Steven Eggert

I don't know when or how it happened, but Config has become one of the major yearly conferences in our industry. Given how much Figma plays a role in the practice, it's common for design system teams to watch with anticipation/excitement/dread for what's coming. In our latest episode, Davy Fung and I discuss the reveals from Config and what that may mean for folks in the design systems practice moving forward. Listen here: https://lnkd.in/gRMBzfbJ #designsystems #figma #congfig #podcast

57

1 Comment

Etsy was recently seen A/B testing AND eventually shipping a more visible search with the help of a higher contrast orange background button color (instead of a ghost-like icon only approach). Full screenshots: https://lnkd.in/gVJThW3W Now I don't work for Etsy (as much as I'd love to design some experiments for/with them) so we can't see the actual numbers beyond this feint probabilistic residue from the public domain. However, the amazing Andrey Andreev who runs experiments at Metro recently ran a very similar one and shared a solid result on GoodUI - with a positive impact to sales and searches. N = 896,009 SRM Check = Pass p-val on Increased Sales = 0.0139 (I know some people might like this to be even lower for a primary metric; but there is also reinforcing evidence on increased searches with an even lower p value at "ultra stat-sig" levels using Ron Kohavi's language; p < 0.00000001). I inverted the a/b test and categorized it under the Filled Vs Ghost Button pattern - further weakening the ghosts and strengthening the high contrast filled button styles on multiple metrics. https://lnkd.in/gmD8Exvc Thanks Andrey for a somewhat of an unintended replication, as I know your source of inspiration came from somewhere else on this one. :) Have you tested anything similar or different. If so, what did you discover? I'd be super curious to see more examples where increases to search have either ❌ failed or ✅ succeeded. Please share independent of outcome. ps. We also have a handful of similar "visible search" experiments where companies tested adding or hiding search: https://lnkd.in/gvWqzrK5 - further supporting this direction. [EDIT]: Note the additional confounders / variables in the Metro test - a stronger border, a change in background color of the search box as per Jakob Nielsen comment. They may be similar along the lines of the increased visibility theme - but possible additional variables nevertheless. #experimentation #ecommerce #replication #optimization #ui #ux

381

28 Comments

🎈Pretty colors aren’t good enough. Color needs to do the hard work of balancing brand voice with utility and accessibility. Fortunately there's color tooling to help designers with the complexity of managing contrast across multiple colors, tints and shades. A valuable color tool should allow you to do a couple of things: → Seed your palette with brand colors → Provide sufficient tints and shades with documented accessibility contrast → Generate output that can be leveraged by design and engineering team members 🌈 Genome Color Tool by Kevin Muldoon The Genome Color web app allows you to create custom color scales from brand colors by leveraging recommended tints/shades from proven color systems. The thinking behind the tool is so smart it makes my head spin and the video tutorial provided makes it super easy to use. Writeup: https://lnkd.in/gjAmBJq6 Tool: https://lnkd.in/ggumsweN Figma plugin: https://lnkd.in/gr6B_Cpi 👨‍🎨 Leonardo by Nate Baldwin Leonardo is a web app that allows you to build out multiple color ramps for a uniform theme that share the same contrast ratios. It offers a number of slick visualizations, accessibility options and ways to export everything. There’s even a plugin that allows you to pull your palette directly into Figma. Writeup: https://lnkd.in/gufQhmFT Tool: https://leonardocolor.io/# Figma plugin: https://lnkd.in/g-FFqJuW 🎨 Supa Pallete by Angelo Libero Supa Pallette is an elegant Figma plugin with a paid option. I’ve tested a fair share of Figma plugins, and this one is packed with features and worth the license. Starting with a seed color, the tool allows you to customize WCAG or APCA accessibility contrast scale, export visual color tiles for preview and even generate Figma variables. Tool: https://lnkd.in/ge92ND2N Figma plugin: https://lnkd.in/gX7sq25U 💎 For as complicated as color can be, we’re lucky to have access to so many resources. Some takeaways our team at Exygy now uses to make color decisions: → Align the team on accessibility guidelines like WCAG or APCA → Meet with designers to find out what Figma tooling and color method, including HEX or HSL, are the most intuitive for them → To ensure adoption, talk with engineering to find out what a valuable output format looks like, whether that’s CSS or tokens I’ve posted about managing color before including more helpful resources: https://lnkd.in/g6E3zYjA #color #ui #design #designsystems #figma #accessibility #branding #color

67

4 Comments

Adobe Researchers have unveiled VideoGigaGAN, a revolutionary model that can upscale videos by up to 8× while preserving intricate details and ensuring smooth temporal consistency. 🎬✨ 🔍 Under the Hood: VideoGigaGAN builds upon the success of GigaGAN, a large-scale image upsampler, and takes it to the next level by addressing the challenges faced by previous video super-resolution methods. 💪 The team has implemented several innovative techniques, such as: 🧩 Temporal Attention Layers: These special layers help the AI pay attention to how the video changes over time, ensuring that the upscaled video looks consistent from frame to frame. 🌊 Flow-Guided Propagation: This feature allows the AI to better understand the motion in the video, resulting in smoother and more accurate upscaling. 🎨 Anti-Aliasing Blocks: Just like how a painter blends colors to create a smooth picture, these blocks help the AI blend pixels together to avoid jagged edges in the upscaled video. 🎢 Direct Shuttling of High-Frequency Features: This technique helps preserve the fine details in the video, making sure that the upscaled version looks sharp and clear. These improvements enable VideoGigaGAN to generate videos with stunning high-frequency details and impressive temporal consistency. We are almost at 'Zoom and Enhance' kind of upscaling levels. » https://lnkd.in/dEshEGh2 This won't be open-sourced. I'm sure of it. Please, Adobe, prove me wrong. Source: https://lnkd.in/dB3K_CyX ↓ ☝️ Follow for more AI-News » https://lnkd.in/esb3PH9W ✌️ Check out our AI Workshops and supercharge your AI Game (so far 90+ successful individual workshops done for companies and corporations) https://lnkd.in/dXGDz58E" 🤘 Test our Gen AI platform creamAI: http://www.creamai.de #ai #genai #videoupscaling #machinelearning #videoproduction #upscale

293

31 Comments

Hey Designers, Just a quick reminder: your NDA isn’t just a piece of paper—it’s a legal obligation. Sharing password-protected portfolios without permission isn’t just risky, it’s illegal. Here's a hypothetical but all-too-real scenario: what if your competitor needed some insights? They reached out to designers at rival companies under the guise of potential job opportunities. As UX designers, we work on ideas and projects that may not be implemented today but in years to come. Imagine your company’s long-term strategies being exposed, some user research data shared, and performance and accessibility KPIs revealed. Sketchy, right? For Recruiters: If a candidate says their portfolio is under NDA, respect that. It shows they value confidentiality and will protect your info too. If they’re willing to break an NDA for you, what’s stopping them from doing the same to you down the line? Remember, many designers work on projects that are never meant to be displayed publicly. Let’s keep things above board and respect each other’s legal boundaries. Remember the Golden Rule: treat others as you would want to be treated. Thanks for understanding!

12

With the rapid adoption curve of AI, users are expecting more ‘magic' from their digital tools and products. Great piece here about the relevancy and reframing of the value of design when building a non-deterministic digital experience: “Sophistication, in prompting, comes from deeply understanding the user — their attitudes and behaviours towards this experience — and integrating that knowledge into the prompt. This is even more important than with traditional software interactions because of the open-ended nature of conversational experiences. As noted by designer, Emily Campbell, “the more autonomy that users have to direct computers to their personal whim, the better they will expect computers to understand them and anticipate their needs.” Logan Kilpatrick, head of developer relations for OpenAI, boils it down to “context.” The model needs context to do its perform a specific task well. This makes prompting an “inherently human” task because it requires that we understand the nuances of what people are trying to achieve.”

8

1 Comment

Here's how to use AI in your UX day-to-day. AI lets us access a ton of scenarios instantly. With the right prompts, you can tap into a historian of user behavior. By leveraging advanced algorithms to predict user behavior and analyze massive datasets, AI anticipates needs before users even realize them. Think of Google's search engine suggesting queries in real-time—that's the power of AI in UX. At Upstory, AI-driven tools aren’t a crucial step in our flow, but we’ll use them on occasion as an extra step in our research process. We prompt our AI with the right story, intended user, and goals, then run through various scenarios to catch potential user behaviors we might’ve missed. This doesn’t replace actual user interviews, but it’s a great way to get quick insights for projects on a tight timeline or budget. AI won’t replace you (yet?), but it can definitely make you better. 🤖

4

Excited about supercharging your design team? 🌟 Don't miss our April Gather & Grow webinar! Learn how DesignOps can transform your design crew into your brand’s MVP and drive sustainable ROI. 🚀 Join Aletheia Delivre from Zapier, Jasmine Price from Airbnb, and Kim Marchant from JPMorgan Chase & Co as they dive into strategies for keeping that creative spark alive, seamlessly integrating brand and product, and why embracing progress over perfection matters. Plus, get insights on leveraging AI for feedback and shifting from centralized to distributed teams effectively. Ready to elevate your design game? Sign up now and let's make magic happen together! ✨ #DesignOps #Webinar #CreativeMomentum

2

1 Comment

Products like Recraft are like the Perplexity for creating images with #GenAI. They are leaning into the benefits of going beyond a conversational prompt-based interface ONLY and integrating traditional, familiar GUI-based workflows with prompts to provide outcomes that creative people seek. In this example, they've notably improved the experience of creating multiple #images in a consistent style using #AI. In addition, they seem to be among the few (only one?) offering SVG export, making them a great option for #illustration. https://lnkd.in/eiGPqbKU

1

1 Comment

I see the potential in AI tools, but let's hold up for a second. This recent mess with Figma's AI tool seemingly cloning Apple's Weather app is a wake-up call. It's a glaring example of what happens when we don't put proper guardrails in place. AI can be a game-changer, but not if it tramples on intellectual property and copyright. We need these tools to be smart enough to avoid ripping off existing work. It's not just about innovation; it's about respecting creativity and ownership. https://lnkd.in/gUHxJd_c

2

Feel like this is the future use cases for AI. Despite all the fear mongering about AI taking everyone's jobs (and of course, legitimate copyright questions), I think it'll ultimately boil down to a great toolset that augments existing tools and platforms, enabling us to make better work. Just because there's a camera in every phone, doesn't make us all photographers. Sure, everyone might be able to make a video in Sora once they release the public version, but great work will still require taste, experience and skill. Exciting times ahead, once the initial mainstream hype dies down. #ai #adobe #sora #runway #pika https://lnkd.in/eP2aYxx8

18

4 Comments

Being Designerly 93 has curated content about Google AI overviews, agentic UX, SONOS redesign fail, US gov's 10K websites in a decade, accessibility for solopreneurs, visualizations of words in The Office, and visual evolution of logos. Read more 👉 Link in comment #Designerly #Design #UX

1

1 Comment

Will AI Diminish Our Capacity for Critical Thinking? Google's new "AI Overview" feature, which compiles search results using AI, has been dishing out some inaccurate—and even dangerous—advice. For instance, suggesting mixing glue into cheese to prevent it from sliding off pizza, or using mustard gas to sanitize a washing machine! Extreme examples like these make people pause, but other incorrect results or AI hallucinations may not be given a second thought. In his book, The Shallows: What the Internet Is Doing to Our Brains, Nicholas Carr argues that the internet’s constant stream of information and demand for multitasking encourages a superficial approach to information processing, leading to scattered thinking and diminishing our capacity for deep thought and contemplation. Critical thinking involves intentionally evaluating information, ideas, or situations from different perspectives to understand and decide how to act on them. Design involves people—their inputs, opinions, and feedback—and critical thinking plays a crucial role in the entire design process. Designers are always asking questions, seeking evidence, and striving to understand deeply. Faced with a design opportunity or “problem,” they don’t dive into a solution without first understanding the real problem, which may lie layers below the surface. How can you think critically? - Apply designerly curiosity: Ask "why," "how," and "what if" questions to delve deeper into ideas, arguments, and situations. - Analyze information: Break it down into smaller parts, identify key points, evidence, and assumptions. Evaluate the credibility and reliability of sources. - Consider multiple perspectives: Train yourself to view issues from different viewpoints to gain a more comprehensive understanding, balance your view of a topic, and recognize potential biases. Developing critical thinking is an ongoing process that takes time and practice. Start by incorporating these techniques into your daily life. Gradually, you'll find yourself becoming more adept at analyzing information, making informed decisions, and engaging with complex ideas. Think Critically! Stay ahead by honing essential #designerly skills, behaviors, and habits: - Follow me here on LinkedIn for regular insights and tips on mastering these skills. - Subscribe to my #BeingDesignerly newsletter for exclusive content, practical advice, and early access to updates about my upcoming book on designerly skills. Elevate your design career and stay competitive. Don’t miss out—follow and subscribe today! Link in comments. #UX #Design #CriticalThinking

13

12 Comments

Had to share this thoughtful article about the conflicting feelings we have as designers with new AI tools.... made me consider the threat of too much standardization over innovation, or letting your own creative instincts atrophy. Check it out and lmk your thoughts

4

Hello fellow designers and creatives. I’ve decided to jump back into writing. Find me on Substack where my musings will focus on industrial design, design, strategy, sustainability, and systems thinking. Thanks for the sub. :) https://lnkd.in/e4Wpfbrp #industrialdesign #designstrategy #systemsthinking #sustainability

3

Many designers fall into the trap of getting way too attached to their own designs. David Fano has seen designers get caught up in all sorts of stuff: — Buttons that look cool but don't really do much — Interfaces that are more flash than substance — Design systems that are more complicated than they need to be The thing is that none of that matters if you're not solving the user's actual problem. The best product managers know that the user is king. They focus on solving the user's needs, not just making things look good. It's easy to get lost in the weeds when you're deep in the design process. You might find yourself going down the rabbit hole of pixel-perfect layouts and animations that are super on-trend. But that’s how to lose sight of why you're designing in the first place. Your job is to make the user's life better, not just to make something that looks cool. As designers, we need to be real with ourselves: — Does this feature actually make the user's experience better? — Are we solving a problem that users actually have? — Or are we just geeking out over our own design skills? Always keep the user front and center. — How do you balance making great designs with actually solving user problems? Watch the full episode with David for more talk on user-centered design, now live on our new content series at minimumviablepodcast.com.

18

1 Comment

How do design system teams prove their value? Finding hard metrics around the impact of design system teams has historically been a challenge. A year ago, I met the design system team at Eventbrite who could confidently say that they saved 534 days of engineering effort with their design system. Here's their calculation: Time spent to develop a component pre-design system - time spent to install the design system component = Time saved In order to extrapolate this number, though, you need to know how many times a component has been used. And, that used to require a savvy dev on the team to build something custom. That's no longer the case. We built something at zeroheight that will allow you to track component use in a single repo or across all of your repos. This should help teams show how valuable they are, plus do things like: see which components to prioritize based on use see which product teams are adopting As an added bonus, anyone using the new feature set can also monitor design system packages, allowing them to quickly see which teams are using outdated packages. On a personal note, I'm hoping with data like this, more teams can confidently quantify their impact. And along the way, we'll get better as design system practitioners at identifying the right things to work on. This new feature set is in beta and totally free. Even if you aren't using zeroheight to document your design system, you can use it. Takes like 15 minutes and you've got rich data. Here's the sign up:

7

4 Comments

Hey friends, I find it challenging to ask directly for support, but I need your opinions on what it means to center the most impacted in design 💬 To "Centre the Most Impacted" is the second principle I'm writing about, following "To be a Designer is to be a Facilitator." I'm passionate about capturing diverse voices on this topic, but I'm struggling to gather people's thoughts 😖 Feedback is at the core of designing for collective spaces - your thoughts, opinions, and ideas are incredibly valuable! The article on what it means to "Center the Most Impacted" is slated for release next Monday. So before then, I would love if you could: 🙋‍♂️ Share your thoughts in the comments, 🔄 Reshare this post, 🗣 And/or use the Miro board linked below to provide feedback! I want to emphasize that every voice matters in this process, and I really cannot write these articles without you - so anything helps! Even suggestions on making it easier to interact 😊 Thank you so much for your support!

4

2 Comments