 Dec 2022 09:28:35 GMT</pubDate>
            <atom:updated>2022-12-12T13:12:01.017Z</atom:updated>
            <content:encoded><![CDATA[<p>By <a href="https://www.linkedin.com/in/roland-schneiders-021a782/">Roland Schneiders</a></p><p><em>I’m Roland Schneiders, Head of Services of Smart O&amp;M — ENGIE group’s operations and maintenance solution. I oversee Smart O&amp;M’s feature development to offer our business activities maximum value and provide ENGIE’s maintenance teams with optimum user experience. Now that our solution is widely considered to be valuable, I want to share some key insights I have gathered over the last decades working hands-on in Digital &amp; Utilities. I hope my experience will be useful for colleagues in the sector who want to develop the most relevant digital solutions with added business value and that you will maybe learn from the challenges I came across.</em></p><p>I studied mechanical engineering and started my career as an Automation Engineer in the Netherlands where I led major industrial projects to optimize production lines in different industries. Over time, as I worked for different entities within the <a href="https://www.engie.com/en">ENGIE</a> group, I became familiar with maintenance activities and service contracts (including SLAs, customer relationship, preventive vs corrective actions, trade-offs, and optimization). I initially focused on OT systems (Operational Technology) and then my career shifted when I had the opportunity to work on a major datacenter design, build and operations project. I discovered that the technology used to monitor and optimize IT and digital system management were far more advanced compared to those in the traditional OT world. It was then that I had my first intuition, which led me some years after, to design <a href="https://digital.engie.com/en/solutions/smart-om">Smart O&amp;M</a>. The idea was to upgrade standard operations with innovative digital capabilities: live data monitoring; energy consumption analysis; and ticketing to manage operators’ actions. The questions at the time were: how can this idea be translated into actual solutions to optimize maintenance operations performance, and would there be a business opportunity for it?</p><h3><strong>How the digital transition affected the operations and maintenance business</strong></h3><p>While I was mulling over these questions, I witnessed the considerable impact the digital revolution was having on the maintenance business. A new era was beginning: on-site operators were equipped with increasingly advanced devices and tools; and sharing procedures and controlling implementation, tracking actions (including time and material), minimizing the volume of unplanned activities were the new ways for optimizing operational performance. Maintenance providers’ benefits that came from digital must then be shared with clients. Competition between actors led prices to drop and forced companies to improve their own processes. Customer expectation for transparency was also impacted by digital technology: detailed operations reports were more frequently required to prove quality of service. This meant that maintenance companies had to design new types of commercial offerings: output-driven or performance-based. The common denominator of the latter was to sell installation/equipment availability or fixed price for a defined scope of service. In the utility sector, the standard was EPC (Energy Performance Contracting), where actors were able to rely on their know-how to guarantee energy savings for customers and offer them best value procurement, meaning the best value for an acceptable price.</p><h3><strong>Digital enablers</strong></h3><p>Digital technologies were very promising enablers in extending this strategy: big data overwhelmed innovation and prospective departments in the utility industry. In a sector where equipment and technicians were generating massive amounts of operations-related data, it was obvious that there would be numerous use cases to create business value: condition-based maintenance; predictive maintenance; on-site augmented reality; and digital twins; etc. And to concretely support these use cases, we could rely on maturing technologies: powerful cheap mobile devices; light IOT sensors; and cloud computing. So, the main challenge was to combine technical capabilities with operational processes to support the new business offerings required to manage operational and financial risks and optimize asset performance whilst containing maintenance costs. In short, we had to design solutions to leverage real-time 360° information, which provided detailed information and guidance for users, and which encouraged transversal communication and collaboration. The main purpose of my digital journey within ENGIE has been to address the following challenge: provide users with an easy, attractive solution that supports them in implementing the overall Group business strategy.</p><h3><strong>How to develop, and run, a digital solution to disrupt maintenance activities</strong></h3><p>I have been busy with this challenge for the last 15 years! And I have developed successive versions of digital solutions to optimize maintenance operations for different types of installations. You could say it was like being agile before “Agile” came along. There are a few take-aways to share from this experience, which are not always intuitive or easy to say openly on LinkedIn!</p><figure><img alt="Infographics of the 4 key takeaways" src="https://cdn-images-1.medium.com/max/1024/1*1xu05drqnrgqqc1Js5cnmg.jpeg" /></figure><h3><strong>A complex stakeholder chain</strong></h3><blockquote>“Any digital product journey requires identifying users and understanding their business activities and expectations.”</blockquote><p>In the utility sector, the business need is often straightforward: everyone wants an installation or equipment to be functioning, ideally in an efficient way. But if we look more closely at the maintenance ecosystem, you can see that the stakeholder chain is quite complex:</p><ul><li>Technicians want to work as quickly as possible — they like tools such as GPSs, and weather forecasts, etc.</li><li>Operations managers want to steer technicians’ activities — they like ticket overviews and installation status</li><li>Contract managers want to maximize profitability — they like automated, insightful dashboards in line with the SLAs defined with customers</li><li>Client asset owners are looking for service provider performance — they like contractual reports, better transparency, and real-time contractor performance overviews</li><li>Client technical managers want to control installation status — they like equipment condition and performance status reports, and regulatory compliance</li><li>Installation users want to benefit from the installation — they like tools to report complaints and request services</li></ul><p>Given this complexity, you must start by building a vision of the digital target. You will have to listen to business managers to understand their company strategy, to operational users to picture their day-to-day constraints, and to sector experts to anticipate context evolutions, and then come up with your own plan of action.</p><p>It is only after this point that you can work on designing a solution, which will support the implementation of required maintenance use cases. And during this process, if you get the feeling that you are just replicating an existing set of business processes, with no other change than a better interface, then stop and review your vision.</p><h3><strong>4 best practices for building an efficient solution</strong></h3><p>The next question is how to build a good solution? Beyond choosing a methodological agile framework, which is always required to structure organization and work methods, I would like to point out a few vital must-dos.</p><h4><strong>1. Feature production</strong></h4><p>When designing features, you must make generic technical capabilities easy to combine and configure, and easily adjustable to support various functional implementation methods. In utility maintenance services, activities, installations, and contracts are often quite similar but their combination is always very specific.</p><h4><strong>2. Team and product management</strong></h4><p>You need to both convince and trust the team to carry out the vision. But it is quite hard to communicate as there are two aspects to consider:</p><ul><li>Use case — Feature implementation — Technical capability configuration is the world of business</li><li>Technical design — UI components — Software engineering methods is the world of products</li></ul><p>Based on the time available and each team member’s specific expertise (business analysts, product owners, UX designers, software architects, scrum masters, developers, and QAs, etc.) not everyone will understand the full picture. So, it makes sense to share the vision, but you must rely on a few trustworthy relays to manage delivery continuity. “A strong network of key people is one of the main factors for success.”</p><h4><strong>3. Anticipating the solution’s implementation</strong></h4><p>You need to provide the user with a solution, which can be open, modular, and flexible, to simplify integration to a specific local context. Don’t forget that your solution will have to:</p><ul><li>Interact with an existing IT eco-system, as green fields are quite rare. You must be ready to deploy useful features within the full constraints of legacy landscapes</li><li>Face actual real-life obstacles. You must apply the 80–20 rule to keep resources for later adjustments rather than spending too many resources trying to prevent all possible edge cases. Make pragmatic and fair trade-offs between time-to-market, user adoption, and business value, before going live and then complete the solution based on priorities</li></ul><h4><strong>4. Mindset and behavior towards stakeholders</strong></h4><p>You will often be torn between pulling opinions and pushing convictions, but the most important thing is how to choose when to do what:</p><ul><li>During the design research phase, don’t listen to user feature suggestions before understanding the end-to-end business process</li><li>During the build phase, stay firm in making decisions aligned with the vision, no decision is worse or more costly than reversing a decision after 3 months</li><li>Pick intermediate presentation steps wisely to be sure to make a good impression, as a negative buzz and distrust can kill your product. Prepare a specific demo speech for the management, business development, and operations teams to show that every version of the solution supports workable and valuable use cases</li><li>Don’t die with your ideas when testing the solution on the field: you need to firmly believe your vision, but you will always have to adjust it to actual stakeholder feedback. There are many ways to shape the features, be ready to change and adapt so that they can be delivered</li><li>When rolling out, be humble with operational users and assertive with strategy managers, this is the best way to promote your solution</li><li>After the roll out, you must remember that product development is a never-ending story: to stay ahead of the game, you must look for more business value rather than technological innovation (this can guide you when deciding whether you must extend horizontal coverage and deepen verticals by taking market standards into account)</li></ul><p>I hope that you can build an attractive, successful solution by following these best practices. Now, let’s see how to ensure that a digital maintenance business tool is a success.</p><h3><strong>Ensure maximum value for utility organizations</strong></h3><p>There is no simple way to know whether a solution is good! You cannot use standard B2C KPIs to assess a digital solution in the maintenance services. Basically, time spent on the solution is time not spent performing technical tasks on equipment, which is the core activity of our users. So, if users seem very involved in the application, it can mean either that it is very insightful or that it is difficult to use and doesn’t suit their daily needs.</p><p>You can easily check if your solution is good by directly asking users if they are happy with it. However, this approach also has its limits. If you choose to launch a quantitative survey, there will be biases, like:</p><ul><li>Misleading questions (Are you happier with this solution rather than another/no solution? Could it be improved? etc.)</li><li>Most respondents are often people with complaints</li><li>People often try to provide the answer they think the management team is expecting</li></ul><p>It makes more sense to opt for a qualitative methodology. The main limit is that you cannot interview all users, so how will you select a representative sample? And how will you assess your solution’s global impact?</p><p>Other indirect indicators help to evaluate feature quality: you can compare overall team and contract performance where your solution is implemented with other similar business contexts. This must reflect whether your solution is contributing to value creation in distinct organizational setups. However, the reasons behind better performance can be other than a digital solution. Plus, in the utilities industry, performance is not always easy to evaluate depending on whether you focus on operational productivity, installation availability, energy efficiency, or financial profitability.</p><figure><img alt="Infographics on KPI to monitor jointly" src="https://cdn-images-1.medium.com/max/1024/1*W6FZz1f8PEEhmYBEpuafxA.jpeg" /></figure><p>It appears that assessing a digital solution’s value requires combining several methodologies. But it also seems that the quality of the solution cannot be assessed by itself, as what is assessed at the end of the day is how users leverage it to perform their own business activities. A solution’s value highly depends on how it is adopted within processes. In the utilities sector, and more precisely in maintenance, where traditionally the experience of technology is quite limited, you could even say that a just-good-enough solution would generate more business value than an optimal one if it is largely used and effectively supports users’ processes.</p><p>As a result, a robust vision is required to build a good solution but also to promote change throughout the stakeholder chain. Besides software development, you need to make a considerable effort to drive adoption:</p><ul><li>Communicate about value for each user profile to make it attractive</li><li>Provide training material and detailed examples on how to implement use cases</li><li>Organize specific teams to spend time on change management to help users get started</li></ul><p>Doing this downwards of software development is a way to efficiently collect useful team feedback to fine-tune the solution and make it even more valuable for users’ business activities, rather than spending months with upfront in-depth research, design thinking, and edge case analysis. This is one of the key specific aspects of our approach with Smart O&amp;M, which makes it successful in delivering real value to ENGIE O&amp;M services.</p><h3><strong>Knowing when your solution is ready</strong></h3><p>I hope my experience will help you when you’ll be working on your next digital applications. I’d like to conclude with a last thought.</p><blockquote>“Even if product development is a never-ending story, at some point you need to stop growing the functional scope of a solution to make it run sustainably.”</blockquote><p>How do you know that a solution is ready? Here are a few tips which you can complete with your own experience!</p><ol><li>When most user feedback is more focused on usability rather than on new features</li><li>When your benchmarks/market studies show that there is no obvious mature feature that competitors include in their own solutions</li><li>When the solution is delivering value in line with the using entities’ business strategy according to its business processes and maturity</li></ol><p>When you reach this point, you need to finalize API and data-sharing features. For maintenance activities in the energy/utilities sector, the future is resolutely being designed by AI: predictive maintenance; schedule optimization; and automatic remote control based on installation usage to automatically optimize energy performance.</p><h3><strong>Next steps…</strong></h3><p>Once your solution is successfully up and running, it must be ready to cope with integration with the next digital game changers when they shift from the advanced innovation stage to market standard.</p><p>And, be prepared for this, because this is more than likely to happen, as going for asset-based and performance as-a-service offerings seems to be an inexorable trend to preserve profitability of large-scale service providers.</p><img src="https://medium.com/_/stat?event=post.clientViewed&referrerSource=full_rss&postId=595bf304720" width="1" height="1" alt="">]]></content:encoded>
        </item>
        <item>
            <title><![CDATA[ENGIE’s Design Studio is all about collaboration]]></title>
            <link>https://engiedigital.medium.com/engies-design-studio-is-all-about-collaboration-888bd2af1598?source=rss-f6d35b915fa5------2</link>
            <guid isPermaLink="false">https://medium.com/p/888bd2af1598</guid>
            <category><![CDATA[designer]]></category>
            <category><![CDATA[design]]></category>
            <category><![CDATA[ux]]></category>
            <category><![CDATA[collaboration]]></category>
            <dc:creator><![CDATA[ENGIE Digital]]></dc:creator>
            <pubDate>Fri, 30 Sep 2022 10:08:46 GMT</pubDate>
            <atom:updated>2022-09-30T10:08:46.678Z</atom:updated>
            <content:encoded><![CDATA[<p>By <a href="https://www.linkedin.com/in/romain-petit-5509484/">Romain Petit</a></p><p><em>The motto at ENGIE’s Design Studio is: view design and UX through a collaborative lens internally and externally. Head of Design Romain Petit explains this innovative approach.</em></p><h3>Building a community of designers at ENGIE Digital…</h3><p><a href="https://digital.engie.com/en">ENGIE Digital</a>’s organization revolves around collaboration. The teams that work on digital platforms developed in-house rely on cross-functional teams of experts like data, tech and operations, but also mostly depend on UX and UI designers from the Design Studio. There are about 20 of these designers on staff, who are all fully integrated for the most part into product teams.</p><p>A practice community called practice Design was started in 2020 to help share experience and expertise. The concept is to pool best practices, provide a safe space where employees have time to co-create and brainstorm, and allow everyone to develop their skills and share common standards such as methodologies, design systems and DesOps.</p><p>Community time was built into the routine to bring together people with different profiles (i.e., UX, UI, front dev, etc.) to discuss cross-functional topics. Every two weeks, “Friday Studio” is held where we address one or more subjects with a variety of formats like a workshop, design crit, demo, feedback, or watch and trends. The goal is to give every designer a chance to bounce things off their counterparts and leverage shared knowledge. And because there are other sources of inspiration and best practices, the Design Studio regularly invites guest designers from other companies for a discussion or even to work on an issue they have in common. For example, they recently held a joint workshop with designers from the RATP studio (Paris’ public transit authority). The theme was “How do we measure the impact of design?” and another workshop with teams from TotalEnergies, Thales and Air Liquide to pool feedback on design systems.</p><blockquote>The practice community’s purpose is to give all the designers a chance to be an integral part of a collective, not just to foster a shared culture and knowledge base, but also and most importantly to support one another, grow together and envision future changes in our industry.</blockquote><blockquote><em>Romain Petit, Head of Design, ENGIE</em></blockquote><h3>…and instilling a design culture and standards throughout the Group</h3><p>In addition to its work at ENGIE Digital, the Design Studio is also managing the rollout of design standards it created for the entire Group on its <a href="http://www.engie.design/">engie.design</a> platform.</p><p>Most notably, the Studio<em> </em>designed and deployed <a href="https://www.engie.design/fluid-design-system/">ENGIE’s Design System called fluid</a> to standardize front end components and templates in line with <a href="https://www.engie.design/fluid-design-system/ux-principles/">UX principles</a> and ENGIE’s brand identity. In barely three years, fluid has become the standard for all the Group’s designers and developers. This tool helps them quickly design digital products and services that deliver a consistent user experience.</p><blockquote>It is an extremely unifying move to provide a design-and-develop tool for digital products that meet ENGIE’s experience and identity standards, not only because it puts the designers and developers in a space where they can share and continuously improve, but also because it optimizes the go-to market for our industry and business stakeholders as well as considerably cuts costs in terms of production time. The ROI is immediate<em>.</em></blockquote><blockquote><em>Romain Petit, Head of Design, ENGIE</em></blockquote><p>Another example of a collaborative project that Practice Design led in the Group is setting up the <a href="https://www.engie.design/experience-index/">Experience Index</a>. It’s a diagnostic tool that teams can use to assess the maturity level of ENGIE’s digital projects in terms of how user-centric they are and then recommend customized improvements. In 2021, over 120 projects across the globe were given detailed assessments. The <a href="https://www.figma.com/proto/0urh6D8hREIvmQLMHq16ir/Exp-Index-Annual-Report-2022---PROD?page-id=0:1&amp;node-id=0:1485&amp;viewport=283,48,0.04&amp;scaling=contain&amp;hide-ui=1">first Experience Index Annual Report</a> also came out. In particular, it shows that 20% more user feedback was shared among the teams that took the test. In 2022, the tool continued being deployed internally to get ENGIE’s teams further aligned on design-related standards. An open source version will soon be released for other companies.</p><figure><img alt="Infographics of the 3 good reasons to approach design cross-functionally" src="https://cdn-images-1.medium.com/max/693/1*FvvOvpl_Am-db1O-Wq4bYg.png" /></figure><h3>Collaborating externally to advocate for design</h3><p>By no means does the collaborative design approach stop at ENGIE’s door. There are many ways it can apply externally, for example at peer exchanges like the RATP workshop and the Design System feedback sessions mentioned earlier, as well as with students and younger generations.</p><p>In this capacity, ENGIE is a strategic partner of the Web School Factory, which trains future managers in digital technology with a major in design. The partnership’s key mission is to develop, accelerate and co-create innovative projects using a collaborative model with a unique method of pairing up operations teams with students. Led by the Design Studio, the partnership gives product teams an opportunity to collaborate with students throughout the year in modules ranging from a tech watch and anti-master class to a hackathon and final project based on various use cases from ENGIE Digital.</p><figure><img alt="Infographics about the benefits for both ENGIE and the Web School Factory’s students (win-win experience)" src="https://cdn-images-1.medium.com/max/693/1*Hg-9UKui_8pfcq_12vs04w.png" /></figure><p>Lastly, collaborating also means promoting or even protecting common interests! In this vein, ENGIE recently reached a new milestone by partnering with Adobe to launch the Innovative Design Board. The concept was to form a group of decision-makers working in design and innovation to share feedback and beliefs on practicing service design as well as speak as a united voice on issues they all share. The founding members come from a wide variety of companies such as Schneider Electric, EDF, BPCE, Decathlon, Renault, Pernod Ricard, L’Oréal and, of course, ENGIE. The goal is to put leaders in digital and product design in the same room so they can collectively shape the industry’s future, positioning, and internal and external recognition.</p><blockquote>The Innovative Design Board aims to share its vision of service design and feedback from sources that are very diverse, yet face similar upscaling issues. Our objective is to pool our beliefs, form a collective and speak as one for the interests of an emerging industry, both to the top of the corporate ladder internally and to all our stakeholders externally.</blockquote><blockquote><em>Romain Petit, Head of Design, ENGIE</em></blockquote><figure><img alt="Infographics on the topics addressed by the Innovative Design Board" src="https://cdn-images-1.medium.com/max/693/1*8JpKV9YHRA1gnBbP0BuWfQ.png" /></figure><p>To learn more about the Design Studio’s methodology, latest news, resources, and skills, visit <a href="https://www.engie.design/">https://www.engie.design/</a></p><img src="https://medium.com/_/stat?event=post.clientViewed&referrerSource=full_rss&postId=888bd2af1598" width="1" height="1" alt="">]]></content:encoded>
        </item>
        <item>
            <title><![CDATA[UX KPIs and user-centricity: please start with something and get results!]]></title>
            <link>https://engiedigital.medium.com/ux-kpis-and-user-centricity-please-start-with-something-and-get-results-6ee37ac1f264?source=rss-f6d35b915fa5------2</link>
            <guid isPermaLink="false">https://medium.com/p/6ee37ac1f264</guid>
            <category><![CDATA[ux-design]]></category>
            <category><![CDATA[user-centered-design]]></category>
            <category><![CDATA[ux]]></category>
            <category><![CDATA[design]]></category>
            <category><![CDATA[kpi]]></category>
            <dc:creator><![CDATA[ENGIE Digital]]></dc:creator>
            <pubDate>Wed, 20 Jul 2022 14:54:46 GMT</pubDate>
            <atom:updated>2022-07-20T14:54:46.705Z</atom:updated>
            <content:encoded><![CDATA[<p>By <a href="https://www.linkedin.com/in/elenaivankina/">Elena Ivankina</a></p><p><em>“Putting human at the center” — how many times have you heard or read this? How many times have you said it in design thinking workshops and other brainstorms? A lot. Too often because it has become a digital buzzword. What does this phrase mean in reality?</em></p><p>In the User Experience (UX) service or product design perspective, human is always at the center. Designers’ ultimate goal is to build experiences that bring them value and solve their irritations. But they are at the center through the whole process: when framing and concepting the idea, during the development and deployment phases; and after, while prioritizing and realizing new cycles.</p><p>Strangely, designers do not often have the reflex to measure the impact of their work. However, there is really no other way to ensure that the user stays at the center while decoding the other major triggers of product evolution such as the effects of technologies, the environment, etc.</p><p>For example, the Zoom application that we all know today was an answer to a user need, but also took advantage of the pandemic to really explode. Then, it almost sank due to technical (or ethical?) weaknesses. The context triggered an opportunity that they seized and ultimately succeeded in keeping their value proposition focused on needs.</p><p>The only way to get there is to find relevant landmarks, but this is tricky. Actually, it is both the lifeblood and the Achilles heel of many products. Yet, once you manage to successfully identify the right KPIs to measure and monitor user experience, these KPIs become a valuable decision-making tool that can be used at strategic levels. But it is also a complicated matter since it involves finding one’s way in an uncertain context in every respect. Here are a few tips to help you define the right metrics to be more user-centric!</p><h3><strong>Measuring user experience to stay user-centric: not always a prerequisite met</strong></h3><p>User centricity is key to delivering a good and seamless user experience.</p><blockquote>You learn more in a day talking to customers than a week of brainstorming, a month of watching competitors, or a year of market research.</blockquote><blockquote><em>Aaron Levie, Box Co-founder &amp; CEO</em></blockquote><p>While we understand globally why “a day talking to customers” is valuable, its implementation is not so easy. Yet, user-centricity is much more than simply saying that users are at the center of our actions. It is a range of practices, rules, and rituals that allow to perfectly understand user needs, irritants, and motivations, and address them through the experience provided by the product.</p><p>In-depth knowledge of your user’s expectations and needs is crucial to building the value proposition of a product, solution, or project. Someone said:</p><blockquote>It’s harder to make people want things than making things people want.</blockquote><blockquote><em>Designer’s proverb.</em></blockquote><p>What we encounter in some cases is that when implementing certain products, the user’s needs are determined at the beginning but are not really followed afterward. Even if it is a good start, it is the opposite of user-centricity. It is a bit like admitting a patient to the hospital, administering them treatment, and then stopping to monitor treatment effects.</p><p>Assuming users’ expectations without checking them throughout the product life cycle does not make any sense<em>. </em>User experience measurement starts by defining the product’s objectives: it is precisely these objectives that will guide indicator choice the rest of the time.</p><p>Unfortunately, indicators are most widely used in the B2C domain (for example, for growth, conversion, and commitment objectives), but in B2B, the stakes are not less high. For example, user engagement, security, and efficiency are decisive indicators for certain products, as is the cost of development and management of legacy.</p><p>A simple user journey audit can save hundreds of thousands of euros in development costs if the journey indicators show that some of its branches are never used, and do not need to be maintained.</p><p>In addition to identifying indicators, the challenge is recognizing the salient clues in the information to be interpreted. For example, what should I do when a news release did not generate the expected engagement, despite the positive user feedback?</p><blockquote>The scientist is not a person who gives the right answers, he is one who asks the right questions.</blockquote><blockquote><em>Claude Levi-Strauss, anthropologist</em></blockquote><figure><img alt="" src="https://cdn-images-1.medium.com/max/1024/1*FHTGdqO6V28vJ-0-YfzCjA.png" /></figure><h3><strong>How to define metrics by asking “the right questions”</strong></h3><h4><strong>What makes a KPI relevant?</strong></h4><p>What is a good KPI? This is a complicated concept, as there are many possibilities and various situations. There is no standard KPI, such as the ideal number of clicks needed to perform a task, because KPIs vary with the product’s nature and complexity. For each product you will have to find the relevant information to monitor. The risk is to be overwhelmed by the mass of information that can be monitored, which leads to bad arbitrations.</p><p>A good KPI is an indicator that can be followed over time, it should not be fixed at a specific point in time. The ability to decrypt relevant trends in a set of KPIs and interpret it properly is key to drawing relevant conclusions and make informed decisions. The main issue is to combine monitoring several types of KPIs to cover major fields of the product<em>.</em></p><p>So, defining the right metrics means finding the right combination of KPIs. There are three main types of KPIs:</p><ul><li><strong>Behavioral KPIs: what is the user’s behavior? </strong>Explore user’s behavior by following the user’s product journey: when why and how do they use the product? Which pages do they use the most, in which order, etc.? This information can be obtained by tracking technical product information, like logging, page requests, etc., but also by tracking more specific UX metrics, such as on-page eye-tracking, heatmaps, etc. This can be measured continuously while using the tool, but also with a focus on a particular feature.</li><li><strong>Appreciation KPIS: how does the user appreciate the product? </strong>Understand what users think and say about the product. This type of information can be obtained via surveys, customer feedback, user meetings, or net promoter score (NPS). It is important to collect user perceptions at crucial moments and give users the maximum opportunity to express themselves. Structuring exchanges with users rather than soliciting them in an alleviatory manner is much more efficient. Also, remember that the moment you ask your users for feedback (in their experience with your product, in their day, etc.) will necessarily have an impact on it.</li><li><strong>Business KPIs: how does the product help to achieve business goals? </strong>For these, the user experience should be adapted to business objectives, such are new user acquisition, revenue growth, cost-saving, etc.</li></ul><p>To find the best combination of KPIs, you should also ask yourself:</p><ul><li><strong>How does the product help to achieve operational goals? </strong>What value does the user get from using the product. This value can be quantified as an appreciation or based on the accomplishment of the jobs to be done.</li><li><strong>What is happening elsewhere? </strong>Has the usage environment changed? How are competitors’ products evolving? Have users’ objectives or constraints changed?</li></ul><h3><strong>Some tips to define the right metrics</strong></h3><p><strong>Identifying your</strong> <strong>product’s objectives</strong> must be defined when designing or developing a product because it is precisely these objectives that will enable you to define the right KPIs. Then the way to approach them is also important:</p><ul><li>Take a <strong>holistic view of the experience</strong>, do not just count the touchpoints, and their own performances.</li><li>As a starting point, <strong>focus on the most critical path</strong> — which steps are essential to delivering the expected result to the user? Put aside for the moment, the options, and the low probability cases (by the way, to know them, you need to measure it first!).</li><li><strong>Assess the risk of the change</strong> you want to implement. Based on your critical path, would this change bring enough value to be attempted?</li></ul><p><strong>Bonus gift tip:</strong> share user information as widely as possible in your organization. Create a dedicated channel for user test results, feedback, and other insights. Connect your prioritization processes to this channel. Share even beyond your team.</p><p>What ultimately matters are what we measure, because it varies according to the nature of the product:</p><ul><li>With a B2B, or internal product (dedicated to employees), the aim is to generate productivity gains using a more secure tool, leaving less room for human error. Therefore, you will measure employee efficiency improvement, comfort, and overall tool appreciation.</li><li>A B2C<strong> </strong>product, on the other hand, generally aims to engage customers, improve conversions, and boost brand loyalty, to increase turnover: in this case, you will measure the number of users reached and their overall level of satisfaction.</li></ul><p>Whatever the target audience, two fundamentals should be emphasized: <strong>simplicity</strong> and <strong>user autonomy</strong>. Tools must be easy to use and intuitive to ensure uncomplicated navigation. You can check these aspects by looking at the time spent on a page, the number of clicks to complete a task, or task and user-request response time, for instance. Finally, we can say that simplicity is a way to approach perfection.</p><blockquote>Perfection is reached, not when there is nothing more to add, but when there is nothing more to take away.</blockquote><blockquote><em>Antoine de Saint-Exupéry</em></blockquote><h3><strong>How to start?</strong></h3><p>There are a lot of frameworks mixing different practices, such as HEART: happiness (appreciation), engagement (behavior), adoption (behavior), retention (behavior), and task success (behavior). But it always should be adopted to the product specificities and objectives.</p><p>If we must mention tools, the ones to get the information about the appreciation are probably the simplest. It is only necessary to send surveys or polls or to add rating scales (NPS type) to the interfaces. In this case, the issue is in the formulation of the questions and the moment of their submission. Ideally, you should focus on a given insight that you want to verify and avoid being too generic.</p><p>Do not ask your users if they find your interface user-friendly, you will leave with a pile of cognitive biases and unactionable conclusions. Instead, ask them what they think of a given page or feature.</p><p>To obtain behavioral indicators, it is often necessary to use specific tools, which are numerous and often specialized. Still, it can be interesting to dig into the monitoring tools that are already operational on the product, they sometimes propose to track specific UX KPIs, or they can be deduced from the logs of page visits, time spent on the pages, past requests, etc.</p><p><strong>Bonus gift tip:</strong> if you are still lost, refer to the solutions and products competing with yours.</p><h3><strong>How ENGIE Design Studio can help to define the best KPIs for digital solutions</strong></h3><p>Many digital solutions have been developed since <a href="https://digital.engie.com/en">ENGIE Digital</a> was created and the Group has moved to industrialize these products. The measurement of their effectiveness and performance has become more than ever a major issue. One of the Design Studio’s tasks is to embed this change in design practices to increasingly demonstrate the value of design and user-centricity through foresight. With our team of experts, we align business goals with the experience that the product should offer.</p><p>The Design Studio was created at ENGIE Digital’s inception since we launched the Experience Index a year and a half ago to allow all ENGIE group users to auto-evaluate their solutions. A lot of data has been collected and we now need to go further. The Experience Index is a lever to identify improvement areas in the team’s organization and its DNA to become more user-centric in the everyday team’s life! We still have a long way to go to improve user experience and make user-centricity a fully-fledged corporate culture.</p><p>Being user-centric is more than just a concept: it is methods, tools, and practices to become a corporate culture.</p><p>Find out more about ENGIE Design Studio <a href="https://www.engie.design/">here</a>!</p><p>And if you wish to learn more about Experience Index, check out <a href="https://engiedigital.medium.com/first-experience-index-annual-report-the-key-in-transforming-our-user-experience-culture-e95bda5c6de3">this article</a>, written by yours truly 😉</p><img src="https://medium.com/_/stat?event=post.clientViewed&referrerSource=full_rss&postId=6ee37ac1f264" width="1" height="1" alt="">]]></content:encoded>
        </item>
        <item>
            <title><![CDATA[L’approche collaborative au cœur du Design Studio d’ENGIE]]></title>
            <link>https://engiedigital.medium.com/lapproche-collaborative-au-c%C5%93ur-du-design-studio-d-engie-206aaa8ae28a?source=rss-f6d35b915fa5------2</link>
            <guid isPermaLink="false">https://medium.com/p/206aaa8ae28a</guid>
            <category><![CDATA[design]]></category>
            <category><![CDATA[designer]]></category>
            <category><![CDATA[collaboration]]></category>
            <category><![CDATA[ux]]></category>
            <dc:creator><![CDATA[ENGIE Digital]]></dc:creator>
            <pubDate>Wed, 20 Jul 2022 10:33:16 GMT</pubDate>
            <atom:updated>2022-09-30T09:48:03.276Z</atom:updated>
            <content:encoded><![CDATA[<p>Par <a href="https://www.linkedin.com/in/romain-petit-5509484/">Romain Petit</a></p><p><em>Envisager le design et l’UX à travers le prisme de la collaboration à la fois interne et externe, tel est le mantra du Design Studio d’ENGIE Digital. Une approche innovante sur laquelle Romain Petit, Head of Design, revient pour nous.</em></p><h3><strong>Développer une communauté de designers au sein d’ENGIE Digital…</strong></h3><p>La collaboration est au cœur de l’organisation d’<a href="https://digital.engie.com/">ENGIE Digital</a>. En effet, les équipes dédiées aux différentes plateformes digitales développées in-house s’appuient sur des équipes transverses d’experts (Data, Tech, Operation…), mais aussi et surtout sur les designers UX et UI du Design Studio. Ce dernier rassemble plus d’une vingtaine de collaborateurs, tous intégrés pour la plupart à 100 % au cœur des équipes produit.</p><p>Pour favoriser le partage d’expérience et d’expertise, une communauté de pratique, appelée <em>practice Design</em>, a été créée en 2020. Le principe ? Mettre en commun des bonnes pratiques, sanctuariser des temps de co-création ou de réflexion et permettre à chacun de développer ses compétences et partager des standards communs (ie. méthodologie, design system, DesOps…).</p><p>La ritualisation de moments dédiés à la communauté a été instaurée pour fédérer les différents profils (UX, UI, front dev…) autour de sujets transverses. Durant le « Friday Studio » qui a lieu un vendredi sur deux, une ou plusieurs thématiques peuvent être abordées à travers différents formats (workshop, design critique, démonstration, retour d’expérience, veille et tendances…). L’objectif : permettre à chaque designer de s’appuyer sur ses pairs et de capitaliser sur un savoir-faire commun. Et parce que l’inspiration et les bonnes pratiques se trouvent aussi ailleurs, le Design Studio invite régulièrement des designers d’autres entreprises pour échanger ou même travailler sur une problématique commune. Par exemple, ce fût le cas dernièrement avec la venue au studio de la RATP pour participer à un workshop commun sur la thématique <a href="https://www.linkedin.com/pulse/comment-mesurer-limpact-business-de-lux-un-workshop-pour-ivankina">« comment mesurer l’impact du design »</a>, et lors d’un workshop avec les équipes de TotalEnergies, Thales et Air Liquide pour mettre en commun des retours d’expérience en matière de design system.</p><blockquote>La vocation de la communauté de pratique est d’offrir à chaque designer l’opportunité de faire partie intégrante d’un collectif, non seulement pour construire une culture et un savoir-faire commun, mais aussi et surtout pour se soutenir entre pairs, grandir ensemble et anticiper les évolutions de notre métier.</blockquote><blockquote><em>Romain Petit, Head of Design chez ENGIE</em></blockquote><h3><strong>… et diffuser les standards et la culture design dans l’ensemble du Groupe</strong></h3><p>Au-delà du travail effectué au sein d’ENGIE Digital, le Design Studio est également à l’origine et en charge du déploiement des standards design pour l’ensemble du Groupe à travers sa plateforme <a href="http://www.engie.design/">engie.design</a>.</p><p>Le studio<em> </em>a notamment conçu et déployé <a href="https://www.engie.design/fluid-design-system/">fluid, le <em>Design System</em> d’ENGIE</a> dont l’objectif est de standardiser les composants front et les templates, en ligne avec les <a href="https://www.engie.design/fluid-design-system/ux-principles/">principes UX</a> et l’identité de marque d’ENGIE. En à peine 3 ans, fluid est ainsi devenu le point de référence pour l’ensemble des designers et développeurs du Groupe. L’outil aide à concevoir rapidement des produits et services numériques, tout en garantissant une expérience utilisateur cohérente.</p><blockquote>Offrir un outil permettant de concevoir et développer des produits digitaux respectant les standards d’expérience et d’identité d’ENGIE est très fédérateur, non seulement pour les designers et les développeurs qui se retrouvent sur un terrain propice au partage et à l’amélioration continue, mais aussi pour nos parties prenantes métier et business qui voient leur go to market optimisé et des coûts significativement réduits en termes de temps de production. Le retour sur investissement est immédiat.</blockquote><blockquote><em>Romain Petit, Head of Design chez ENGIE</em></blockquote><p>Autre exemple de projet collaboratif mené par la <em>practice Design</em> au sein du Groupe, la mise en place de l’<a href="https://www.engie.design/experience-index/">Experience Index</a>. Cet outil de diagnostic, auquel on peut répondre en équipe, vise à évaluer la maturité des projets digitaux d’ENGIE en termes de centricité utilisateur et propose des recommandations personnalisées pour s’améliorer. En 2021, plus de 120 projets ont ainsi été examinés à la loupe à travers le monde. Le <a href="https://www.figma.com/proto/0urh6D8hREIvmQLMHq16ir/Exp-Index-Annual-Report-2022---PROD?page-id=0%3A1&amp;node-id=0%3A1485&amp;viewport=283%2C48%2C0.04&amp;scaling=contain&amp;hide-ui=1">premier rapport annuel de l’Experience Index</a> est également sorti. Il révèle en particulier que le partage des retours utilisateurs a augmenté de 20 % au sein des équipes qui ont répondu au test. En 2022, l’outil poursuit son déploiement en interne avec pour but de continuer à aligner les équipes ENGIE sur les standards en matière de design. Il offrira également sous peu une version open source pour d’autres entreprises.</p><figure><img alt="Infographie sur les 3 bonnes raisons d’adopter une approche transversale du design" src="https://cdn-images-1.medium.com/max/693/1*f7qWj7umbrCNmhOHIY5lrw.png" /></figure><h3><strong>Porter la voix du design en collaborant hors les murs</strong></h3><p>L’approche collaborative du design, loin d’être limitée aux frontières d’ENGIE, s’ouvre à l’externe sous plusieurs formes : à travers des échanges entre pairs (comme nous l’avons vu plus haut avec la RATP ou les REX sur la thématique Design System), mais aussi avec les étudiants et les plus jeunes générations.</p><p>À ce titre, ENGIE est partenaire stratégique de la Web School Factory qui forme les futurs managers du numérique avec une majeure en design. L’objectif principal du partenariat est de développer, d’accélérer et de co-construire des projets innovants par un modèle de collaboration et de confrontation original entre équipes opérationnelles et étudiants. Ce partenariat, animé par le Design Studio, permet ainsi aux équipes produit de bénéficier de collaborations avec les étudiants tout au long de l’année à travers différents modules (veille, anti master class, hackathon, projet de fin d’année…) en lien avec les différents cas d’usage portés par ENGIE Digital.</p><figure><img alt="Infographie sur les bénéfices pour ENGIE et pour les étudiants de la WSF (un partage d’expérience gagnant-gagnant)" src="https://cdn-images-1.medium.com/max/693/1*6RKxiyfZe7qUkUfpYabknA.png" /></figure><p>Enfin, qui dit collaboration externe dit aussi promotion, voire défense d’intérêts communs ! Ainsi, ENGIE a franchi une nouvelle étape dernièrement avec le lancement, en partenariat avec Adobe, de l’<em>Innovative Design Board</em>. L’idée : créer un cercle qui fédère des décideurs engagés dans le design et l’innovation pour partager des retours d’expérience et des convictions sur la pratique du design de service, mais aussi pour parler d’une seule voix sur des enjeux communs. Les membres fondateurs sont issus d’entreprises aux horizons variés telles que Schneider Electric, EDF, BPCE, Decathlon, Renault, Pernod Ricard, L’Oréal et bien sûr ENGIE. L’objectif étant de faire se rencontrer des responsables du design digital et du design produit pour influencer collectivement l’évolution du métier, son positionnement et sa reconnaissance interne comme externe.</p><blockquote>L’Innovative Design Board vise à partager une vision du design de service et des retours d’expérience venant d’horizons très différents, mais avec des enjeux similaires de passage à l’échelle. L’objectif est de mettre en commun nos convictions, faire émerger un collectif et porter d’une seule voix les intérêts d’une filière naissante, tant en interne au plus haut niveau de l’entreprise, qu’à l’externe vis-à-vis de l’ensemble de nos parties prenantes.</blockquote><blockquote><em>Romain Petit, Head of Design chez ENGIE</em></blockquote><figure><img alt="Infographie sur les thématiques abordées au sein de l’Innovative Design Board" src="https://cdn-images-1.medium.com/max/693/1*lJun5VOP2_wU7WZ6-h_f_Q.png" /></figure><p>En savoir plus sur la méthodologie, l’actualité, les ressources et les compétences du Design Studio : <a href="https://www.engie.design/">https://www.engie.design/</a></p><img src="https://medium.com/_/stat?event=post.clientViewed&referrerSource=full_rss&postId=206aaa8ae28a" width="1" height="1" alt="">]]></content:encoded>
        </item>
        <item>
            <title><![CDATA[The stakes of data cleaning for anomaly detection and predictive maintenance]]></title>
            <link>https://engiedigital.medium.com/the-stakes-of-data-cleaning-for-anomaly-detection-and-predictive-maintenance-fd5819e54d92?source=rss-f6d35b915fa5------2</link>
            <guid isPermaLink="false">https://medium.com/p/fd5819e54d92</guid>
            <category><![CDATA[predictive-maintenance]]></category>
            <category><![CDATA[data-cleaning]]></category>
            <category><![CDATA[anomaly-detection]]></category>
            <category><![CDATA[data]]></category>
            <dc:creator><![CDATA[ENGIE Digital]]></dc:creator>
            <pubDate>Mon, 27 Jun 2022 16:05:40 GMT</pubDate>
            <atom:updated>2022-06-27T16:05:40.874Z</atom:updated>
            <content:encoded><![CDATA[<p>By <a href="https://www.linkedin.com/in/paulponcet/">Paul Poncet</a></p><p>To improve the efficiency of ENGIE’s renewable energy assets (such as wind turbines), one of our key levers is to create and deploy predictive maintenance algorithms. This is part of our daily work at <a href="https://digital.engie.com/en">ENGIE Digital</a>, and we keep on deploying machine learning models on thousands of industrial assets, as explained in this <a href="https://engiedigital.medium.com/predictive-maintenance-from-prediction-to-decision-making-13a459dfb5da">former article</a>.</p><p>In this context, <strong>data cleaning is a mandatory step.</strong> Starting from our functional and technical understanding of potential data quality issues, we need to identify data artefacts and remove them whenever necessary, before training a machine learning model that might be biased otherwise, or before delivering new predictions.</p><p>In this article, we explain what we mean by “data artefacts” or “data surprises”, how we proceed in practice, and what challenges are still to be addressed.</p><h3><strong>Three kinds of data surprises in time series</strong></h3><p>Since many of our data science algorithms are made to detect <strong>real anomalies</strong> that occur in our wind turbines or solar farms, we need to distinguish between real anomalies and <strong>data artefacts</strong>: data or predictions that are anomalous for reasons independent of the asset’s health or performance.</p><blockquote><strong>What is a “real anomaly”? </strong>A real anomaly is an unexpected event occurring in an industrial asset that might cause an unavailability of the asset and/or a loss of energy production. Examples of such events (among many others) are:</blockquote><blockquote>- mechanical issues (accelerated wear and tear of a bearing in a wind turbine component for instance);</blockquote><blockquote>- sensor issues (some sensors might be essential for the safe operation of the asset);</blockquote><blockquote>- abnormal control loops (a wind turbine has its own internal regulation loops, should it be for facing main wind direction, restarting, stopping, etc. These loops need to be parametrized to work correctly).</blockquote><p>At a conceptual level, we make the distinction between <strong>three types of data surprises</strong>:</p><ul><li>“outlier values”;</li><li>“uncommon values”;</li><li>“unsupported values”.</li></ul><p><strong>Outlier values</strong> correspond to physically impossible or very unlikely data records. This is the case e.g. for a wind speed at 1,000 m/s, or a sequence of successive identical values (we talk about frozen values) due to a loss of communication in the data acquisition chain (see Figure 1).</p><figure><img alt="Diagram of a wind turbine’s active power against wind speed" src="https://cdn-images-1.medium.com/max/572/1*I3RoguTs71_UYo6QUMvNYA.png" /><figcaption>Figure 1: This “S” shape is well known in the wind power industry. It displays active power (in kW) against wind speed (in m/s) measured on a wind turbine with a 10 minutes timestamp. Here, the vertical line (points in red) reveals the presence of frozen values in wind speed measurement, due to a loss of communication in the data acquisition chain.</figcaption></figure><p><strong>Uncommon values</strong> are plausible data points related to an unusual situation (say the occurrence of a curtailment on a wind turbine for instance, see Figure 2). They may or may not be the sign of a real anomaly having occurred in the asset. But, in any case, <strong>at training time</strong> we do not need to know the cause of these uncommon values: we just remove them for our anomaly detection model to capture the normal behavior of the asset.</p><figure><img alt="Diagram of a wind turbine’s active power against wind speed" src="https://cdn-images-1.medium.com/max/572/1*norIHnyTPcRjiTtfiTOicw.png" /><figcaption>Figure 2: Data points in red correspond to a situation where the wind turbine is <em>curtailed</em>. Curtailment may be part of normal operation (and may occur to limit noise emissions or comply with bat and bird protection legislation for instance) or may be anomalous. At training time we don’t need to know if these points are normal or anomalous: we merely remove them.</figcaption></figure><p><strong>Unsupported values </strong>correspond to data points that, with respect to a specific topic or business question, must be left aside. Think of data points corresponding to a regular wind turbine stop: an asset being stopped is part of normal operation (see Figure 3). Hence, these points are not “outliers” nor “uncommon”, but are better filtered out to detect real anomalies.</p><blockquote><strong>How to define “unsupported values”? </strong>A data scientist fitting a regression model of the form <strong>y=f(x) + 𝜀</strong> may think of unsupported values as lying outside the domain of the regression function f ; this includes cases where y values are missing.</blockquote><figure><img alt="Diagram of a wind turbine’s active power against wind speed" src="https://cdn-images-1.medium.com/max/572/1*JIgHCLCCJb89oKWyrD_beg.png" /><figcaption>Figure 3: Data points in red correspond to situations where the wind turbine is stopped, either due to a lack of wind (wind speed less than 3.5 m/s typically), a maintenance action, or an unexpected event. Usually, we consider these points as “unsupported” and filter them out to avoid including them in the training of our models.</figcaption></figure><h3><strong>Dealing with data surprises</strong></h3><p>Outlier values are usually cleaned up by combining business rules (e.g. thresholds beyond which a data point is very unlikely) and simple statistical algorithms, then completed with appropriate algorithms if necessary.</p><p>To remove uncommon values, we may use either deterministic approaches (for instance we may know from our databases that there was a curtailment on a wind turbine during a given period) or statistical algorithms. In the latter case, simple algorithms are often not sufficient, and custom approaches are required.</p><p><strong>We choose not to fill data gaps created by the cleaning of uncommon values.</strong> Indeed, the occurrence of uncommon values may tell us that the industrial asset as a whole is in a “uncommon” state, in which case filling gaps would not be robust at all and would disturb the training of an anomaly detection model.</p><p>To manage these tasks, we have developed a simple, yet useful <strong>dedicated software package</strong> that powers all of our downstream predictive maintenance algorithms. This software includes methods for detecting:</p><ul><li>frozen values (notably related to a loss of communication with the asset / farm);</li><li>duplicate values;</li><li>simple outliers;</li><li>records (highest or lowest values ever observed);</li><li>outlier values, with statistical algorithms (e.g. <a href="https://en.wikipedia.org/wiki/Local_outlier_factor">Local Outlier Factor</a>, Influenced Outlierness, <a href="https://en.wikipedia.org/wiki/Isolation_forest">Isolation Forest</a>, and variants of these algorithms);</li><li>abnormal values, relying on the comparison of assets meant to behave similarly (think e.g. of wind turbines of the same wind farm).</li></ul><h3><strong>Conclusion</strong></h3><p>Data cleaning is a mandatory part of any machine learning pipeline. In practice, we observe that:</p><ul><li>it acts as a safeguard against undesired behavior of machine learning models at scoring time, and make our predictions more robust and reliable;</li><li>it contributes to reducing false positive rates when detecting anomalies on our assets; false positives are to be avoided, since we do not want wind turbine operators or maintainers to trigger a potentially expensive inspection or maintenance action for no reason;</li><li>it helps identifying data quality issues that, once confirmed, may be treated in the upstream data acquisition and storage workflow.</li></ul><p>These data cleaning steps can be applied either on-the-fly or at an early stage within databases, as part of a global data quality management process.</p><p>Yet, it remains uneasy to value data cleaning algorithms economically. Also, it might be a problem for an end-user to get no predictions/no alerts at a given moment because of data artefacts that have been cleaned up. These are some challenges that we have identified as next steps.</p><h4><strong>Acknowledgments</strong></h4><p>I thank my colleagues Céline Mallet and Régis Lavisse for their valuable suggestions to improve the content of this article.</p><img src="https://medium.com/_/stat?event=post.clientViewed&referrerSource=full_rss&postId=fd5819e54d92" width="1" height="1" alt="">]]></content:encoded>
        </item>
        <item>
            <title><![CDATA[First Experience Index annual report: the key in transforming our user experience culture]]></title>
            <link>https://engiedigital.medium.com/first-experience-index-annual-report-the-key-in-transforming-our-user-experience-culture-e95bda5c6de3?source=rss-f6d35b915fa5------2</link>
            <guid isPermaLink="false">https://medium.com/p/e95bda5c6de3</guid>
            <category><![CDATA[design]]></category>
            <category><![CDATA[feedback]]></category>
            <category><![CDATA[ux-design]]></category>
            <category><![CDATA[ux]]></category>
            <dc:creator><![CDATA[ENGIE Digital]]></dc:creator>
            <pubDate>Mon, 13 Jun 2022 11:13:28 GMT</pubDate>
            <atom:updated>2022-06-13T11:29:15.184Z</atom:updated>
            <content:encoded><![CDATA[<p>By <a href="https://www.linkedin.com/in/elenaivankina/">Elena Ivankina</a></p><p>ENGIE Design Studio has published its first Experience Index annual report, a year and a half after the Experience Index launch. Hundreds of ENGIE digital projects were analyzed, revealing key learnings on where we stand today in terms of user experience (UX) maturity and how to improve our digital product performance.</p><h3><strong>Back to basics: it all starts with the end-users</strong></h3><p>What is the point of spending time and money on building a digital solution if it does not fix user pain points, nor generate sufficient value? None! This is where UX comes into play: it ensures that a solution is useful, efficient, user friendly and viable for the company. User Experience is a strong lever to improve our digital projects’ business efficiency.</p><p>It is now obvious that UX makes a decisive contribution to rapid product adoption. It also opens numerous opportunities for competitive advantage, customer engagement and loyalty, employee efficiency and comfort, and the optimization of our organizations and processes. UX is a powerful asset in improving your digital projects’ business efficiency and, in fine, the Group’s performance.</p><h3><strong>The Experience Index: measuring the user experience’s delivery in digital products across ENGIE</strong></h3><p>Where does your project stand in terms of user experience? This is the question behind the <a href="https://www.engie.design/experience-index/">Experience Index</a> launched in 2021 throughout all ENGIE’s project teams. The team self-assessment barometer was created by <a href="https://digital.engie.com/en">ENGIE Digital</a>’s Design Studio to measure the user experience of digital products and services across <a href="https://www.engie.com/en">ENGIE</a>.</p><h4><strong>So, how does it work?</strong></h4><p>By answering a series of questions on key fundamentals, you get a score indicating how far along you are in your design journey. You then get a UX maturity level for your project, based on the understanding of, and responsiveness to your users’ needs, as well as how aligned your team is around these aspects. Four levels of maturity have been determined:</p><ul><li><strong>Emerging:</strong> very low priority is given to UX, as projects are managed by tech and business KPI’s;</li><li><strong>Tactical:</strong> UX is integrated into the project’s lifecycle, but without vision or regularity;</li><li><strong>Mastered:</strong> UX strategy, standards, and processes are followed and fully integrated into the project life cycle;</li><li><strong>Vital:</strong> design is a strategic asset and part of the project’s DNA.</li></ul><p>Based on your score, the Experience Index will give you personalized advice based on four key fundamentals: User Experience, Team, Methodology, and Leadership.</p><figure><img alt="Picture of the Experience Index’s four key fundamentals" src="https://cdn-images-1.medium.com/max/628/1*W-krVJ4oO-0ZwZDphMw4lw.png" /><figcaption>The Experience Index’s four key fundamentals</figcaption></figure><h3><strong>A first Experience Index Annual report to understand where ENGIE stands in terms of UX maturity</strong></h3><p>Our goal was to help ENGIE Digital’s product teams focus on user issues by designing this tool. Then we decided to make it available to all the digital projects in the Group. It’s an important step towards scaling UX practices and culture. Today, we get requests even from outside of ENGIE, and we are thinking about offering this tool to some external partners. It’s great! As a large industrial Group, our first annual report shares UX state of the art, but other organizations will also find their solutions to their issues here.</p><p>The results of <strong>126 projects</strong> and services from 298 respondents working in four Global Business Units (GBU’s) across ENGIE’s main regions have been consolidated into the <strong>first Experience Index annual report</strong>. With an overall score at the “<strong>Tactical</strong>” level, the data analysis highlighted <strong>four major insights </strong>that will help us go further and push user experience as a real value provider to our projects.</p><h4><strong>1. The UX paradox</strong></h4><p>User experience and user data issues are well known by respondents, with 83% collecting users’ feedback. But 69% do not consider it as a high priority for their products! The UX paradox can be explained by the fact that we still trust traditional business and tech indicators for regular projects, yet they are not relevant enough to track user product perception. The challenge now is to update these metrics so that they can improve our products and our business efficiency.</p><h4><strong>2. It’s a match!</strong></h4><p>Sharing user feedback has increased by 20%. Our teams are bonding over user knowledge, but not at the same rhythm. In some of the most UX advanced projects, user insights are increasingly seen as a major issue for product or project development. This is a good point, but team expectations regarding user perspective remain vague, as team members do not know concretely how to exploit this, and what the designer’s role and actions are in this respect. When the intentions match, the next step is to get to know each other better!</p><h4><strong>3. Connecting the dots</strong></h4><p>Only 23% of respondents ignore whether UX principles are applied, meaning that design fundamentals are extensively shared. From now on, competitive advantage can only be achieved if these principles are put into practice and connected to tech and business issues. A great deal of information is available about users, product usage, and technical issues, etc. So, let us connect them with specific experience indicators.</p><h4><strong>4. Power up!</strong></h4><p>58% of respondents perceive and understand how design impacts product adoption. Next on the agenda is to help them realize that user experience is a business performance lever. Respondents have underlined the necessity to create a common UX strategy, with more synergies between business, design, and tech teams. A common vision and framework are key to clarifying management and team expectations in terms of the experience we deliver and the increased efficiency we can generate through user insights at all product life cycle stages.</p><h3><strong>How to go one step further: what is next?</strong></h3><p>Developing a UX approach implies an active engagement for the team and the business stakeholders. Here are some basic principles to focus on from the start:</p><ul><li>UX design is a rigorous, rational approach, so <strong>plan, budget, and organize UX phases in your product lifecycle</strong>.<strong> Structure how the team interacts with each other in this respect;</strong></li><li>Measure the UX impact on your business KPIs;</li><li><strong>Designate a user experience leader</strong> to share user data within the team regularly, and build together a sharing ritual or channel that fits with the team’s organization;</li><li><strong>Systematize advanced interactions</strong> with users by usability testing, feedback loops, and behavior analytics etc.</li></ul><p>Wish to read the full report? Check it out <a href="https://www.figma.com/proto/0urh6D8hREIvmQLMHq16ir/Exp-Index-Annual-Report-2022---PROD?page-id=0%3A1&amp;node-id=0%3A1485&amp;viewport=283%2C48%2C0.04&amp;scaling=contain&amp;hide-ui=1">here</a>!</p><p>Be part of the next edition: assess or re-assess your project’s user centricity, receive your score, and get tailored recommendations that will allow you to go further and increase your design maturity! Get your score <a href="https://www.engie.design/experience-index/">here</a> (ENGIE account required, sorry!).</p><img src="https://medium.com/_/stat?event=post.clientViewed&referrerSource=full_rss&postId=e95bda5c6de3" width="1" height="1" alt="">]]></content:encoded>
        </item>
        <item>
            <title><![CDATA[Designing keyboard interactions to improve accessibility: the eCare Monitoring use case]]></title>
            <link>https://engiedigital.medium.com/designing-keyboard-interactions-to-improve-accessibility-the-ecare-monitoring-use-case-222da6a799f8?source=rss-f6d35b915fa5------2</link>
            <guid isPermaLink="false">https://medium.com/p/222da6a799f8</guid>
            <category><![CDATA[accessibility]]></category>
            <category><![CDATA[designer]]></category>
            <category><![CDATA[design]]></category>
            <category><![CDATA[use-cases]]></category>
            <category><![CDATA[keyboard-shortcuts]]></category>
            <dc:creator><![CDATA[ENGIE Digital]]></dc:creator>
            <pubDate>Tue, 07 Jun 2022 11:16:46 GMT</pubDate>
            <atom:updated>2022-06-07T11:16:46.956Z</atom:updated>
            <content:encoded><![CDATA[<p>By <a href="https://www.linkedin.com/in/celinegbouillon/">Céline Bouillon</a></p><p>Last February, I shared an article to <a href="https://engiedigital.medium.com/advocating-for-accessibility-by-design-cc204f095028">advocate for accessibility by design</a>. If you don’t know what accessibility is, or have any doubt about its importance, feel free to read this paper first. Don’t hesitate to share your thoughts and engage in the conversation! In this follow-up, as promised, I would like to share how we’ve been working on including keyboard navigation at <a href="https://digital.engie.com/en/solutions/ecare">eCare</a>.</p><h3>Confession: we (too) used to not take keyboard into account</h3><p>When we started working on improving accessibility for eCare Monitoring — a piece of software which enables experts to remotely monitor connected installations — some of our blinders needed to be removed. Amongst these was how people would navigate the Web. We used to design and develop mainly thinking about mouse navigation. And by doing so, we were totally forgetting about keyboard users and creating massive accessibility issues.</p><p>You might wonder what the link is between keyboard and accessibility. Some users, with or without disabilities, rely heavily on keyboard navigation. For instance, you can think of:</p><ol><li>people using <a href="https://en.wikipedia.org/wiki/Screen_reader">a screen reader</a> (e.g., with visual impairments or learning disabilities) — which may be coupled with a keyboard to navigate from one focusable element to the next,</li><li>individuals who have difficulties using a mouse — it requires fine motor skills,</li><li><a href="https://en.wikipedia.org/wiki/Power_user">power users</a> who are used to relying on a keyboard for efficiency…</li></ol><figure><img alt="A desktop keyboard with a refreshable Braille display" src="https://cdn-images-1.medium.com/max/1024/1*qmCohrS7BKBRRP6p4K7xWA.jpeg" /><figcaption>People with visual impairments may use a screen reader along with a keyboard, or even a <a href="https://en.wikipedia.org/wiki/Refreshable_braille_display">refreshable Braille display</a>, to navigate the web. Photo credits: Unsplash</figcaption></figure><p>Unfortunately, keyboard navigation keeps being trampled over notably because:</p><ol><li>the focus indicator is seen as an unsightly addition to user interfaces (UI) and is consequently hidden in the development phase,</li><li>knowledge is lacking on how people interact with components (e.g., links, buttons, carousels…) when they use a keyboard,</li><li>designing and implementing these interactions are seen as bringing unnecessary complexity to the table.</li></ol><p>This was the case at eCare. Missing keyboard usage entailed that some people were just not able to use our software. As I wrote in my first article, I think empathy is key in design. And here was an opportunity to learn and work on this overlooked type of navigation to make our software more inclusive.</p><h3>Bringing elements of keyboard navigation to our design process</h3><p>To add keyboard navigation into eCare Monitoring’s experience, I first worked on making the focus indicator visible. To do so, I leveraged an excellent article written by Sara Soueidan: <a href="https://www.sarasoueidan.com/blog/focus-indicators/">A guide to designing accessible, WCAG-compliant focus indicators</a>. I mixed her takeaways on <a href="https://www.w3.org/WAI/standards-guidelines/wcag/new-in-22/#guideline-24-navigable">contrasting areas</a>, based on the Web Content Accessibility Guidelines (WCAG), with <a href="https://www.engie.design/fluid-design-system/">Fluid — ENGIE’s Design System</a> to try and find the “right” focus indicator for various components we use in our software.</p><figure><img alt="Six button design attempts at finding the right balance for the focus indicator" src="https://cdn-images-1.medium.com/max/638/1*i7IZP5RwTo9hdisc9KHc1w.jpeg" /><figcaption>Here is an example of my attempts at finding the right balance (contrast, solid outline width, outline offset…) for our focus indicator. The one on the bottom right, with its widest offset and solid border, seemed to be the most visible</figcaption></figure><p>Once the focus was visible, I needed to design keyboard interactions for complex components (e.g., a listbox) to help and guide further front developments. As designers and developers, we can count on a technical specification, <a href="https://www.w3.org/TR/wai-aria-1.1/">WAI-ARIA</a> — which stands for Web Accessibility Initiative — Accessible Rich Internet Applications — to help us define user-expected keyboard interactions.</p><p>Including these interactions to eCare Monitoring’s user experience required back-and-forth discussions between several roles in the team: design, accessibility expertise and development. This was especially true for custom combinations of complex components. The help of an accessibility expert was critical here so that we could benefit from their past hands-on experiences. These included impact on development and “real-life” usage of complex components.</p><p>The following process would typically unfold:</p><ol><li>The accessibility expert and I would work on keyboard interaction proposals,</li><li>We would discuss and challenge our proposals iteratively to land on a common one that would be specified along the way. The key here was to find the right balance between User Experience (UX) and development complexity,</li><li>Developers would be included to refine our common proposal,</li><li>Support would be provided once the component was in the development phase — some interactions could be adjusted if needed.</li></ol><p>The first step was quite a challenge, given that I had few knowledge on keyboard experience at the beginning. Here is how I would work on my design proposals:</p><ol><li>Learn more about keyboard interactions and try to navigate on my own, based on components I knew were accessible. For instance, this is the case in official design systems like the <a href="https://designsystem.digital.gov/">US Web Design System</a>, the <a href="https://design-system.service.gov.uk/">Gov UK Design System</a> and let’s not forget about the <a href="https://www.systeme-de-design.gouv.fr/">Système de Design de l’État</a></li><li>Deep-dive into relevant complex components and their associated <a href="https://www.w3.org/TR/wai-aria-practices-1.1/#intro">WAI-ARIA authoring practices</a></li><li>Prototype various interactions in figma (keyboard interactions are indeed supported 🎉) and check for consistency in terms of user experience</li></ol><figure><img alt="Interaction details for a given complex component in Figma with a key combination as trigger" src="https://cdn-images-1.medium.com/max/496/1*ugbihiyplqB7_XlhdrlGkQ.png" /><figcaption>This is an example of interaction details for a given complex component in figma. The use of <em>Shift+Tab</em> — so that the user may move focus back to the previous focusable element — is prototyped here</figcaption></figure><p>Let’s take a more concrete example to illustrate the whole process. In eCare Monitoring, a given category of filters helps experts select one type of connected installations in a data table. This select filter could be perceived as a button triggering a list of buttons, seeing how pressing one in the list would trigger the filtering of the table. Alternatively, this filter looks like a dropdown, so some keyboard users might expect to navigate a list of options. These 2 proposals emerged in the discussions and seemed to be technically feasible. So, we decided to integrate, prototype and specify both ways of interacting:</p><ul><li>in a list of buttons — using Tab to move focus to the next button in the list, for example</li><li>in a collapsible dropdown — using the down arrow key to move from one option to the next for instance</li></ul><figure><img alt="An example of filter on types of installations in eCare Monitoring. It could be a list of buttons or a list of options in a dropdown menu" src="https://cdn-images-1.medium.com/max/610/1*mfPamW-B1G2HPpGuR8qjQg.jpeg" /><figcaption>User-expected keyboard interactions will depend on how the component may be perceived (visually, by hearing or touch). Here this filter “type of installations” (boiler, photovoltaic panels, heat pump) could either be a list of buttons or a list of options in a dropdown menu</figcaption></figure><h3>User testing is crucial…and we are not there yet</h3><p>I believe this adjustment in our production process was important and worthwhile. Constant discussions between different roles in the team and accessibility expertise proved key to ensure experience consistency in our monitoring service. It is worth noting though that this is just one step in our process. There are still several steps in which accessibility could be improved and embedded.</p><p>I think user testing is the next phase we should enhance. For example, include keyboard and assistive technologies (AT) users to collaborate in our prototyping phase, and once the experience is live, could validate our proposals even further. I believe that relying solely on accessibility expertise is not enough and represents a bias to mitigate with additional user feedback. After all, what is design without testing, bias mitigation and iterations?</p><p>And you, how do you proceed? Do you prototype and test all types of interactions? Don’t hesitate to share if you have any advice or thoughts on this! 😊</p><img src="https://medium.com/_/stat?event=post.clientViewed&referrerSource=full_rss&postId=222da6a799f8" width="1" height="1" alt="">]]></content:encoded>
        </item>
        <item>
            <title><![CDATA[Measuring the performance of DevOps teams at ENGIE Digital]]></title>
            <link>https://engiedigital.medium.com/measuring-the-performance-of-devops-teams-at-engie-digital-883a6fe8220b?source=rss-f6d35b915fa5------2</link>
            <guid isPermaLink="false">https://medium.com/p/883a6fe8220b</guid>
            <category><![CDATA[dora-metrics]]></category>
            <category><![CDATA[devops]]></category>
            <category><![CDATA[technology]]></category>
            <category><![CDATA[software-company]]></category>
            <category><![CDATA[cto]]></category>
            <dc:creator><![CDATA[ENGIE Digital]]></dc:creator>
            <pubDate>Tue, 31 May 2022 14:37:36 GMT</pubDate>
            <atom:updated>2022-06-14T10:19:49.228Z</atom:updated>
            <content:encoded><![CDATA[<h3>Measuring the performance of DevOps teams at ENGIE Digital or how we automated DORA Metrics for ENGIE’s digital factory</h3><p><strong>Feedback</strong>, by <a href="https://www.linkedin.com/in/davidhenry1979">David Henry</a></p><p><em>At ENGIE Digital, the ENGIE group’s software company, the CTO’s team (Chief Technology Officer) automated a system to measure delivery performance across the entire organization. Putting the concept into practice was a chance to ask ourselves a long list of questions and advance our thought process on the topic in general. We share with you here our assumptions, decisions and lessons learned:</em></p><ul><li><em>The measurement choices — DORA metrics and others — and how to automate them (from </em><a href="https://www.devops-research.com/research.html"><em>DORA:</em></a><em> </em><a href="https://www.devops-research.com/research.html"><em>DevOps Research and Assessment</em></a><em> that is now part of Alphabet/Google),</em></li><li><em>The advantages for management: views of anomalies and trends, situation analysis, and observations on the impact of improvement actions.</em></li><li><em>The unintended outcomes: the project introduced more control and rigor in our practices.</em></li></ul><h3>Introduction</h3><p>Among their many missions, ENGIE Digital’s CTO Vincent Derenty and his team are tasked with supporting product teams to boost operational efficiency. The product teams design and develop strategic digital solutions for the ENGIE Group. The CTO’s team consults them as experts on topics like cybersecurity, agile methodologies, software architecture and development practices.</p><p>One of the first issues that comes up about improvement is how to measure it. It’s impossible to manage and coordinate an improvement action without knowing where you’re coming from, determining where you’re going and being able to see the results of our initiatives.</p><blockquote>“If you can’t measure it, you can’t manage it,”<em> </em>as <a href="https://en.wikipedia.org/wiki/Peter_Drucker">Peter Drucker</a> would say.</blockquote><p>When it comes to measuring the performance of DevOps teams, the gold standard is the <a href="https://www.devops-research.com/research.html#reports">State of DevOps Reports</a>and its resulting analysis in the <a href="https://itrevolution.com/accelerate-book/">Accelerate</a> research, which draws a correlation between a company’s operational performance and its business performance. It defines operational performance as a set of <a href="https://itrevolution.com/24-key-capabilities-to-drive-improvement-in-software-delivery/">24 capabilities</a> classified into five categories: Continuous Delivery, Architecture, Product and Process, Lean Management and Monitoring, and Culture. These capabilities are a list of drivers for improving efficiency in an organization’s software delivery.</p><p>Believing it was a solid approach, we decided to set up an automated data collection system that now helps us compute and track the four recommended basic measurements commonly known as DORA metrics:</p><ul><li><strong>Two speed indicators:</strong></li></ul><p>-Deployment frequency</p><p>-Lead time for changes</p><ul><li><strong>Two stability indicators:</strong></li></ul><p>-Mean time to restore</p><p>-Change failure rate</p><p>It wasn’t the decision itself that took the most effort so much as the implementation. The concept behind what each indicator needs to measure is clear enough, but how it works on the ground is not as obvious. Questions arose very quickly: What operational data do we measure? How do we get these data? How do we process them? And, as is often the case, we realized that the devil is in the details.</p><p>Our goal here is to tell you how we took these measurements from broad concept to real-life implementation by creating a link to the operational practices that teams use in the field.</p><p>We also want to take this opportunity to introduce four more indicators that we added to the DORA metrics. These new ones are more focused on workflow within teams:</p><ul><li>Deployment chronology</li><li>Activity split</li><li>Feature flow</li><li>Defect flow</li></ul><p>By “flow” we mean a quest for fluidity, which <a href="http://reinertsenassociates.com/about/">Don Reinertsen</a> does a good job explaining and illustrating in his book <a href="http://www.startuplessonslearned.com/2009/07/principles-of-product-development-flow.html?ref=https://product-frameworks.com"><em>Product Development Flow</em></a>.</p><p>All of these indicators were designed to give teams a clear bird’s-eye view of their actual operations, to map out their situations, and pick out anomalies or trends in their own delivery.</p><p>The teams access these metrics on a <a href="https://developer.atlassian.com/server/confluence/confluence-plugin-guide/">Confluence plugin</a>, and we will show you screen shots of the interface. <a href="https://www.atlassian.com/software/confluence">Confluence</a> is one of the most commonly used knowledge management tool in this context.</p><h4>Background</h4><p>To gain a better understanding of the issues we were tackling, you need a brief overview of the background.</p><p>As we mentioned, the CTO’s team at ENGIE Digital supports a number of teams making digital products that give the ENGIE Group a competitive edge in its markets. The majority of these products and the teams that produce them stem from various entities in the Group that predated ENGIE Digital, where they were brought together after being founded.</p><ul><li>ENGIE Digital has about a dozen of these products.</li><li>They may be produced by teams ranging from 10 to over 60 people that include product management and tech.</li><li>The teams are all separate and accountable. They choose their own development and collaboration practices and workflows, and to some degree their tools, technology and architecture. While this was definitely not a distinct factor for us, we should point out that the history of all these teams meant there were relative disparities in their practices.</li></ul><p>We collected raw data from tools that the teams used the most:</p><ul><li><a href="https://github.com/">GitHub</a> (source code manager): The best way to see the codebase activity.</li><li><a href="https://www.atlassian.com/software/jira">Jira</a> (organizational tool): A closeup view of team workflows.</li></ul><h4>High-level implementation</h4><p>Below is the process we chose.</p><p>The program relies on a <a href="https://en.wikipedia.org/wiki/Representational_state_transfer">REST API</a> to collect data:</p><ul><li>Pushed by the projects themselves using scripts they have at hand when the data come from GitHub’s <a href="https://en.wikipedia.org/wiki/Continuous_integration">continuous integration</a> (CI) factory.</li><li>Pulled by a <a href="https://en.wikipedia.org/wiki/Cron">cron</a> when the data come from Jira.</li></ul><p>And that exposes them for display in the Confluence plugin.</p><figure><img alt="Diagram of the high-level architecture for the delivery performance metrics reporting system." src="https://cdn-images-1.medium.com/max/988/1*MTl87b2Y7rGyT4baTyk5jg.png" /><figcaption>High-level architecture</figcaption></figure><h4>Position</h4><p>As we will see, a lot of the thought process revolved around choosing the data that would be processed to build each indicator.</p><p>These considerations also involved a related objective of making it easier for the teams to adopt our approach. In other words, it had to be the least intrusive and prescriptive in terms of their existing practices, as well as adaptable to the wide range of routines in place. To varying degrees of success, we tried to ask as little effort of the teams as possible when collecting the data.</p><p>Our decision was partly based on how to walk this tightrope between availability, ease of acquisition and data accuracy.</p><h3>Definitions, calculations and views</h3><p>So, the same process was used for each metric: transition from intent to technical implementation by defining calculations and concepts, and making sure they were still adapted to the team practices and routines.</p><p>Let’s see how this worked for each indicator.</p><h4>1. Deployment frequency</h4><h4>Concept</h4><p>The goal was to calculate the deployment frequency over a given time period, which meant counting the number of <strong>deploys to</strong> <strong>production </strong>for new <strong>product versions</strong> under assessment during that period.</p><p>Beyond deployment frequency, these two concepts structured what came next.</p><h4>Deploy to production (DTP)</h4><p>What is a DTP? Seemingly simple at first, the matter became more complex when we factored in, for example, the technical reality of micro-service architectures and multi-repository organizations.</p><p>We decided on the following definition for deploy to production: Providing a new version of a product to users by simultaneously deploying one or more <a href="https://docs.github.com/en/repositories/creating-and-managing-repositories/about-repositories">repositories</a>.</p><p>So a DTP must be differentiated from:</p><ul><li><strong>A repository deployment:</strong> When several repositories are deployed in the same DTP, we consider it to be a single DTP event.<br><em>For example: If I push new code to production every two weeks by updating five repositories each time, the frequency would be two DTPs per month, not 10 (2 x 5 repositories).</em></li><li><strong>Commissioning (COM):</strong> Even when the new code to production wouldn’t alter how the application behaves for the end user (i.e., <a href="https://en.wikipedia.org/wiki/Feature_toggle">feature flipping/flags/toggles</a>), we count one DTP event for each update.<br><em>For example: If I push 10 deployments in a row into production for a new functionality’s various components over a period of one month and ultimately activate the finished functionality on the 10th deployment. I would count a frequency of 10 DTPs per month, not one (which would be the COM).</em></li></ul><p>So the idea is to count the deliveries of <strong>new product versions</strong>. We also had to work on defining what a version is and putting it into practice to be able to automate it.</p><h4>The product version</h4><p>Therefore, counting DTPs would mean counting the number of product versions deployed within a given timeframe. But we still had to effectively version out the product, which may seem easy, but it’s not all that clear-cut.</p><p>In a rare prescriptive move, we asked product teams to systematically name the versions, typically by using the standard <a href="https://semver.org/">Semantic Versioning</a> (SemVer): X.Y.Z where X is the major version number, Y is the minor version number and Z is the fix number for a patch version. Having everyone use this convention made it easier for us to tally all the X.Y.Z combinations that were brought to production, and thus to count the DTPs.</p><p>Again, we had to be mindful of the multi-repository organizations because some of them don’t even use this product version concept. They only had repository versions, whereas product versions implicitly remained with the combination of repository versions and had no naming convention in the source codes. This could be improved in view of a need to restore systems to a previous state.</p><p>In this case, we had to offer these teams a procedure for <a href="https://git-scm.com/book/en/v2/Git-Basics-Tagging">tagging</a> application repositories when DTPs occurred with the same annotated tag that contained an incremental version number and its deployment date. On the ground, this was a <a href="https://www.jenkins.io/doc/">Jenkins</a> job done after the deployments in production.</p><h4>Calculating deployment frequency</h4><p>GitHub was our go-to source. Once the version tagging job was set up, we will count all the product version tags (different X.Y.Z combinations qualified by date) within a certain time period to find out the deployment frequency for that period.</p><h4>2. Lead time for changes</h4><h4>Concept</h4><p>The idea was to calculate how much time passed between when a development is <strong>considered finished</strong> and when it is <strong>actually deployed</strong> <strong>to production</strong>. This indicator is meant to measure a time period, i.e., how much time does a code change take to be deployed in production?</p><p>We had the most trouble figuring out this indicator.</p><h4>A composite indicator</h4><p>First, bear in mind that there are two major types of factors impacting the time a development takes to go to production:</p><ul><li><strong>Technical efficiency:</strong> The capacity to quickly deploy code in the production environment by, for example, automating the <a href="https://www.atlassian.com/continuous-delivery/pipeline">deployment pipeline</a> or the <a href="https://en.wikipedia.org/wiki/Regression_testing">regression tests</a>.</li><li><strong>Organizational efficiency:</strong> Best collaboration practices that minimize time spent verifying or waiting for compliance verification (code review, functional review, business-side approval) and reduces the amount of redos needed by opening an effective communication channel between functional and technical upstream of developments.</li></ul><p>Seen from this lens, “lead time for changes” is a composite indicator that, depending on the case, demonstrates both of these aspects to varying degrees. Therein lies why what is measured will change from one team to another.</p><h4>A sensitive indicator in a team dynamic</h4><p>The relationships between workflow, state of the source code (mostly <a href="https://www.atlassian.com/git/tutorials/using-branches/git-merge">merge</a> on main branches) and deployment environments vary widely depending on the organizational and technical choices teams make — largely based on <a href="https://reviewpad.com/blog/github-flow-trunk-based-development-and-code-reviews/">branching strategies</a> (Git Flow, GitHub Flow, trunk-based, etc.), setting up an <a href="https://docs.gitlab.com/ee/ci/review_apps/">app review</a> process, whether regression testing is automated, and if related fixes are immediate or delayed.</p><p>Therefore, depending on a team’s organization and options, the indicator’s various components come into play based on the methods and at different times. Consequently, they will be accurately represented to a greater or lesser degree, which is what we used as the starting point to calculate “lead time for changes.”</p><p>This shows how important it is to be mindful when using this indicator to compare teams with disparate organizational situations and technical options.</p><h4>Determining version age</h4><p>As we saw earlier, a product version is what goes to production. Yet a version can contain several increments like user stories and fixes whose precedence often varies by factors such as complexity, work hours and industry value.</p><p>As recommended <a href="https://cloud.google.com/blog/products/devops-sre/using-the-four-keys-to-measure-your-devops-performance">in the literature</a>, we decided to use <a href="https://www.atlassian.com/git/tutorials/saving-changes/git-commit">commits</a> as the standard increment. The advantage was we could assess consistent unitary components from a technical implementation perspective because it saved us from having to match up the more abstract concepts of tasks and user stories.</p><p>That then brings up the question of version “age.” Using a version’s commits as a basis, how could we figure out the time between when the code is considered finished and when the version goes to production?</p><p>Once again, we turned to the literature and determined that a version’s age is the average age of its commits.</p><p>[RESERVE] There are a few limitations to an approach based on the average commit age. The most noteworthy is that disruptive workflow practices like late corrections create commits on main/master branches that make the versions look younger than they are.</p><p>Let’s look at the example of a development that’s lagging behind on a branch and built in a context where I’m forced to merge to get to delivery and acceptance. If, after merging the development, eight anomalies are found during delivery and acceptance and 12 commits are made to correct them, the correction commits wipe out the effect of the branch’s age. In this example, the more fixes done downstream in the workflow, the better the indicator’s value. But we are looking for the opposite.</p><p>To get around this, we are considering not equating the average commit age with the version age, but rather the time that passes between the version’s oldest commit and its DTP. Then the best practices will be endorsed (<a href="https://readwrite.com/2021/04/24/stop-starting-start-finishing-with-agile/">Stop starting, start finishing</a>: limit work in progress and stock up on ready code).</p><h4>Calculating lead time for changes</h4><p>While dealing with the commit concept, we naturally turned to GitHub for guidance.</p><p>To calculate an age, you have to measure the time between a start date and an end date.</p><ul><li>The start date that we chose for each commit was the <a href="https://alexpeattie.com/blog/working-with-dates-in-git/">committer date</a> (GIT_COMMITTER_DATE), which is the date of a commit’s last edit. The thinking was that this data point gets us as close as possible to the <em>considered finished</em> idea.</li><li>The end date is the deployment date shown on the tag for the product version where the commit is embedded.</li></ul><p>Then you just have to calculate the version’s age by figuring out the average age of its commits.</p><h4>3. Mean time to restore</h4><h4>Concept</h4><p>The idea was to determine the mean time needed to restore a service when there’s an <strong>incident</strong>. The impacted service could be all or part of the product in question.</p><h4>Incident</h4><p>Insofar as an incident is an event whose resolution time has to be deducted, our definition of an incident is related to its operational outcome. If a product malfunction disrupts users so much that one or more team members must drop whatever they’re doing to take care of it, then that is an incident.</p><p>This definition allows us to remain contextual, and to accommodate each team’s uses and everyday realities when it comes to incidents.</p><h4>Calculating mean time to restore</h4><p>We used Jira for guidance on this type of event. It actually helped us find tickets in the workflow that could be described as incidents and to monitor their resolution time.</p><p>Since our approach used workflow instead of technical monitoring, we were able to stick to our definition of incident based on user disruption, and not solely on availability or performance criteria.</p><p>For each team, then, the task was to see which tickets fit our definition of incident. In Jira, we used:</p><ul><li>Ticket type (<a href="https://support.atlassian.com/jira-cloud-administration/docs/what-are-issue-types/">issueType</a>)</li><li>Priority level (<a href="https://support.atlassian.com/jira-cloud-administration/docs/what-are-issue-statuses-priorities-and-resolutions/">priority</a>)</li></ul><p>The issueType/priority pairs specific to each team are what make the tickets identifiable as incidents for the team.</p><p>And for each of these tickets, we calculate the time between the date the ticket was opened (creationDate) and the date it was resolved (resolutionDate).</p><p>It should be noted that you have to make sure the resolution date (resolutionDate) is determined within the Jira configuration by entering a state that means the code was deployed in production.</p><p>We will take the average resolution time in a given period for all the incidents resolved during that period.</p><p>[RESERVE] Since a ticket’s lifecycle in Jira may not reflect what actually happens on an operational level, the times calculated are still representative because a ticket’s creation date cannot be the same time the issue detected resulted in an incident, let alone when it first happened. Similarly, the resolution date won’t necessarily be the exact time the fix itself went to production.</p><h4>4. Change failure rate</h4><h4>Concept</h4><p>This was about determining the proportion of <strong>failed changes</strong> within a given time period.</p><h4>Failed changes</h4><p>To calculate DTPs, we leveraged the product version tagging convention in place and counted each DTP as a change, then each failed change as a DTP that was immediately followed up with the fixed version’s deployment.</p><p>It’s very easy to figure this out using the standard X.Y.Z SemVer. It’s a corrected version when Z ≠ 0. We will, therefore, deduce that the previous version was a failure.</p><h4>Calculating the change failure rate</h4><p>We used the same GitHub deployment frequency data as a basis.</p><p>We calculate the proportion of X.Y.Z corrected versions where Z ≠ 0 out of all the X.Y.Z versions that had a DTP in that period.</p><p>We should note that a corrected version could itself be a failure, i.e., X.Y.1 then X.Y.2, in which case there would be two failed versions (X.Y.0 and X.Y.1) and one successful version (X.Y.2). So the change failure rate for this period would be 2/3 (two failed DTPs out of a total of three DTPs)</p><p>[RESERVE] The last example clearly shows that this indicator can be quite inconsistent for short periods or when the deployment frequency is low.</p><h4>5. Visualizing delivery performance</h4><p>A team’s delivery performance is shown on a graph with four curves that each represent a DORA metric. This is an on-demand graph built for any date range the user chooses. It can calculate an average in increments of a week, a month or a quarter.</p><p>This visual reveals how the indicators correlate to one another and provide an overview for a given period that includes all the operational performance components.</p><figure><img alt="Image of the four delivery performance components (DORA metrics) plotted as a graph." src="https://cdn-images-1.medium.com/max/1024/1*T9HxXTw0LtGU7OSF16uTyA.jpeg" /><figcaption><em>Graph showing DORA metrics</em></figcaption></figure><h4>6. Flow monitoring</h4><p>We used the data available to us to establish a set of four indicators with a more direct link to what the teams do and their workflow, and ones that were comparable.</p><p>We tried to make it easy to discern:</p><ul><li>The proportion of the various development tasks the teams do (<strong>Activity split</strong>)</li><li>The chronology of deployments over time (<strong>Deployment chronology</strong>)</li><li>The flow of functional deliveries (<strong>Feature flow</strong>)</li><li>The flow of corrections (<strong>Defect flow</strong>)</li></ul><p>Deployment chronology uses GitHub data that identifies the deployments, whereas the other three indicators (activity split, feature flow and defect flow) are taken from Jira. Each activity type represents one or more ticket types (issueType). These are the tickets counted to calculate the indicators.</p><p>[RESERVE] It can get difficult to figure out the approximation because the unit is a ticket in teams where tickets have a very wide range of granularity in terms of complexity and time spent.</p><h4>Activity split</h4><p>Activity split tracks <strong>the distribution of development activities</strong> divided into three main categories:</p><ul><li><em>Features</em> for functional increments (mostly user stories)</li><li><em>Bugs and problems</em> for things like corrections, support and production incidents</li><li><em>Technical activities</em> for technical tasks like research and design, technical hygiene, refactoring and dependency management</li></ul><p>The idea is to give the team a view of how its work is broken down and set the conditions for team members to see if this breakdown matches their strategy or if it is tolerating this breakdown, i.e., spending too much or not enough time on fixes or tech, etc.</p><figure><img alt="Graphic representation as a bar graph of the number of tickets delivered by period, broken down into three main activity categories: Features, Bugs and problems, Technical activities." src="https://cdn-images-1.medium.com/max/1024/1*uoMYj8PLJAgo5ceUTcCHBw.jpeg" /><figcaption><em>Graphic representation of activity split</em></figcaption></figure><h4>Deployment chronology</h4><p>By showing deployments over time and their lead times for changes, deployment chronology provides a granular view of events in each deployment. This graphic representation makes it possible to distinguish <strong>variations and accidents in the pace of deployments by correlating them with the age of their developments</strong>.</p><figure><img alt="Each deployment is plotted by the date of its lead time for changes. A visual distinction is made between successful deployments (green dots) and failed deployments (red triangles)." src="https://cdn-images-1.medium.com/max/1024/1*BmGJ06O7p3tf7s7G_Ajmtw.jpeg" /><figcaption><em>Graphic representation of deployment chronology</em></figcaption></figure><h4>Feature flow</h4><p>This entailed overseeing feature flows (functional increments) and the relation between the system’s inflows and outflows. It’s a view into <strong>the balance between capacity to do and what’s left to do</strong> (not started and in progress).</p><p>This provides a clear picture of issues related to the pulled flows concept: Is our <a href="https://www.agilealliance.org/glossary/backlog/">backlog</a> too bloated and does it push our flow? Is our in-progress bloated? Or is our backlog drying up? And so on.</p><figure><img alt="Presented as a graph of changes in the amount of features in progress or to be done and finished features." src="https://cdn-images-1.medium.com/max/1024/1*sT49oh5VDljbu-WUgh57FQ.jpeg" /><figcaption><em>Graphic representation of feature flow</em></figcaption></figure><h4>Defect flow</h4><p>Here is where we track corrective activity and the effects of building reliability by plotting <strong>changes in the inventory of anomalies in terms of work on accepted fixes</strong>. Practically speaking, this compares the number of tickets for anomalies (bugs, incidents, etc.) within a time period either left to do or in progress that are on deck at the end of that period with the number of fixes done during the period. In other words, it’s the number of tickets with the same type that have been closed.</p><p>Once again, it’s a matter of flow management since the change in inventory is a reflection of the effects of inflow and the number of completed fixes represents the outflow.</p><p>This view helps teams stay on track setting the right priorities for fixes or building reliability.</p><figure><img alt="Bar graph of changes in the anomalies stock compared to the number of fixed anomalies represented as a curve." src="https://cdn-images-1.medium.com/max/1024/1*kjssxISaHZeLjOiyhB87Nw.jpeg" /><figcaption><em>Graphic representation of defect flow</em></figcaption></figure><h3>Conclusion</h3><p>It is first important to reiterate that <strong>these indicators act as a system</strong>, especially for DORA metrics. They are tightly interwoven. Several indicators will probably change in response to a team’s action or modification.</p><p>Next, <strong>reading metrics only scratches the surface of the search for meaning</strong>. Metrics provide clues, they help assess things like the landscape, events and constraints that will give you outcomes and explanations, and ultimately an improvement roadmap.</p><p>With these disclaimers out of the way, we suggest reviewing the advantages that we observed after setting up these indicators. We detected two main types:</p><ul><li>Expected advantages that we had anticipated.</li><li>Unintended outcomes that are more unexpected yet noteworthy.</li></ul><h4>Expected advantages</h4><p>As we have just explained, the top expected advantages result from a <strong>systematic view</strong> of the situations within teams that the indicators provide us. The serve to simplify:</p><ul><li><strong>The determination of anomalies and trends</strong> that need to be explored.<br>Example: The chart below clearly demonstrates the need to review how much effort is allocated for fixes after a product is initially put into service:</li></ul><figure><img alt="Right after it’s put into service, you can quickly see an increase in detected anomalies as a consequence of user influx and thereby a spike in use cases." src="https://cdn-images-1.medium.com/max/1024/1*iAH6OCnxu1TSFmbJrVHn1Q.jpeg" /><figcaption><em>Example of the visible effect of putting a product into service on its defect flow</em></figcaption></figure><ul><li><strong>Factualizing situations</strong> and <strong>structuring discussions</strong> on areas of improvement.<br>Example: Many teams opted to always open their <a href="https://www.scrum.org/resources/what-is-a-sprint-retrospective">Retrospectives</a> with a group review of the indicators to detect anything outstanding that needs action, discussion or celebration.</li><li><strong>Noting the impact of improvement actions</strong>.<br>Example: The chart below shows the impact of delivering a deployment automation project around 8/20/2021.</li></ul><figure><img alt="Notice that the delivery of a deployment automation project significantly reduces the values for deployment frequency and lead time for changes." src="https://cdn-images-1.medium.com/max/1024/1*a03JM1b5cdCT3ygNwerSrA.jpeg" /><figcaption><em>Graphic representation of the impact of delivering a deployment automation project on the deployment frequency-lead time for changes pair.</em></figcaption></figure><figure><img alt="Notice that the delivery of a deployment automation project significantly increases the number and frequency of deployments." src="https://cdn-images-1.medium.com/max/1024/1*DsRbHGhNREJIrZum_uA-SQ.jpeg" /><figcaption><em>Graphic representation of the impact of delivering a deployment automation project on the number and frequency of deployments.</em></figcaption></figure><figure><img alt="Notice that the delivery of a deployment automation project significantly brings the features to do and finished features curves closer together." src="https://cdn-images-1.medium.com/max/1024/1*4jiiZSW7q2i2YWLNB8zDwQ.jpeg" /><figcaption><em>Graphic representation of the impact of delivering a deployment automation project on feature flow.</em></figcaption></figure><h4>Unintended outcomes</h4><p>The mere process of setting up the reporting system brought situations to the surface and led to changes that trended towards <strong>increasing control and rigor in practices</strong>:</p><ul><li>As we have mentioned, <strong>installing SemVer</strong> for all products was an improvement in itself that helped better control how the codebase in production was arranged.</li><li>In terms of mean time to restore and flow monitoring, we were able to <strong>advance some workflows</strong> because we are basing the resolution concept of releasing a given code to production on the Jira ticket lifecycle data. Teams that were not using it advanced their visual management system and added the missing steps all the way through production. This made it possible to <strong>supplement, clarify and realize the actual state of flows downstream of developments</strong>.</li><li>Similarly, in terms of organizing work and divvying up tasks, setting up a system that counts tickets to monitor flows incentivizes teams to <strong>reduce the degree of granularity among tickets and focus more on the finer points</strong>. This in itself is <a href="https://www.scrum.org/resources/blog/9-reasons-slice-user-stories-better">a best practice that leads to many benefits</a>.</li><li>Lastly, the simple act of shedding light on a few things like defect stock, the number of items in backlogs and the age of some commits (and thereby of some branches) led to <strong>sorting and cleaning actions </strong>that could prove healthy and <strong>enhanced everyday diligence</strong> in this area.</li></ul><h4>Takeaways</h4><p>To conclude, here are two takeaways that we think are important to remember if you want to automate DORA metrics or other indicators in a mixed organization like ours:</p><ul><li>Takeaway 1: Do not skimp on the discovery phase and <strong>take the time to really understand all the different uses in your organization</strong>. That way you can strike a good balance between adaptability (hence complexity) and the prescriptive nature of the anticipated solutions. And keep acceptability and team adoption in your line of sight.</li><li>Takeaway 2: <strong>Tap into the real-life data as soon as possible</strong> to check whether you’ve made the right choices or if your theories match realities on the ground. It goes without saying that you should also get feedback from your first users. Keep in your line of sight improvements for your solution and adapting it to real uses.</li></ul><img src="https://medium.com/_/stat?event=post.clientViewed&referrerSource=full_rss&postId=883a6fe8220b" width="1" height="1" alt="">]]></content:encoded>
        </item>
    </channel>
</rss>