Pixels of the Week – April 14, 2024

Accessibliity & typography, research when you can't access users and 100 Figma tips

My curated weekly-ish online newsletter, where I share interesting articles, tools, and resources I found during the week. You can expect content about UX, design, user research, accessibility & tech, but also some processes, some inspiration, sometimes books, and a couple of videos and podcasts. Also, don’t forget to, subscribe to the newsletter to get notified, you will get the weekly links directly in your mailbox, and be notified when I publish other articles.

Now: what I’m currently up to

This week I run a brand-new workshop on bias mitigation for UX researchers. I’ve created new Miro boards for this one, and new cards with different bias mitigation strategies. The feedback was amazing. I plan on releasing those new cards as well at some point, once they are tested a little bit more. Meanwhile, enjoy a sneak peek on LinkedIn

Discount code for UX London!

As a reminder, I will be giving a workshop titled Designing adaptive reusable components and pages at UX London on June 18 & 20. I have the discount code JOINSTEPHANIE people are welcome to use for 20% off the 3-day event. Full details can be found here. I would love to see you there!

Get my ticket

Most popular content this week

Designing Accessible Typefaces with Eleni Beveratou: super insightful 1hour 30 video on the different aspects of typefaces and accessibility, including the difference between legibility and readability and a little bit of debunking on serif fonts and fonts for dyslexia. I love how Eleni introduces the concept of “likability”, where people got used to certain fonts in digital environments (arial, verdana), so even if they are not the most legible, people still prefer those.

Interesting articles that caught my attention

UX research, design

  • 5 ideas for user research when you can’t talk to users (10min) Direct access to users can be tricky sometimes. What can you do then? Chris Myhill offers 5 ways to do research when you can’t talk to users: check forums (or any place where people talk about your product online), read review (from appstores, aggregators, etc.), use proxies and speak to the people who have direct access to users (support, sales, training people), check analytics data and use remote session recording tools.
  • The UX dilemma: do complex products need instruction manuals? I’m tired of hearing “good products don’t need instructions because they are intuitive”. When the product is complex (often enterprise products), users might need and expect instructions. And there’s nothing bad about this. Kike Peña spits product complexity in 3 categories: low, medium and high. For low product complexity: the action itself explains the product usage. For medium product complexity: they require some practice, but complexity is overcome over time, usage, discovery and adoption. For high product complexity, people will expect instructions, support, and a learning curve. Assessing your product complexity can help you prepare, in advance, the type of help that will be needed.
  • Surveys and the “True Score” Mistake (10min) by Chris Chapman: “Survey responses are motivated communication, not expressions of some other latent truth. Our research efforts must be designed to reflect this. Surveys do express truth … but it is not the truth of some true score, as usually assumes. Instead, it is the truth of what respondents tell us, reflecting many influences.” I really like his example of a salary satisfaction survey, and how people will game their answers based on motivation and if the survey is anonymous or not. Personal anecdote: I once was supposed to answer an “anonymous” survey at a company. BUT, they asked for your gender, for how long you were working there and your department as mandatory questions. Answering those 3 questions meant it’s not anonymous because I was the only women working in that department.
  • Hide or Disable (3min) Should we hide or disable a specific element? Disable the element, if a user will be able to interact with it at some point. Hide, if they won’t. For example, if a feature will never be available to a user, hide it. by Sam Solomon
  • Optimising Human-AI Collaboration (12min) by Maria Panagiotidi. If you are going to build an AI tool that helps people with writings, you might be interested in the  Scaffolding Theory: how much supportive guidance provided by AI will augment the writer’s capability. And how it might impact output quality. Adjustments might be needed, and you can let the user customize it.
  • Redesigning navigation with a user-centric approach (15min) Want to improve your navigation? Do user research, using card sorting and tree testings, then test the prototypes with users, and iterate. I’m happy to see a big emphasis on getting the information architecture right, before jumping into screen design in this interesting case study of how welcome to the jungle improved their navigation.
  • Contrasting Aesthetics (10min)  sometimes unexpected combinations can work nicely, if well executed they create memorable experiences and moments of surprise. I love the bitmap typefaces juxtaposed with Renaissance art example in here. By rauno
  • Robin Rendle — Design is a process of getting stuck (3min) “So a whiteboard is the best tool for getting unstuck—perhaps the best prototyping tool of all time—and, although it won’t solve every problem, it will certainly soothe the most painful ones.” – Robin Rendle. I agree so much with this, yet I think it’s the most difficult thing to teach to beginners. They are so eager to jump into design tools and start drafting an interface solution. Also, visual communication, the ability to draw idea, even imperfect is what helps us get unstuck.
  • How to write a UX portfolio case study (+4 common mistakes) (10min) Erik D. Kennedy gives 5 easy to put in place tips to improve your portfolio, with nice examples to illustrate them: start with the end, frame the project, show your decisions (not your process), then show the results of your decisions, and give the reader 2 clear next steps.


Subtitles, Closed Captions, and Open Captions: What’s the Difference? (10min), by Ben Myers. Subtitles transcribe only dialogue (often for translation). Captions transcribe all meaningful audio, including dialogue, music, and background noises. Open captions are embedded in the video, they are nice if people share the video because they will share the captions too, but less flexible. Closed captions are usually supplied as a separate file / metadata. This offers more flexibility like the ability to toggled on /off, change contrast, etc.

Inspiration: fun experiments, beautiful art, and great ideas

Surf the web like it’s 1999 with these old-school cursors are you team 8-bits, Y2K, Windows Aero or Apple Skeuomorphic? As April Fun Day, Figma brought some cursors back from our favorite digital design areas. I can’t decide because I love how cute the 8-bits are, but the Y2K are awesome too. I hated Win Aero though, so, no thanks.

Useful tools & resources

  • Blip Finally, I found an equivalent of airdrop that lets me share files between my Android phone and my Macbook / iPad. It’s free for personal use too, so that’s a nice plus.
  • 100 Figma Design Tips this is a very nice Figma file, that contains 100 Figma tips and tricks to boost your workflow. Love the inception!

Cool and Interesting Videos

Death of the Follower & the Future of Creativity on the Web with Jack Conte the interesting journey of the creator of Patreon and how internet started as a platform that democratized creative distribution. Then came the glorious area of the “subscribe” button, that enables creators to go beyond reach. Sadly, with the rise platform-focused algorithms (Facebook’s ranking, TikTok’s “for you” curation), creators cannot reach their following and true fans. And this shift has been devastating for content creators. Honestly, I think it has been devastating for people on social media too, the quantity of noise and fluffy low level content that come with “pleasing the LinkedIn” algorithm for example is huge.