<?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 Anton Abramovich on Medium]]></title>
        <description><![CDATA[Stories by Anton Abramovich on Medium]]></description>
        <link>https://medium.com/@abramovichanton?source=rss-70b94fa17d09------2</link>
        <image>
            <url>https://cdn-images-1.medium.com/fit/c/150/150/1*lbaqrAmPMfiqqu_7XQTSpg.jpeg</url>
            <title>Stories by Anton Abramovich on Medium</title>
            <link>https://medium.com/@abramovichanton?source=rss-70b94fa17d09------2</link>
        </image>
        <generator>Medium</generator>
        <lastBuildDate>Mon, 16 Nov 2026 20:30:13 GMT</lastBuildDate>
        <atom:link href="https://medium.com/@abramovichanton/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[7 Non-Obvious Consequences of AI in UX]]></title>
            <link>https://medium.com/@abramovichanton/7-non-obvious-consequences-of-ai-in-ux-d5ba0d6dd815?source=rss-70b94fa17d09------2</link>
            <guid isPermaLink="false">https://medium.com/p/d5ba0d6dd815</guid>
            <category><![CDATA[ux]]></category>
            <category><![CDATA[content-design]]></category>
            <category><![CDATA[ai]]></category>
            <category><![CDATA[ux-writing]]></category>
            <category><![CDATA[ux-design]]></category>
            <dc:creator><![CDATA[Anton Abramovich]]></dc:creator>
            <pubDate>Sun, 14 Dec 2025 16:05:36 GMT</pubDate>
            <atom:updated>2025-12-14T16:05:36.750Z</atom:updated>
            <content:encoded><![CDATA[<figure><img alt="" src="https://cdn-images-1.medium.com/max/1024/1*C5okhiJP0F8EOFrB-rlwZg.png" /></figure><p>The most important part of integrating a new technology is the effects that show up over time. They’re like radiation: invisible at first, but eventually they change the DNA of your product, your processes, and your team’s habits.</p><p>And the changes aren’t always positive. One day you can wake up and realize your product now looks like all your competitors at once — and your team has forgotten how to think. Here are seven such consequences, through the lens of UX writing.</p><h3>1. AI reduces sensitivity to language</h3><p>When AI “polishes” text, the team stops feeling it. They stop noticing subtle differences in tone across contexts — because the text looks fine. Smooth, correct, sterile. But users don’t feel grammar — they feel emotion.</p><p>At first the changes are hard to spot. The problem surfaces over time, when it starts affecting the user experience.</p><p>Right now, AI tends to make text emotionally flat, while users perceive the same tone differently across scenarios. Using an interface like that is like eating a plastic apple.</p><blockquote><strong><em>How to prevent it</em></strong><em>: read the entire text out loud. If you stumble or catch a “robot intonation,” rewrite it. It also helps to spot-check the model’s favorite words — things like “ensures,” “enables,” “immerse yourself,” “unique” — and add them to a stop list.</em></blockquote><h3>2. A quiet breakdown of tone of voice</h3><p>AI suggests harmless edits: “made it a bit simpler,” “removed repetition.” But the model was trained on the average output of the entire internet — not on your brand book. A brand voice doesn’t break suddenly; it just slowly leaks away, like sand slipping through your fingers.</p><p>A few months later, your product starts sounding… just okay. Like everyone else — not like a brand with its own voice.</p><p><strong>Example</strong>: your product is fintech for high school and university students. You write: “Your money — your rules.” AI suggests: “Manage your finances efficiently.” Seems like the same thing. Product and marketing might even argue the second version describes the feature better. But after six months of edits like this, a youth startup starts sounding like the tax office.</p><blockquote><strong><em>How to prevent it</em></strong><em>: work through demonstration. Don’t upload a pile of guidelines hoping the machine will “figure it out.” Instead, put examples directly into the prompt: the same message in three tones — dry, neutral, and in your product’s voice. Then ask it to write in the style of example #3.</em></blockquote><h3>3. Meaning drift in long flows</h3><p>AI handles short push notifications really well. But if you let it loose in a long flow, chaos begins. AI forgets what it called a button three screens ago, and meaning starts to drift.</p><p><strong>Example</strong>: on the first screen the product asks the user to fill the “Cart,” on the third screen AI calls it “Your order,” and in checkout it suddenly becomes the “Shopping list.”</p><p>This might be a term swap “for variety,” even though what matters is consistent, easy navigation. Or a slightly shifted tone. The result: navigation becomes harder, cognitive load increases.</p><blockquote><strong><em>How to prevent it</em></strong><em>: don’t generate copy screen by screen. Feed the whole scenario at once. Also create a glossary and give it to the model with a strict instruction: “To refer to the purchase, use only the word ‘Cart.’ Any synonyms are forbidden.”</em></blockquote><h3>4. Loss of the team’s contextual memory</h3><p>AI helps you move faster, which means you can rely on it without articulating for yourself — or for the team — why a particular decision was made.</p><p>— Why did we set a 140-character limit here?<br> — No idea. GPT suggested it, so we used it.</p><p>The model suggested it, it must know better. A few months later, no one remembers why the limit exists, why that term was chosen, what the error wording depended on. Product history turns into a string of random actions.</p><blockquote><strong><em>How to prevent it</em></strong><em>: don’t paste AI outputs into mockups or production without commentary — and don’t allow the team to do it either. If a model proposed the copy or solution, the owner must write in the task tracker or in Figma comments: “We chose this version because it’s shorter and reduces the user’s anxiety about payment.” If you can’t explain the choice in words, you can’t use it.</em></blockquote><h3>5. Excessive AI “optimization” increases users’ cognitive load</h3><p>AI loves shortening things. For it, brevity is the soul of wit — but in interfaces it can be the soul of misunderstanding. The model doesn’t understand the “weight” of words and removes what looks like filler, even if it’s actually a user’s lifeline. Even if it sounds a bit rough.</p><p><strong>Example</strong>: a modal says, “Enter the code from the SMS we sent to +44…,” and AI happily shortens it to “Enter the SMS code.” But the user may not have logged in for a long time, may have changed their number, or may have registered the account under someone else’s phone. Now they’re staring at their phone, waiting for a message, slowly boiling inside.</p><blockquote><strong><em>How to prevent it</em></strong><em>: add Safety Constraints to your prompts. For example: “Cut this text in half, but you must keep the mention that the code will be sent to email, not via SMS. If that can’t be preserved, don’t shorten it.” Use hallway testing as validation for decisions like this.</em></blockquote><h3>6. The illusion that “content has become easier”</h3><p>AI creates the impression that deep expertise in specific domains is no longer needed — that you can work faster and reduce certain roles. Why hire a UX writer if Claude can produce copy in a second?</p><p>But with that approach, problems get solved superficially and flow conflicts start popping up. It becomes important to communicate up the chain the value of humans in places where AI will never be able to replace them.</p><p><strong>Example</strong>: AI generates a legal disclaimer. It sounds convincing, serious. But when a lawyer reads it, they go grey — because legally the text promises the user half the company’s shares as a gift. Without a human expert, content generation speed turns into problem generation speed.</p><blockquote><strong><em>How to prevent it</em></strong><em>: periodically show stakeholders what the model produces. For example, one slide with hallucinations, style blunders, legally risky wording — and another slide with what the UX writer actually shipped to production.</em></blockquote><h3>7. Erosion of expertise</h3><p>AI is like an exoskeleton: wear it long enough and your own muscles atrophy. It speeds specialists up, but in exchange it slowly weakens independent thinking. If many people use it this way, the team’s overall level starts to drop too. It can show up like this: fewer questions, fewer debates, fewer corner cases worked through. Ideally, AI shouldn’t be used at the start — it should expand what humans have already thought through first.</p><blockquote><strong><em>How to prevent it</em></strong><em>: don’t start work by opening ChatGPT. First draft a structure or ideas on paper or in notes. Formulate the problem. Only then give it to the model and ask it to critique, expand the idea list, or suggest alternative wording. You can take the best parts, edit them, and validate them. In this setup, AI plays the role of an opponent — not the author.</em></blockquote><p>All these risks aren’t a reason to reject technology — they’re a reason to rethink how you use it. AI is great at routine work, but it still can’t replace strategy, empathy, or deep contextual understanding.</p><p>The main challenge for teams right now is learning to use neural networks as an amplifier of professional skills — not as a crutch that replaces critical thinking.</p><p>Ultimately, the value of a UX specialist shifts from the ability to simply “write text” or “draw a screen” to the ability to validate it and hold the bigger picture in mind. The more accessible generative tools become, the higher the bar for the outcome should be. Let AI propose options — but the final decision, responsibility, and protection of the user’s interests must always remain with a human.</p><p>🔗 Feel free to connect me in <a href="https://www.linkedin.com/in/anton-abramovich">LinkedIn</a></p><img src="https://medium.com/_/stat?event=post.clientViewed&referrerSource=full_rss&postId=d5ba0d6dd815" width="1" height="1" alt="">]]></content:encoded>
        </item>
        <item>
            <title><![CDATA[Burnout in Remote Work: How It Shows Up and How to Fight It]]></title>
            <link>https://medium.com/@abramovichanton/burnout-in-remote-work-how-it-shows-up-and-how-to-fight-it-ef3fa30f814d?source=rss-70b94fa17d09------2</link>
            <guid isPermaLink="false">https://medium.com/p/ef3fa30f814d</guid>
            <category><![CDATA[design]]></category>
            <category><![CDATA[team-building]]></category>
            <category><![CDATA[burnout]]></category>
            <category><![CDATA[management]]></category>
            <category><![CDATA[teamwork]]></category>
            <dc:creator><![CDATA[Anton Abramovich]]></dc:creator>
            <pubDate>Fri, 05 Dec 2025 15:19:34 GMT</pubDate>
            <atom:updated>2025-12-05T15:19:34.463Z</atom:updated>
            <content:encoded><![CDATA[<figure><img alt="" src="https://cdn-images-1.medium.com/max/1024/1*dKaU9PqsLkMpKnknJZBMHA.png" /></figure><p>This will be a long conversation — but an important one. Over the past five years, the topic, unfortunately, hasn’t lost relevance; it has only intensified. So let’s proceed from the assumption that fighting burnout is the new normal. Let’s begin.</p><p>Not long ago, burnout was considered the employee’s problem — “learn to say no,” do yoga, breathe in a square.</p><p>Now, however, there’s research showing that treating burnout with personal “remedies” leads nowhere. The WHO has officially linked burnout to the work environment, shifting responsibility for addressing it from the individual to the company — calling it “a phenomenon exclusively related to the professional context.”</p><p>This means the responsibility lies with the organization, not the employee. <strong>It’s not the canary’s fault that it was pulled out of the mine half-alive.</strong></p><figure><img alt="" src="https://cdn-images-1.medium.com/max/1024/1*0vrlY5B41-Ap1FqkXClI8Q.png" /></figure><p>Burnout is a syndrome characterized by three core symptoms:</p><ul><li>Emotional exhaustion</li><li>Cynicism (or detachment)</li><li>A sense of personal inefficacy (reduced professional accomplishment)</li></ul><p>Gallup analysts conducted a large study and identified five key factors that directly correlate with burnout:</p><ul><li>Unfair treatment at work (bias, favoritism, lack of transparency)</li><li>Unmanageable workload that exceeds capacity</li><li>Blurred role boundaries and unclear expectations</li><li>Lack of support and communication with one’s manager</li><li>Unrealistic deadlines and excessive time pressure</li></ul><p><strong>In distributed teams, each of these factors intensifies.</strong> Why?</p><p>A manager doesn’t see a colleague’s dimmed expression over a coffee break. Without regular direct contact, overall opacity grows, quickly turning into poor task allocation and lack of support when a person is left alone with a problem.</p><p>If a team lead doesn’t notice or ignores warning signs, burnout becomes a management failure — and beyond harming the individual, it risks losses for the organization, from a decline in the quality of design decisions to the loss of valuable specialists. According to McKinsey, burnout is tightly linked to attrition: a toxic work climate is the strongest predictor of intention to leave, and employees exposed to high toxicity are <strong>8 times more likely</strong> to experience burnout symptoms and consider quitting.</p><figure><img alt="" src="https://cdn-images-1.medium.com/max/1024/1*jhzMUozkMDeAFGqM2e_a7g.png" /></figure><p>Burnout doesn’t happen instantly. It’s a slow, steady erosion that accumulates over time. Managers need to know which stage an employee is in to intervene in time. You can rely on a simplified five-stage model often used in research.</p><h3><strong>1. Honeymoon</strong></h3><p>The employee is full of enthusiasm and inspiration, ready to take on more than they can carry. This is a dangerous moment for a manager: it’s tempting to encourage this behavior to showcase high performance, but that’s how work habits that lead to overwork are formed.</p><p><strong>Signs:</strong> the person readily takes on tasks beyond their normal scope.</p><p><strong>Manager’s task:</strong> help set priorities and closely monitor workload so the “honeymoon” doesn’t turn into chronic stress.</p><h3><strong>2. Accumulating stress</strong></h3><p>Enthusiasm fades and productivity starts to dip. Irritability and anxiety appear, focus becomes an issue. Feedback that isn’t constructive — or extra rounds of iteration that were previously ignored — can now ruin the day.</p><p><strong>Signs:</strong> the employee is quieter on calls, proposes fewer ideas, occasionally misses internal deadlines.</p><p><strong>Manager’s task:</strong> offer active support — revisit timelines, clarify expectations, so stress doesn’t become chronic.</p><h3><strong>3. Chronic stress</strong></h3><p>Fatigue is felt almost daily. At this stage, cynical or apathetic attitudes toward work may emerge.</p><p><strong>Signs:</strong></p><ul><li>The specialist stops “caring deeply” about the work, suggests easy “top-shelf” solutions</li><li>May drift into social isolation: avoids team chats or calls. In remote settings, this is easy to miss — the person is technically present but emotionally absent</li><li>Switches to denial mode, insisting “it’s just a lot of work right now”</li></ul><p><strong>Manager’s task:</strong> at this point, act proactively and autonomously (without waiting for the employee’s buy-in) — identify and address the “pain point” (process, task, colleague).</p><h3><strong>4. Burnout</strong></h3><p>Symptoms become critical, and the quality and timeliness of work plummet.</p><p><strong>Signs:</strong></p><ul><li>Pessimism and cynicism dominate</li><li>The employee may “disappear” for an hour or two or stop responding on time</li><li>Colleagues start coming to you with questions about their behavior</li></ul><p><strong>Manager’s task:</strong> insist on a full break — a long vacation or sabbatical, with all work tasks taken off their plate.</p><h3><strong>5. Chronic burnout</strong></h3><p>This is the worst-case scenario, often leading to clinical depression and psychosomatic issues. The employee is essentially lost to the team: initiative and creativity vanish, cognitive abilities decline.</p><p><strong>Manager’s task:</strong> at this point, there’s nothing you can do within the team. The only option is prevention long before this stage. If it’s reached, the person likely needs medical professional support.</p><figure><img alt="" src="https://cdn-images-1.medium.com/max/1024/1*Hd8f8rtCJFYlRm2JI40JcA.png" /></figure><p>To move from “managerial intuition” to systematic management of the problem, use proven frameworks.</p><h3><strong>Questionnaires</strong></h3><p><strong>Maslach Burnout Inventory (MBI)</strong> — the most reliable, globally recognized tool for measuring burnout. It evaluates the severity of the three key dimensions:</p><ol><li><strong>Emotional exhaustion</strong></li><li><strong>Cynicism</strong> (depersonalization) — indifference or distancing toward work and clients</li><li><strong>Reduced professional efficacy</strong> — negative self-assessment of one’s effectiveness</li></ol><p><strong>What this gives you:</strong> MBI helps you see the problem as systemic rather than individual — for example, if 80% of your team shows high exhaustion. You can also use simplified tools like the <strong>Copenhagen Burnout Inventory (CBI)</strong> or <strong>Oldenburg Burnout Inventory (OLBI)</strong> for regular “thermometer checks” of team well-being.</p><h3><strong>The “Six Areas of Worklife” model</strong></h3><p>To understand <em>why</em> burnout arises, use the model by Christina Maslach and Michael Leiter, which identifies six key mismatches between the person and the work environment:</p><ul><li><strong>Workload:</strong> too much work</li><li><strong>Control:</strong> lack of autonomy and influence over decisions</li><li><strong>Reward:</strong> lack of recognition and fair compensation</li><li><strong>Community:</strong> poor team relationships or a toxic atmosphere</li><li><strong>Fairness:</strong> a sense of injustice (e.g., favoritism)</li><li><strong>Values:</strong> misalignment between personal and corporate values (e.g., sacrificing UX quality for short-term gains)</li></ul><p><strong>This model is useful because it directs efforts toward concrete systemic changes.</strong> Audit each of the six categories: survey the team to see where it “hurts” most. Then address the issues found — that’s core prevention.</p><p><strong>Important:</strong> proactive steps (e.g., granting autonomy and flexibility) can make employees <strong>70% less likely</strong> to experience high burnout.</p><p>And by regularly gauging team climate, you can turn burnout risk from an unexpected crisis into a manageable, adjustable metric.</p><figure><img alt="" src="https://cdn-images-1.medium.com/max/1024/1*Ymzqrjcq_AE2jbLZ5Od7fQ.png" /></figure><p>Once you’ve identified the stage, move to action. Here’s your toolkit.</p><h3>Prevention</h3><p>The ideal scenario is eliminating burnout precursors early. In remote work, this requires deliberately building a culture that encourages balance and well-being.</p><p><strong>When to apply:</strong> before the honeymoon stage.</p><p><strong>What to do:</strong></p><ul><li><strong>Encourage balance.</strong> Offer flexible schedules, recovery time (e.g., “meeting-free Fridays”), and cultivate open dialogue about workload</li><li><strong>Train managers.</strong> Gallup’s research shows direct managers have the strongest influence on burnout levels. Managers must be able to detect and remove stressors</li><li><strong>Set “quiet hours.”</strong> In distributed teams across time zones, define quiet hours after 6 p.m. local time so no one feels compelled to stay online</li><li><strong>Give autonomy.</strong> Autonomy is key against micromanagement and burnout. Set clear expectations but let people control <em>how</em> they work — schedule, location, methods</li><li><strong>Maintain alignment.</strong> Record meetings, share summaries, and attach them to tickets or chats so teammates who miss meetings (time zones!) don’t accumulate anxiety</li></ul><p><strong>Pros of prevention:</strong></p><ul><li>Sustains high productivity and motivation</li><li>Improves organizational reputation: trust and engagement grow</li><li>Cheaper and easier to implement (plan adjustments, process tweaks) than treating a burnt-out employee</li><li>Universally beneficial and efficiency-boosting</li></ul><p><strong>Cons of prevention:</strong></p><ul><li>Requires ongoing attention and manager time; one training won’t fix it forever</li><li>Hard to convince leadership to invest before a crisis hits</li><li>External shocks (layoffs, crunches) can raise stress regardless of internal prevention</li></ul><h3>Restructuring</h3><p>If warning signs are already present, change internal processes to remove stressors.</p><p><strong>When to apply:</strong> during the honeymoon and accumulating stress stages.</p><p><strong>What to do:</strong></p><ul><li><strong>Rebalance workload:</strong> redistribute projects or hire a temporary specialist to reduce acute pressure</li><li><strong>Clarify expectations:</strong> only <strong>60%</strong> of workers feel sure they know what’s expected. Clarity calms. Also, set expectations upward — it’s okay to tell stakeholders about burnout risk when they bring too much work</li><li><strong>Optimize processes:</strong> for high-stakes, time-sensitive projects, add weekly alignment calls and “freeze” changing requirements after a set deadline if stakeholders keep bringing edits</li></ul><p><strong>Pros of restructuring:</strong></p><ul><li>Targets root causes instead of symptoms</li><li>Can deliver quick wins — people see real changes and feel supported</li><li>Retains the team and typically raises overall effectiveness</li><li>Builds a culture of continuous improvement and trust in leadership</li></ul><p><strong>Cons of restructuring:</strong></p><ul><li>Requires resources and time: changes may temporarily slow productivity</li><li>Doesn’t always guarantee individual recovery: serious burnout won’t vanish immediately</li><li>May face organizational resistance (budget, cross-team changes)</li><li>Not a cure-all: if the problem is deeply personal (loss of interest in the profession), better conditions won’t be enough</li></ul><h3>Rotation or changing tasks</h3><p>UX burnout often stems from monotony. Rotation can restore a sense of growth and motivation.</p><p><strong>When to apply:</strong> during accumulating and chronic stress.</p><p><strong>What to do:</strong></p><ul><li><strong>Temporarily switch routine tasks.</strong> For example, a UX designer might spend a couple of sprints on UX research or UI animation if interested; a researcher could join product-strategy brainstorming</li><li><strong>Assign a different project.</strong> If the org structure allows, temporarily move the person to an adjacent project with a different team and task set</li></ul><p>Crucially, frame this as an opportunity, not a punishment: “Let’s relieve some routine and give you something new.” Remote makes rotation easier — “moving” to another team is a couple of calls, no physical relocation.</p><p><strong>Pros of rotation:</strong></p><ul><li>Injects novelty, breaking the Groundhog Day feel</li><li>Supports skill development and broadens perspective, boosting confidence</li><li>Can reveal talents and help someone find a better-fit role</li><li>Flexible in duration — temporary or permanent</li></ul><p><strong>Cons of rotation:</strong></p><ul><li>Doesn’t eliminate root causes. Systemic issues will return if unaddressed</li><li>Risk of lost focus in projects, especially without proper backfill</li><li>Not for everyone: those who value stability may find rotation stressful</li><li>If constant, can miss expectations in the new role and add frustration</li></ul><h3>A pause</h3><p>In remote work, where office and home blur, a <em>real</em> vacation is critical — otherwise a person is physically away but mentally still stewing in work, checking email and chats. A manager should help organize a proper break: temporarily redistribute responsibilities, assure the person the world won’t collapse without them for a couple weeks. This could be an unscheduled paid leave or a sabbatical.</p><p>Studies show that <em>short-term rest can significantly reduce burnout symptoms</em>, though it’s not a radical cure.</p><p>In severe cases of exhaustion, give time to rebuild — not “return to normal,” but build a <em>new</em> normal that doesn’t require ignoring the body’s signals.</p><p><strong>When to apply:</strong> during chronic stress and burnout.</p><p><strong>What’s needed for recovery:</strong></p><ul><li>Proper rest and nutrition</li><li>More sleep and reduced work intensity</li><li>Permission to pause without guilt</li><li>Gentle physical activity</li></ul><p><strong>Pros of a pause:</strong></p><ul><li>Immediate relief: rest quickly reduces acute stress and emotional overheating</li><li>Prevents errors and further damage to the product that could be costly</li><li>Signals support: the manager values the person over short-term gain, boosting loyalty</li><li>Gives leadership time to fix systemic issues so the employee returns to improved conditions</li></ul><p><strong>Cons of a pause:</strong></p><ul><li>Doesn’t tackle root causes: if the chaos remains, burnout will recur — pair the break with restructuring</li><li>Load shifts to others: one person’s leave can stress the rest of the team</li><li>Risks to timelines and clients: cover gaps and communicate constraints to stakeholders</li><li>The employee may struggle to disconnect unless required to turn off notifications</li></ul><p><strong>Critically important:</strong> if someone goes on leave, they should fully disconnect from work.</p><p>And remember: if they return to the same chaos, burnout will repeat. Pair the pause with internal changes you implement while they rest.</p><h3>Leaving the team</h3><p>This is a last resort when attempts to restore the person within the current team have failed, or it’s clear the role is no longer a fit.</p><p>We’re talking chronic burnout here. Severe burnout isn’t a sign of weakness — it’s a sign someone has been strong for too long without enough support.</p><p>Chronic mental and physical fatigue may lead someone to stop working for a time to recover. Professional help (e.g., a psychologist) may be needed.</p><p>Since burnt-out employees are <strong>2.6 times</strong> more likely to resign, managers should respectfully support a decision to change career paths if the previous environment — or the profession itself — has become toxic. Exits can be structured as an internal transfer to a different department or a separation from the company.</p><p>Recovery from deep exhaustion takes time and may require a therapist to develop a plan aligned with the person’s limits and goals.</p><p><strong>Pros of leaving the team:</strong></p><ul><li>Chance for full recovery in a different environment. A new team or job can shed negative associations and offer a fresh start</li><li>Reduced team toxicity. If burnout reached cynicism, that person can drag others down; removing this factor protects the team from “emotional contagion”</li><li>Focus on those ready to work. Leadership can invest in developing others rather than “resuscitating” someone set on leaving</li><li>With an internal transfer, the company retains a valuable specialist in a different role</li></ul><p><strong>Cons of leaving the team:</strong></p><ul><li>Loss of experience and knowledge. The departure of a seasoned UX professional, especially with deep product context, is a loss of capability</li><li>Organizational warning sign. A burnt-out exit signals to HR and leadership that conditions aren’t optimal</li><li>Emotionally hard on the team. Colleagues may feel guilt or fear; it’s important to discuss reasons and emphasize steps being taken to improve conditions</li><li>Formal hurdles. Burnout per se isn’t a legal basis for termination (outside a medical diagnosis), which may require negotiations and compensation; finding an equivalent replacement quickly isn’t guaranteed</li></ul><p>When weighing an exit, a manager must be guided by both business logic and empathy. If all internal attempts fail, it’s better to support the person’s decision to change course than to keep them at the cost of their health.</p><figure><img alt="" src="https://cdn-images-1.medium.com/max/1024/1*_T0gWP-q5PxZGs8pVPMygQ.png" /></figure><ol><li><strong>Burnout is a professional phenomenon</strong> tied to chronic stress, and <strong>responsibility for it lies with the organization,</strong> not the individual.</li><li><strong>Symptoms:</strong> emotional exhaustion, cynicism, and a sense of inefficacy — developing gradually across five stages.</li><li><strong>Remote work increases risks</strong> by raising opacity and a sense of isolation.</li><li><strong>Core causes are managerial:</strong> unrealistic deadlines, excessive load, unclear roles, and lack of manager support.</li><li><strong>Diagnosis must be systemic:</strong> use tools like the Maslach Burnout Inventory (MBI) or the “Six Areas of Worklife” model to identify root causes in the work environment.</li><li><strong>Management must be proactive:</strong> invest in <strong>prevention,</strong> <strong>restructure processes,</strong> use <strong>rotation</strong> to renew motivation, and when necessary, insist on a <strong>full pause.</strong></li></ol><p>If you recognize your team in these symptoms, don’t wait for chronic burnout. Schedule a 1:1 with your key specialist right now and start with what’s in your control: <em>“What can I, as your manager, change in our processes to give you more control and transparency?”</em></p><figure><img alt="" src="https://cdn-images-1.medium.com/max/1024/1*ncB1hdlDK2h6xwMWlBeV_A.png" /></figure><p>And here’s a checklist for individual contributors. Use it for yourself or to gauge a new team’s climate:</p><ul><li>You feel <strong>tired almost every day;</strong> sleep doesn’t help</li><li>You’re <strong>emotionally drained</strong> and can’t give of yourself psychologically as before</li><li>You have <strong>trouble focusing;</strong> small things that once went unnoticed now irritate you</li><li><strong>Daily routines break down;</strong> you stay up late trying to “catch up”</li><li><strong>Time off doesn’t fix it</strong></li><li>You’ve become <strong>apathetic or cynical</strong> about your work</li><li>You’ve <strong>stopped fighting</strong> for UX quality and are running on autopilot</li><li>You <strong>avoid team chats</strong> or calls, withdrawing into social isolation</li><li>You feel <strong>indifferent, negative, or even sarcastic</strong> toward colleagues, clients, or the project</li><li>Work feels like <strong>meaningless routine</strong></li><li>You feel you’re <strong>losing confidence</strong> in your abilities</li><li>You <strong>miss internal deadlines more often</strong></li><li><strong>Your quality drops sharply,</strong> or you make more mistakes</li><li>You <strong>think about quitting</strong> more frequently</li><li>You <strong>switch to denial</strong> and brush off help</li><li>You feel <strong>no longer able to make a positive impact</strong> on people or the product</li></ul><p>If you spot 3–4 patterns that keep repeating, try to address them before the problem grows beyond easy control.</p><p>Take care of yourselves, value what you do, and hydrate more. Peace to all 🖖</p><p>🔗 Feel free to connect me in <a href="https://www.linkedin.com/in/anton-abramovich">LinkedIn</a></p><img src="https://medium.com/_/stat?event=post.clientViewed&referrerSource=full_rss&postId=ef3fa30f814d" width="1" height="1" alt="">]]></content:encoded>
        </item>
        <item>
            <title><![CDATA[9 Libraries for UX/UI Research and Inspiration]]></title>
            <link>https://medium.com/@abramovichanton/9-libraries-for-ux-ui-research-and-inspiration-4fa8ce1c5f78?source=rss-70b94fa17d09------2</link>
            <guid isPermaLink="false">https://medium.com/p/4fa8ce1c5f78</guid>
            <category><![CDATA[ux]]></category>
            <category><![CDATA[apps]]></category>
            <category><![CDATA[product-design]]></category>
            <category><![CDATA[product-research]]></category>
            <category><![CDATA[research]]></category>
            <dc:creator><![CDATA[Anton Abramovich]]></dc:creator>
            <pubDate>Thu, 27 Nov 2025 10:53:55 GMT</pubDate>
            <atom:updated>2025-11-27T10:53:55.293Z</atom:updated>
            <content:encoded><![CDATA[<figure><img alt="" src="https://cdn-images-1.medium.com/max/1024/1*FBUkmBHAKcXSMqytU9fn8g.png" /></figure><p>I’ve often heard the bias that using interface libraries is “wrong”: as if you’re copying others’ solutions, losing originality, and risking conflicts with your own product.</p><p>It seems that won’t happen if you work with references like this:</p><ol><li>Decompose the task — not “we need a beautiful user profile,” but “we need a solution for fitting security settings, personal data, and order history onto one screen without it feeling overloaded.”</li><li>Look for relevant products — Dribbble and Behance are great for visuals, but when you need to find a working product you don’t yet know about, libraries are the perfect tool.</li><li>Diversify sources — analyze direct competitors (a must), indirect ones (e.g., from other countries, highly recommended), and even non-competitors (excellent for building your internal library of UX solutions).</li></ol><p>And don’t forget Jakob’s Law:</p><blockquote>Users spend most of their time on sites that aren’t yours, so they expect familiar interface patterns.</blockquote><p>I’ve collected a stack of libraries that still make my work much easier today:</p><h3>Apps &amp; Web</h3><ul><li><a href="https://mobbin.com/">Mobbin</a> — the most popular one; it simply has to be on this list.</li><li><a href="https://refero.design/">Refero</a> — by their account, the largest library of web and iOS screens.</li><li><a href="https://pageflows.com/">Page Flows</a></li></ul><h3>Web</h3><ul><li><a href="https://www.saasframe.io/">SaaSframe</a> — a category-structured archive.</li></ul><h3>Russian Apps</h3><ul><li><a href="https://scrn.gallery/apps">SCRN</a> — interesting feature patterns section: a single page gathers screens with the same purpose from different apps.</li><li><a href="https://benchmarkee.ru/">Benchmark</a></li><li><a href="https://liba.tilda.ws/library">Liba</a></li><li><a href="https://screenoteka.com/">Screenoteka</a></li><li><a href="https://screenbook.ru/?os=iOS">Screen Book</a></li></ul><h3>Bonus</h3><ul><li><a href="https://dsgn.zone/">Dsgn.zone</a> — an archive of design resources: design systems, links to icon, font, illustration, and inspiration sites.</li><li><a href="http://scrnshts.club/">Scrnshts</a> — promo screenshots published in the app stores.</li></ul><p>Pro tip: these services work great as training grounds. Open a screen from a product that’s close to you and think through how the product team arrived at that solution — reverse-engineer it all the way down to the underlying job to be done.</p><img src="https://medium.com/_/stat?event=post.clientViewed&referrerSource=full_rss&postId=4fa8ce1c5f78" width="1" height="1" alt="">]]></content:encoded>
        </item>
        <item>
            <title><![CDATA[Toxic or Comfort Zone: 16 Questions to Find Out Who You’ll Be Working With]]></title>
            <link>https://medium.com/@abramovichanton/toxic-or-comfort-zone-16-questions-to-find-out-who-youll-be-working-with-28ba699a7cb8?source=rss-70b94fa17d09------2</link>
            <guid isPermaLink="false">https://medium.com/p/28ba699a7cb8</guid>
            <category><![CDATA[management]]></category>
            <category><![CDATA[team]]></category>
            <category><![CDATA[hiring]]></category>
            <category><![CDATA[communication]]></category>
            <category><![CDATA[interview]]></category>
            <dc:creator><![CDATA[Anton Abramovich]]></dc:creator>
            <pubDate>Sat, 22 Nov 2025 12:45:57 GMT</pubDate>
            <atom:updated>2025-12-06T11:11:37.693Z</atom:updated>
            <content:encoded><![CDATA[<figure><img alt="" src="https://cdn-images-1.medium.com/max/1024/1*wxOBKwhJORqSRQrNNwUmnQ.png" /></figure><p>The job platform <strong>Glassdoor</strong> surveyed its users and compiled <strong>16 questions</strong> worth asking a potential manager to understand the team’s atmosphere.</p><p>After all, an interview isn’t an exam — it’s a <em>mutual check for sanity.</em></p><h3>Leadership and Management Style</h3><ul><li>How would you describe your leadership style?</li><li>How do you communicate with your team? How do you set expectations and track progress?</li><li>How do you typically give feedback?</li></ul><h3>Handling Mistakes and Difficult Situations</h3><ul><li>How do you respond when someone on your team makes a mistake?</li><li>What do you do if a deadline is missed?</li><li>When projects don’t go as planned, how do you deal with it?</li><li>How do you support people when they’re overwhelmed?</li></ul><h3>Growth Mindset and Self-Reflection</h3><ul><li>How do you identify your own areas for growth?</li><li>What has been your biggest failure?</li><li>Do you want your team to challenge you?</li></ul><h3>Work–Life Balance</h3><ul><li>How often does your team have to stay late?</li><li>What does your team usually do on a Friday at 6 p.m.?</li><li>How do you help your team maintain work–life balance?</li></ul><h3>Team Culture</h3><ul><li>On average, how long do people stay on your team?</li><li>What qualities make someone successful in your team?</li><li>What do you enjoy most about managing people?</li></ul><p>These questions reveal a wide range of important aspects about your potential workplace:</p><ul><li>how involved the manager is in the team’s daily routine and whether their approach aligns with yours</li><li>how the team handles stressful periods</li><li>whether the manager treats the team as real people or expects flawless performance at all times</li><li>whether the real team culture matches the job description</li><li>how friendly and supportive the overall environment is</li></ul><p>And, of course, these questions are <em>smartly framed.</em> They’re open-ended — and some of them subtly invite the manager to reveal toxic behaviour. For example, <em>“How do you identify your areas for growth?”</em> puts the spotlight on self-awareness. Or <em>“How do you support people when they’re overwhelmed?”</em> quietly assumes that overwork happens — often as a result of ineffective management.</p><p>The only caveat: not every manager will have an immediate answer — and that’s fine. Not everyone becomes a leader of that level right away, able to respond thoughtfully without prior reflection. In that case, a healthy, non-toxic response might sound like: <em>“That’s an interesting question — I’ve never really thought about it before. I don’t have a ready answer right now.”</em> From there, depending on how much time you have in the interview, they can either think aloud about it — or you can move on.</p><p>Red flags in these conversations tend to speak for themselves.</p><p>Here are a few worth paying extra attention to:</p><ul><li>visible signs of irritation (closed posture, microexpressions, weary sighs)</li><li>bragging</li><li>vague or overly abstract answers</li><li>changing the subject</li><li>refusing to answer</li><li>cutting the interview short</li></ul><blockquote>And most importantly — <strong>trust your intuition.</strong></blockquote><img src="https://medium.com/_/stat?event=post.clientViewed&referrerSource=full_rss&postId=28ba699a7cb8" width="1" height="1" alt="">]]></content:encoded>
        </item>
        <item>
            <title><![CDATA[Where Should the UX Writing Team Live — in Design, Content, or on Its Own?]]></title>
            <link>https://medium.com/@abramovichanton/where-should-the-ux-writing-team-live-in-design-content-or-on-its-own-fc937326741a?source=rss-70b94fa17d09------2</link>
            <guid isPermaLink="false">https://medium.com/p/fc937326741a</guid>
            <category><![CDATA[product-design]]></category>
            <category><![CDATA[ux]]></category>
            <category><![CDATA[ux-design]]></category>
            <category><![CDATA[team]]></category>
            <category><![CDATA[ux-writing]]></category>
            <dc:creator><![CDATA[Anton Abramovich]]></dc:creator>
            <pubDate>Sat, 15 Nov 2025 13:23:54 GMT</pubDate>
            <atom:updated>2025-11-15T13:23:54.573Z</atom:updated>
            <content:encoded><![CDATA[<h3>Where Should the UX Writing Team Live — in Design, Content, or on Its Own?</h3><figure><img alt="" src="https://cdn-images-1.medium.com/max/1024/1*gw_bSnK2e_KSZiUYglSB0g.png" /></figure><p>Every time a company decides where to “place” UX writers, somewhere a content lead sighs quietly — and a Head of Design argues loudly. It’s understandable: writers are like cats — they love everyone but aren’t ready to live with anyone outside their own pride.</p><p>Still, they have to sit <em>somewhere.</em></p><figure><img alt="" src="https://cdn-images-1.medium.com/max/1024/1*vdP9PSvCAefiC3n2C76DsQ.png" /></figure><p>Seems like a win-win: one pipeline, shared planning, faster reviews. Writers work side by side with designers, decisions are made on the fly, and copy makes it into mockups on time.</p><p>But speed has a side effect — <strong>depth gets lost.</strong> Instead of meaning, people start discussing the visual weight of a headline. The writer becomes “the person who fits words into the button width,” not the one defining the user logic behind it.</p><p>It’s like sitting a translator next to a designer and asking them to “make the interface sound a bit prettier.”</p><figure><img alt="" src="https://cdn-images-1.medium.com/max/1024/1*aCJIvi9WsEad5-YC1K7F1g.png" /></figure><p>Here everything’s neat: guidelines, style culture, even a touch of academia. Finally, the UX team stops writing by instinct — a unified language appears, with consistency, system, and tone that doesn’t jump from <em>“Hey there!”</em> to <em>“Dear user.”</em></p><p>Almost too good. The downside? You pay for that order with speed. Decisions get over-discussed and over-iterated until they lose momentum — and designers start quietly tweaking text in mockups because they “can’t wait for content.” UX turns into bureaucracy with beautiful wording.</p><figure><img alt="" src="https://cdn-images-1.medium.com/max/1024/1*qwiDyfAjY96M30SPbD1eCw.png" /></figure><p>Now that’s the most ambitious scenario. Here, UX writing exists as a horizontal discipline: it builds its own guidelines and architecture, mentors designers, co-creates flows, and directly shapes product strategy. Sounds great — like a dream.</p><p>And such teams <em>do</em> exist — but only where the company truly understands the value of writers as a strong, autonomous unit. Otherwise, the team either dissolves between departments or drowns in endless cross-functional threads.</p><p>There’s also the scenario where one writer serves the entire design department — but let’s not talk about that. Let’s just… sympathize.</p><h3>So, where <em>should</em> a UX writing team live?</h3><p>It doesn’t matter. What matters is that it has the right to ask not “where do we place the text on this screen?” but <strong>“why does the user even need this?”</strong></p><p>Structure and hierarchy don’t matter when the writer is involved early in the interface-thinking process. That’s where trust in their expertise grows — and that trust, in turn, lays the foundation for future writers.</p><p>A UX writing team is not a department or a service. It’s a way of <em>thinking</em> about the product through language. While companies debate where to put writers, users don’t care — they just want to be <strong>understood the first time.</strong></p><p>If you draw a parallel, design is like speech — there’s spoken and written, two perspectives on the same thing, each with its own rules and principles. The same goes for visuals and text.</p><blockquote>So the real question isn’t <em>“where should UX writing live?”</em> It’s <em>“where can UX writing be most useful?”</em></blockquote><img src="https://medium.com/_/stat?event=post.clientViewed&referrerSource=full_rss&postId=fc937326741a" width="1" height="1" alt="">]]></content:encoded>
        </item>
        <item>
            <title><![CDATA[Metrics for UX Writers: Why They’re Hard to Measure — and Which Ones Actually Matter]]></title>
            <link>https://medium.com/@abramovichanton/metrics-for-ux-writers-why-theyre-hard-to-measure-and-which-ones-actually-matter-f4754a1d5e77?source=rss-70b94fa17d09------2</link>
            <guid isPermaLink="false">https://medium.com/p/f4754a1d5e77</guid>
            <category><![CDATA[analytics]]></category>
            <category><![CDATA[product-design]]></category>
            <category><![CDATA[metrics]]></category>
            <category><![CDATA[ux]]></category>
            <category><![CDATA[ux-writing]]></category>
            <dc:creator><![CDATA[Anton Abramovich]]></dc:creator>
            <pubDate>Sat, 15 Nov 2025 12:04:53 GMT</pubDate>
            <atom:updated>2025-11-15T12:04:53.749Z</atom:updated>
            <content:encoded><![CDATA[<h3>Metrics for UX Writers: Why They’re Hard to Measure — and Which Ones Actually Matter</h3><figure><img alt="" src="https://cdn-images-1.medium.com/max/1024/1*CLfK9RnK6wBpuXI_JaytEw.png" /></figure><p>You probably know this situation: everything works, the interfaces are clear, users are happy — but at some point, someone turns to the writer and asks: <em>“What about the numbers?”</em> That’s when the real nightmare begins.</p><p>Should a UX writer be responsible for product metrics like conversion, retention, or NPS? Or does their job end when the user gets what they need from the product with minimal effort? Let’s unpack that.</p><h3>Text Is Subject to Context</h3><p>The idea of a UX writer having ownership over metrics sounds tempting:</p><ul><li>they’re no longer just the person proofreading and fixing commas</li><li>content design becomes more influential</li><li>everyone can see firsthand that copy impacts the funnel as much as design does</li></ul><p>All of this gives writers a certain kind of control — albeit disguised as a data-driven approach. Is that a bad thing? Not at all.</p><p>But on the other side of that approach lie metrics that aren’t as obedient as we’d like. Interface copy, by its very nature, doesn’t live in a vacuum — it competes with visuals, flows, server response speed, and the user’s Tuesday-morning mood when they open the app out of habit, wanting to do absolutely nothing.</p><blockquote>For instance, you change the button from <em>“Place order”</em> to <em>“Confirm order,”</em> and nothing happens. Why? Because at that exact moment, designers added another step to the flow. Or because users, as it turns out, don’t even read that screen.</blockquote><p>In the end, the writer stares at analytics tables, fighting not for meaning or clarity but for a few tenths of a percent in A/B tests. That’s no longer UX — that’s a <strong>cult of analytics</strong>.</p><h3>What You Can Measure</h3><p>That doesn’t mean all metrics are useless. On the contrary — when backed by observation, they work. Here are four that do — if you interpret them correctly:</p><ul><li><strong>Task success rate</strong> — the percentage of users who successfully complete a task. Example: changing <em>“Confirm order”</em> to <em>“Place order”</em> increased conversion.</li><li><strong>Comprehension rate</strong> — the level of understanding.<br> Measured through testing, completion time, and the number of clarifying actions. Like when <em>“Reinstall the app”</em> performs better than <em>“Try again later.”</em></li><li><strong>Trust index</strong> — literally, trust. Measured not by surveys but by behavior: repeated actions without hesitation, absence of re-check clicks.</li><li><strong>Micro-conversions</strong> — small wins on the path to a big one: moving to the next step, enabling notifications, saving settings.</li></ul><p>You don’t have to use all of them — and you don’t have to limit yourself to just these. Large companies can even afford to develop their own metrics. The key is that a metric should <em>actually work</em> and be <em>used correctly</em>, so that cause and effect don’t get swapped.</p><h3>Metrics Are Feedback</h3><p>It really does help for a UX writer to understand numbers — but not to live by them. What matters most is understanding the context in which text influences behavior, and treating numbers as feedback. Because metrics show <em>what happened</em>, but never <em>why.</em></p><p>A UX writer doesn’t have to count, but they do need to identify <em>where</em> and <em>why</em> those numbers drop.</p><blockquote>For example, when the <em>“Try for free”</em> button suddenly stops converting — it’s not always about the wording. Sometimes people just don’t believe in <em>“free”</em> anymore.</blockquote><p>Metrics exist to help us understand the world — not to justify ourselves to it. A UX writer with ownership doesn’t chase numbers but can interpret them without losing the human meaning behind them.</p><h3>Don’t Let Numbers Replace Meaning</h3><p>Let’s hope the industry has finally moved past the point where interface copy was seen as just another ad banner designed to sell. Its real purpose has always been simpler — <em>to not get in the way, and to help.</em></p><p>That’s why the KPI for text isn’t CTR — it’s whether it <em>reduced cognitive friction</em> and <em>helped the user achieve their goal.</em></p><p>Because good UX copy doesn’t always increase conversion. Sometimes, it simply restores trust — and that, eventually, <em>will</em> increase conversion.</p><img src="https://medium.com/_/stat?event=post.clientViewed&referrerSource=full_rss&postId=f4754a1d5e77" width="1" height="1" alt="">]]></content:encoded>
        </item>
        <item>
            <title><![CDATA[Figma for UX Writers: 12 Plugins That Make Your Work Easier]]></title>
            <link>https://medium.com/@abramovichanton/figma-for-ux-writers-12-plugins-that-make-your-work-easier-a52e4a8e6052?source=rss-70b94fa17d09------2</link>
            <guid isPermaLink="false">https://medium.com/p/a52e4a8e6052</guid>
            <category><![CDATA[ux-design]]></category>
            <category><![CDATA[ux-writing]]></category>
            <category><![CDATA[product-design]]></category>
            <category><![CDATA[ux]]></category>
            <category><![CDATA[productivity]]></category>
            <dc:creator><![CDATA[Anton Abramovich]]></dc:creator>
            <pubDate>Fri, 31 Oct 2025 16:01:53 GMT</pubDate>
            <atom:updated>2025-10-31T16:01:53.856Z</atom:updated>
            <content:encoded><![CDATA[<figure><img alt="" src="https://cdn-images-1.medium.com/max/1024/1*k8EPb02zxgP98mZcZDW6rA.jpeg" /></figure><p>If you work with text directly in Figma, these plugins will save you hours of routine work — and your last nerve cell. All of them are tested in real projects and integrated into my editorial workflow.</p><h3>Working in Figma</h3><p><a href="https://www.figma.com/community/widget/1158809166636728308"><strong>DONE!</strong></a> — a small panel that helps you track the status of an entire design task. It’s easy to adapt to your team’s workflow and assign contributors. Great for keeping editors, designers, and developers in sync.</p><p><a href="https://www.figma.com/community/widget/1389895253265745876"><strong>Checkmark</strong></a> — looks like a simple to-do list, but the devil’s in the details. I’ve tested dozens of plugins looking for the right balance between simplicity and capability. Checkmark supports nested lists and full-list editing via Markdown, shows overall progress, and even allows you to add hyperlinks (a surprisingly rare feature). You can also create checklist templates — a huge time-saver when the team moves into detailed design or pre-development prep.</p><p>If you need the simplest possible to-do list, try <a href="https://www.figma.com/community/widget/1342178141900653188"><strong>Simple Tracked ToDo</strong></a> or <a href="https://www.figma.com/community/widget/1477952869464627187"><strong>Checklist</strong></a>.</p><h3>Prototyping and Concepting</h3><p><a href="https://www.figma.com/community/plugin/742764242781786818"><strong>Wireframe</strong></a> — a large collection of wireframes for both web and mobile, plus components for flow maps. Extremely handy when you need to sketch a structure quickly, brainstorm, or explore early concepts.</p><p><a href="https://www.figma.com/community/plugin/731627216655469013"><strong>Content Reel</strong></a> — an asset library for early-stage design work when you need to fill screens quickly with realistic content. It includes text assets like formatted dates and phone numbers, as well as visual ones — photos and icons.</p><p><a href="https://www.figma.com/community/plugin/733902567457592893/autoflow"><strong>Autoflow</strong></a> — a super-convenient connector tool for organizing layouts. You can customize color, thickness, and rounding; to draw a connection, just select two frames while holding Shift.</p><h3>Working with Text</h3><p><a href="https://www.figma.com/community/plugin/1238151538920262810"><strong>Frontitude</strong></a> — an AI-powered plugin for writing and improving text. Useful when you need to “tweak” text, explore alternative versions, or fit it within a character limit. It also supports style guides — you can upload your tone-of-voice, component guidelines, and text-category recommendations (e.g., for error messages), and the plugin will apply them when generating text.</p><p><a href="https://www.figma.com/community/plugin/1044632054433601396"><strong>UX writing status</strong></a> — a very simple but incredibly useful plugin. It includes four statuses: <em>To do</em>, <em>In progress</em>, <em>Review</em>, and <em>Done</em>. Select text, click the relevant status, and the plugin labels the text layer in the left panel. It prevents confusion when a flow has multiple states and texts — and lets designers instantly see what’s at which stage.</p><p><a href="https://www.figma.com/community/plugin/754026612866636376/spelll-spelling-grammar-checking-for-figma-figjam"><strong>SPELLL</strong></a> — checks spelling and grammar. You can add your own glossary so it doesn’t flag the same words repeatedly. Helps catch typos that Figma’s built-in spell checker might miss.</p><p><a href="https://www.figma.com/community/plugin/745519632050796775"><strong>Typograf</strong></a> — formats punctuation according to typographic standards: adds correct quotation marks (« »), recognizes nested quotes, removes double spaces, formats fractions, and adds non-breaking spaces between numbers and preceding words. Lots of useful small fixes. The only drawback — it lacks flexible customization for individual editorial standards.</p><p><a href="https://www.figma.com/community/plugin/977968775104001765"><strong>Space&gt;&gt;cleaner</strong></a> — removes duplicate spaces but doesn’t replace non-breaking ones. Handy if <em>Typograf</em> conflicts with your editorial documentation but you still need to clean up the text.</p><h3>Other Tools</h3><p><a href="https://www.figma.com/community/plugin/739922281164562258/better-font-picker"><strong>Better Font Picker</strong></a> — a separate, non-closing panel with all installed fonts. Useful when tidying up layouts or fixing inconsistent font use — no more searching through Figma’s right-hand panel each time.</p><p><a href="https://www.figma.com/community/plugin/801195587640428208/design-lint"><strong>Design Lint</strong></a> — primarily a designer’s tool, it checks for style errors in layouts. But it’s also valuable for editors who work directly in Figma — to verify text styles and maintain consistency across mockups.</p><p>Let’s connect in <a href="https://www.linkedin.com/in/anton-abramovich">LinkedIn</a></p><img src="https://medium.com/_/stat?event=post.clientViewed&referrerSource=full_rss&postId=a52e4a8e6052" width="1" height="1" alt="">]]></content:encoded>
        </item>
        <item>
            <title><![CDATA[UX 3.0 — how AI is changing the way we design]]></title>
            <link>https://medium.com/@abramovichanton/ux-3-0-how-ai-is-changing-the-way-we-design-ed44880e386f?source=rss-70b94fa17d09------2</link>
            <guid isPermaLink="false">https://medium.com/p/ed44880e386f</guid>
            <dc:creator><![CDATA[Anton Abramovich]]></dc:creator>
            <pubDate>Mon, 27 Oct 2025 09:49:54 GMT</pubDate>
            <atom:updated>2025-10-29T20:05:59.897Z</atom:updated>
            <content:encoded><![CDATA[<h3>UX 3.0 — how AI is changing the way we design</h3><figure><img alt="" src="https://cdn-images-1.medium.com/max/1024/1*Tg9-Udz7XdVArCsJDtomew.png" /></figure><blockquote>UX is becoming less about interfaces — and more about relationships between humans and self-learning systems that make decisions on their own.</blockquote><p>Over the past forty years, UX has come a long way: from early computer operating systems, to the internet, to mobile devices. The next stage is artificial intelligence. AI won’t simply integrate into existing UX practices, as it may seem now — it will introduce an entirely new element: autonomous system behavior.</p><h3>How UX evolved</h3><p>In nearly half a century, UX has gone through three major stages:</p><figure><img alt="" src="https://cdn-images-1.medium.com/max/1024/1*1o1XGR03YuLMcPRWXnS-RA.png" /></figure><p>In short: UX 1.0 made technology understandable, UX 2.0 made it convenient, and UX 3.0 makes it human. But “human” here refers not to the designer’s idea of humanity — it’s about the user’s experience of it. More on that in a moment.</p><p>The latest transition hasn’t been easy. Poorly designed AI systems can already cause harm — think of accidents involving self-driving cars. The reason lies in technology-centered design rather than user-centered design. This is a new challenge, and it can’t be solved with traditional UX tools, because their principles were built for a completely different mode of human–product interaction. They don’t account for the complexity of AI systems.</p><p>That’s where the new paradigm begins — one in which UX design must make AI transparent, controllable, and trustworthy. A note of scope: UX 3.0 applies primarily to products that are fully AI-based and operate at the system level — not to those that merely use AI for specific functions like Spotify’s recommendation algorithms, customer-support chatbots, or image-generation tools.</p><h3>The components of UX 3.0</h3><ul><li><strong>Ecosystem design</strong>. UX extends beyond individual products and is now designed at the ecosystem level — across connected scenarios that span mobile, desktop, wearable, and AI systems. What matters most is their coherence. UX design covers the entire lifecycle of AI — from model design and training to behavior monitoring, updates, and even the end-of-life stage. Beyond the interface, UX now includes business logic and data layers: the user experience depends on solution integration, infrastructure, and data quality. Finally, UX is embedded not only in the technical but in the socio-technical context: design must account for how AI integrates into broader social and cultural environments.</li><li><strong>Innovation</strong>. Technology should serve the user, not the other way around. Innovation shifts from technology as an end in itself to the creation of distinct, useful, and easily learnable experiences. It helps not only to address existing pain points but also to uncover unmet needs, anticipate them, and close the loop instantly. At the system level, UX must connect across all user touchpoints.</li><li><strong>Cooperation</strong>. AI changes the role of the user: they’re no longer a passive recipient of solutions but an active participant in co-creation. AI enables a new level of collaboration — large language models, assistants, and agents allow for shared work and distributed roles. It also brings UX into real time: recommendations, predictive personalization, and assistants turn interaction into a continuous two-way dialogue where the interface adapts during user actions, not after them.</li><li><strong>Interactivity</strong>. Here we return to the idea of “humanity.” Today, designers often interpret it as caring for the user through their own professional lens — guided by an internal compass that determines how to soften business needs within the interface without frustrating the user. And yet, we still see unsubscribe flows stretched across seven screens. In UX 3.0, humanity means predictability, transparency, control, and, above all, trust. For most users, AI remains a black box. As it becomes part of everyday life, UX design must create more coherent and understandable interaction experiences.</li></ul><h3>The UX 3.0 framework</h3><p>Here’s what this means in practice for UX professionals:</p><ul><li>Design will have to integrate across the entire AI lifecycle — starting with data collection.</li><li>UX teams will become cross-functional. Alongside designers, writers, and researchers, they’ll include engineers, analysts, product managers, and domain experts from the specific fields the product serves.</li><li>UX work will operate on multiple levels simultaneously: at the product level, the system level, and the forward-looking level. Think of it like an open-world game — developers don’t design every single thing players end up doing. But with AI, they’ll have to (or at least try to), because it directly affects the real world.</li></ul><h3>What changes for UX professionals</h3><p><strong>UX leads</strong> will no longer focus on content management or flow design — their role will expand to building full architectures of human–AI interaction. They’ll also help define the product’s guiding principles: predictability, safety, ethics, and beyond. And it seems that generalists may have an advantage here.</p><p><strong>Designers and writers</strong> will need to understand the principles of explainability — in other words, control: how to create predictability within a fundamentally new mode of interaction with a fundamentally new category of products.</p><p><strong>Researchers</strong> will move closer to system analysts.</p><p>The most interesting challenge lies with <strong>companies</strong>: they’ll have to rethink strategy at a structural level — add new metrics, and reorganize departments and teams.</p><p>In essence, product thinking will have to expand into system thinking, and the new goal of UX professionals will be to preserve humanity and trust in a world where technology is no longer just a tool — but a partner.</p><p>Let’s connect in <a href="https://www.linkedin.com/in/anton-abramovich">LinkedIn</a></p><img src="https://medium.com/_/stat?event=post.clientViewed&referrerSource=full_rss&postId=ed44880e386f" width="1" height="1" alt="">]]></content:encoded>
        </item>
        <item>
            <title><![CDATA[“Let’s research this more” doesn’t work]]></title>
            <link>https://medium.com/@abramovichanton/lets-research-this-more-doesn-t-work-caff37a77dee?source=rss-70b94fa17d09------2</link>
            <guid isPermaLink="false">https://medium.com/p/caff37a77dee</guid>
            <dc:creator><![CDATA[Anton Abramovich]]></dc:creator>
            <pubDate>Thu, 23 Oct 2025 17:32:49 GMT</pubDate>
            <atom:updated>2025-10-27T09:51:20.734Z</atom:updated>
            <content:encoded><![CDATA[<figure><img alt="" src="https://cdn-images-1.medium.com/max/1024/1*RsWmzCDQ2bhdVX5dw5allw.png" /></figure><p>We often hear this when we’re unsure about a decision: let’s run one more interview, test the hypothesis on another audience. It seems logical — more information equals less risk. But in a dynamic environment, it’s the opposite: the longer we research, the higher the chance of being wrong.</p><p>In 1987, cognitive psychology researchers Don Kleinmuntz and James Thomas studied when it’s better to act and when to keep gathering information. They modelled a situation where every action affected the state of a “system,” and the system itself responded with a delay. In other words, the system was constantly changing, but feedback didn’t come immediately.</p><p>Participants were asked to manage the condition of a patient (a metaphor for a complex system). At each step, they could either clarify the patient’s state (i.e., research more) or apply treatment without full certainty (i.e., make a decision). The results appeared with a time lag — just like in real products, where the effects of changes are not visible right away.</p><h3>Here’s what the researchers found:</h3><ul><li><strong>People cling too tightly to certainty</strong><br> Most participants, afraid of making a mistake, spent too much time researching — even when the delay itself made the situation worse.</li><li><strong>The most effective strategy is a hybrid one</strong><br> Some participants gathered information only where it genuinely reduced uncertainty, and then immediately moved to action. They didn’t try to clarify everything — only what truly affected the outcome.</li><li><strong>People lose to randomness</strong><br> Because of cognitive biases (overvaluing new information, overcorrecting for minor changes, avoiding visible mistakes), over time, the average human performance turned out worse than random choice.</li><li><strong>Time isn’t free</strong><br>We experience errors more sharply than delays. We fear making the wrong decision because an error is visible, tangible, and leaves a trace. The cost of delay, however, is stretched over time and harder to feel. While participants were seeking certainty the system kept changing. By the time they were ready to act, they were solving yesterday’s problem.</li></ul><p>UX teams often fall into the same trap. We can spend weeks debating a phrasing or running tests. Meanwhile, the market and users change, and the context shifts. By the time a decision is ready, it fits perfectly — with the previous version of the product.</p><h3>What we can take away from this</h3><ul><li>The desire to know as much as possible can be an extremely inefficient strategy and doesn’t guarantee a better decision — especially since “a good decision” means different things to an editor, a product manager, and a designer.</li><li>When time and information are limited, an intuitive decision can be more effective than a meticulous analysis and weeks of research.</li><li>Inaction is also a decision — and it comes at a cost.</li></ul><p>Sometimes, the best way to reduce uncertainty around you is simply to act and see what happens. Because action is a form of research.</p><p>Let’s connect in <a href="https://www.linkedin.com/in/anton-abramovich">LinkedIn</a></p><img src="https://medium.com/_/stat?event=post.clientViewed&referrerSource=full_rss&postId=caff37a77dee" width="1" height="1" alt="">]]></content:encoded>
        </item>
    </channel>
</rss>