21

A little while ago, we started a conversation about evolving the look and feel of Stack Overflow. Your feedback was clear: you wanted less marketing and more product. You were right. We're sorry we missed the mark, and we're here today with a different approach.

We want to share a first look at the design changes we’re considering, but first, we want to be transparent about why we're doing this now.

The "Why": Responding to Feedback and Preparing for the Future

This is more than a simple cosmetic update. It's a response to direct community feedback and a necessary step to modernize our platform.

1. You've been telling us the design needs an update.

We've been tracking your feedback on what you'd most like to improve about Stack Overflow. "Site usability"—which includes visual design—has been steadily climbing that list. It went from being ranked 12th in January 2024 to being ranked as high as 2nd this past July in the Site Satisfaction Survey. We've read every piece of feedback, including comments like:

  • “The site looks poorly designed on most modern work displays.” (Site Satisfaction Survey, June 2024)
  • “The UI is pretty full of junk, could reduce the visual noise a bit.” (SO Improvement Survey, 2024)
  • “The visual can improve, despite being very well organized; longer code examples and bigger code boxes would be welcome.” (Site Satisfaction Survey, Aug 2025)

Your feedback is the primary reason we are working on this project.

2. We need to increase our technical agility.

Parts of our front-end codebase are outdated, which makes it slow and difficult to ship improvements and fix bugs. This initiative includes modernizing our codebase and updating our design system to make it easier for us to respond quickly. For you, this means a more consistent experience across the site and a faster turnaround for future feature development and bug fixes.

3. We need to adapt while reinforcing what makes this community valuable.

The world of technology is changing, but our role as a community of humans helping other humans is more critical than ever. This redesign is an opportunity to lean into that by improving clarity and reducing friction, ensuring the focus remains on high-quality knowledge. It is also an opportunity to meet the needs of both tenured and new users.

What's in Scope?

To achieve these goals, we're planning a widespread update. Here are the key areas we’re focusing on for this initial phase:

  • An updated Design System: We're updating our library of reusable UI components (buttons, forms, etc.), introducing a new color palette, a new typography system, and refreshed icons.
  • A wider layout: We know many of you use wide monitors. We're increasing the width of the site to make better use of your screen real estate while still maintaining our responsive layout.
  • Updated core navigation: We're simplifying the top bar and side navigation to make it easier to find what you're looking for. This also includes an update to our filtering capabilities.
  • Refreshed User Profiles: We're rethinking the user profile page to better empower you to showcase your expertise and contributions within the community. We're even exploring things like a free-form text field and the ability to link to external projects.

However, this update will not hit the whole network at once. We will start with Stack Overflow, focusing on the homepage, questions, AI Assist, and profile. As mentioned before, Stack Exchange sites will continue to exist, but under the Stack Overflow name and brand. This change will not be effective for about ~14 months.

A Series of Conversations

This post is the first in a series where we’ll share our progress and ask for your feedback on specific parts of the redesign. As we’re sharing these designs early in our process, you may find that they look rough in places, especially with placeholder icons. We believe getting your feedback now is more important than waiting for a perfectly polished version.

To make sure we get each piece right, we’ve broken the conversation down into following dedicated posts. We invite you to read through them and share your feedback in each post.

What’s Next?

This is a massive project, and we are committed to working with you. Your feedback on each of these posts will influence the next iteration of these designs. We will read every comment and will post updates to keep you informed of our progress.

Thank you for your time and your continued dedication to this community. We’re looking forward to your thoughts.

Improve this question
4
  • 3
    As the design system is changed, how will this impact other projects that use Stacks, such as the accessibility dashboard? Commented 8 hours ago
  • @cocomac They will remain intact Commented 6 hours ago
  • "our role as a community of humans helping other humans is more critical than ever" Sounds good but I wonder if that's really true. Why should it have been less critical in the past? Commented 5 hours ago
  • 1
    "ranked 12th in January 2024 to being ranked as high as 2nd this past July" That is a strong jump within 1.5 years. The site must have aged significantly in between. Just out of curiosity what was 2nd in January 2024? Commented 5 hours ago

6 Answers 6

Reset to default
31

Stack Exchange sites will continue to exist, but under the Stack Overflow name and brand.

So - this is a problem for a lot of us. Historically, the Stack Exchange Sites - whether the bigger ones with our own branding like Super User, Server Fault or Ask Ubuntu, or the smaller foo.stackexchange.com sites have been neglected, and treated as, well third class citizens.

The company already operates as Stack Overflow for most part, and getting pushed under that banner... doesn't really do us much good. As we've said before - we don't have the baggage SO has, and many of our sites have scopes that have nothing to do with programming.

So - what do we need to do to protect our identity? Way back when there was a campaign to preserve the Mascots of world building - Pandora and Slartibotfast, 'cause the company literally failed to understand that even with interconnected sites, our identities were important. As much as keeping the 'bodies' of the sites running, keeping their 'souls' healthy is important as well, and our site identities matter.

Over time - it's also the one thing that distinguishes smaller sites from the massive monolith that Stack Overflow was. A lot of per site branding has been on the backburner over time and while I get the desire to redesign the big site, it's worth remembering that we might benefit from our own site designs and branding, and our own distinctiveness. As a pets mod, what do canines have to do with C, or Rabbits with Rust? It also means potentially site promotion, should it be a thing will entirely be focused on Stack Overflow, over smaller sites, simply cause that's the only brand that matters. Maybe its been the case for too long, but that cements it.

And historically we've had more people confused that non programming sites exist (or asking programming questions on an inappropriate site).

The rest might have value, but I feel like pushing us under Stack Overflow's branding and name will do more harm, and cause more confusion.

There's been significant feedback about this through every single stage of consultation - pointing out issues with this, so I'd ask you take these into consideration.

Improve this answer
5
  • 8
    I completely understand your frustration and I am doing everything in my power to ensure you can keep what makes your site special. Now, not everything lies within my power, but I will continue to represent our users (ALL users) to the best of my ability. And to be clear, I do not have an answer right now. We are doing research on what users want and what this might look like in the future, but there are product decisions that will need to happen as well that haven't happened yet. Commented 10 hours ago
  • 16
    The company's track record with the needs of the smaller sites has been pretty abysmal over the years - we have a few bright spots and things get back to the status quo. From where I am, as a regular user with some experience on meta - I'm really going to have to speak out because I don't really trust the company to understand. A lot of decision makers don't really seem to have the context on the ground - so hopefully you can pass this up to them if it isn't your decision. Commented 10 hours ago
  • 4
    "I am doing everything in my power...." That's odd wording. When Stack Overflow makes any random decision, does it matter to users whether any particular employee was for or against it? I don't understand. Commented 7 hours ago
  • 1
    Well hence me asking what do we need to do? It feels that the decision is made and locked in . My options are organising a moderation strike (which honestly is terrible for everyone involved, and I'm not good at these things), or claiming this is a major decision made without moderator input, triggering that process... which is also not good for everyone involved, or somehow hoping (with no evidence to that) someone in the company can and will put the brakes on that. I think we can make the latter work and there's time to organise but I'd rather not open up the nuclear option. Commented 2 hours ago
  • @Piper This is not a request to you as an individual. This is a request to the company as a whole. When the response from the companies agent is 'not in my power', that implicitly means 'the company is not interested in consulting the community on this matter'. Its definitely within the companies power. If its not yours, then whoever holds that power is should be taking this feedback into account. If they do not, then well, that is ignoring community feedback and as Journeyman Geek said, a direct violation of the agreement after the strike. Commented 1 hour ago
20

An updated Design System: We're updating our library of reusable UI components (buttons, forms, etc.), introducing a new color palette, a new typography system, and refreshed icons.

Do you intend to address the technical design flaws that result in you consistently producing inaccessible UI components? A lot of problems are solved by semantic HTML (including semantic use of classes), and the CSS cascade: the current implementation of Stacks unsolves those problems, and the company lacks the resources or skill to re-solve all of those problems on a case-by-case basis.

If you're updating the Design System, this would be a good opportunity to reimplement your component library according to sound web principles. If you don't know how to do that, this would be a good opportunity to contribute some excellent questions to Stack Overflow (or other Stack Exchange sites, depending on exactly which parts you don't yet know how to do).

"Site usability"—which includes visual design—has been steadily climbing that list.

A large part of this is because you've been reducing site usability, thus making it a more urgent problem. Unless you reflect on why that is, you're not going to successfully make things better.

Improve this answer
13
  • 4
    Destroying Stacks and starting over would be a big win. Commented 8 hours ago
  • "If you don't know how to do that, this would be a good opportunity to contribute some excellent questions to Stack Overflow" Or, hire someone with that expertise. And, ideally not someone who just handles everything HTML/CSS-wise through JS frameworks/components, either. Commented 8 hours ago
  • 4
    @TylerH Hiring someone new would not be sufficient (since internal politics could scupper that), and it would not be necessary (since I believe the current designers and developers are perfectly capable of learning the "new technology" of the Lost Arts of HTML/CSS). It might help, but only once they know enough to know how to distinguish someone with that expertise from someone without it. Commented 8 hours ago
  • Do you have any specifics on how Stacks doesn't use semantic HTML or what exactly isn't accessible about Stacks specifically? We realize it does have some issues, but it is how we lead improving accessibility here at Stack Overflow since every developer uses when building UI. Commented 6 hours ago
  • 4
    @Piper What is the semantic meaning of <div class="d-flex fd-column jc-space-between pb4 mb16 ml8">? Commented 6 hours ago
  • And… it's hard to explain exactly what isn't accessible, but imagine trying to re-colour something using these classes in a userstyle! You could consult the second half of this previous answer, especially the Codeless Code references, for further explanation, although I find it hard to explain how things are bad (except by comparing them to the better approaches). Commented 5 hours ago
  • 1
    If they rewrote Stacks to use meaningful classes like the long abandoned Semantic UI project I would be sooo happy Commented 5 hours ago
  • @GammaGames If you mean this Semantic UI project, I'm afraid they're also doing it very wrong (worse than Stacks!). <button class="ui active button">One</button> is at least three kinds of wrong, and they introduce a fourth when you observe the demo version is actually <div class="ui active button">. (That example should, I believe, be labelled radio buttons that share a name.) Commented 5 hours ago
  • @wizzwizz4 I do, and I agree with you in that case! They even have a few styles where the class order is important 🤢 But I still think the idea has merit, being able to read how a div is styled in plain terms makes maintaining much easier than d-flex fd-column jc-space-between pb4 mb16 ml8 (for me at least) Commented 5 hours ago
  • 1
    @GammaGames HTML classes aren't about styles. Sure, clear CSS is good and useful, but the misnamed "Semantic UI" project still uses classes primarily for styling. <a class="ui black pointing below ignored label">Run Code</a> is not good HTML, for reasons too numerous to list here. Commented 5 hours ago
  • 3
    @GammaGames reading that something has a yellow background because it has the style bg-yellow is just style="bakground-color: yellow" but more compressed. It doesn't explain why it is yellow. For example if one bookmark has a yellow background - why is it different to others? In one usersciript I had to colour elements by adding fc-red-500 to match it to existing styles. After a redesign, I had to go update my code to a different class. Instead of, you know, having the class flag-cast and that having the correct colour for a flag that was cast Commented 4 hours ago
  • @Piper See VLAZ's most recent comment for a good description of an accessibility issue caused by unsemantic HTML. (I know this is a power user thing, not an ordinary user thing, but the issues faced by non-power-users are effectively the same as this, just with various layers of browser configuration and assistive tech as an intermediary.) I can imagine writing an NVDA plugin to expose class="recently-active" to the user, but .bg-yellow-050 could mean too many different things, and isn't stable enough to colour scheme changes, to be useful. Commented 4 hours ago
  • @VLAZ Yup that's the type of semantic-ness I was initially trying to talk about, classes like .avatar, .card, .question, .answer, etc. A few bad examples from the home page of the site derailed the convo Commented 4 hours ago
16

Reminder: in addition to wide layouts, test on narrow layouts. Many of us are using cell phones as browsers.

Pet peeve that I hope you can address: the pop-up menu that appears when text has been selected sometimes obscures the function bar at the top of the text entry box. This makes selecting text and then applying an operation to it, such as bold or hyperlink or quote, significantly more difficult than it has to be. I can work around by entering the markup manually, but this strikes me as something that could be done better.

Improve this answer
2
  • 1
    I'd pretty much say 720p (or lower) and UHD/4k or 5K on the high end would be good places to ensure good support. Commented 10 hours ago
  • Also, portrait or vertical monitors, or just windows that are taller than wide. Commented 33 mins ago
12

I have to say that I really don't understand the logic behind rebranding everything behind Stack Overflow name and I am mostly Stack Overflow user. For me Stack Exchange was very logical brand name encompassing whole network including Stack Overflow.

For non technical sites, Stack Overflow is meaningless. For technical sites, pushing Stack Overflow brand means that more users will be confused and instead of asking question on smaller, more appropriate site, they will end up asking off topic question on Stack Overflow.

When it comes to the actual sites redesign, I hope that you will not be using that dreadful colors and style you have shown us at your introductory post. Those are really hard on the eyes. Some of us will be able to use various user scripts to make it more usable, but we are only a fraction of your user base. Horrible design choices will not make sites more popular.

Improve this answer
3
  • 1
    The name changes are not in my purview. However, this post series does explore how we (product design) took the marketing assets and imagined them for our products. I'd recommend you check out those posts. Commented 8 hours ago
  • I note that many of us have spent the last 10+ years promoting "Stack Exchange" sites all over the Internet; those links are still functional, so hopefully they don't break. There are also issues with Meta.SE and Meta.SO being different things (and likewise with Math.SE and MathOverflow). Commented 4 hours ago
  • 1
    @Piper we keep hearing that. I'm going to "can we speak to the manager" here and maybe get the person whose purview its under to either post here on why its so massively important to do this, sit down with folks and try to find a solution that works better for the communities involved? We have a SO mod, and a SE/Network mod both pointing out this is a problem - so if you can't do anything, please get someone who can do something come out. Considering we represent affected communities, and are speaking on behalf of them, who would the rebranding be for then? Commented 2 hours ago
6

Is this actually going to be a full redesign, where we go, on day one, from what we have to a new design? or is this going to be more of a slow, over time roll out where now instead of two to three different UI ideas being maintained/in use across the site it's three to four?

Improve this answer
4
  • 4
    an xkcd 927 situation? Commented 10 hours ago
  • 5
    "We're gonna solve our tech debt issue by replacing the existing 2-3 not entirely rolled out site redesigns with a new redesign!" yea, exactly. We have the home page, which is entirely different from the rest of the site, we have the staging ground which was built at an earlier site design phase, we have articles which was built at yet another design phase, and... review queues... a partial in progress chat redesign kinda? Commented 10 hours ago
  • 5
    It won't be a full replacement on day one. My hope is that we can do a simple design system swap across the board and then the changes that live outside the design system will happen more slowly. Devs just finished discovery on how we might approach this so we are simply too early for me to say exactly how much will change with that initial push. Commented 10 hours ago
  • 1
    @Piper I'd be interested to hear about the results of that discovery: I'm all but certain they've missed some fundamentals, which any of a few members of the community could fill in for them. (And perhaps I'd be pleasantly surprised.) Commented 9 hours ago
4

Will users have the ability to opt for legacy layout/design once the redesign goes live?

Improve this answer
New contributor
UncleBigBay is a new contributor to this site. Take care in asking for clarification, commenting, and answering. Check out our Code of Conduct.
1
  • 3
    In the experiment phase, probably? I don’t think it's likely at full release though. Commented 6 hours ago

You must log in to answer this question.

Start asking to get answers

Find the answer to your question by asking.

Ask question

Explore related questions

See similar questions with these tags.