I maybe struck a nerve in Mastodon when mentioning that I think the djangoproject.com homepage would benefit from a fresh redesign. So, to broaden the request beyond the fediverse, I decided to also post here.

After a mention from @wsvincent that it might be useful to have something for people to react to, I threw together some HTML and Tailwind CSS this morning:

To be clear, this is just a first draft after a few hours. There is a whole page with lots more included, but I don’t want to give people the wrong idea.

If anyone is interested in working on the homepage, please reach out to me with a GitHub username and I can add you to a private repo where I’m experimenting. Or if you want to see a deployed version, DM me and I will send it to you.

I have thoughts about what would be useful and some particular goals I’d like for the homepage, but I’m sure there are others who do this for a living, e.g. designer/UX/marketer/DevRel, and have better approaches than I do.

Some unknowns upfront:

  • I do not know the process to make a redesign official or deploy it – I’m mostly just trying to push the ball forward and then going to figure out the next steps later
  • Redesigning the whole site, i.e. blog, news, docs, etc is out of scope (for now)

I look forward to hearing your ideas and feedback!

18 Likes

Thanks for getting the ball rolling here!

I’ve been following the conversation on Mastodon, and really agree strongly with the goals expressed over there (which I’d roughly paraphrase as: “make it more obvious what Django is and what it can do for you”).

One bit of prior art you might want to look at: about a year ago, 20tab did some user research and made some recommendations; you can find their report here: djangoproject.com – 20tab deliverables – Google Drive. I think you’ll find some stuff in there that supports your line of thought.

Mostly though I wanted to speak to this:

I do not know the process to make a redesign official or deploy it – I’m mostly just trying to push the ball forward and then going to figure out the next steps later

You have a few options here, but I think your best bet here is to take the design pretty far – like “early beta” stage – and then present it to the Board for feedback and approval. I can’t speak unilaterally for the Board, but I really think if you show up with a clear vision, and something that’s a concrete improvement of the status quo, there’s a very high chance you’ll get an enthusiastic “go ahead”.

This is obviously risky: I’m asking that you put in work (possibly lots of work) with a possibility of it going to waste. But the problem is that without a clear vision, it’s going to be really hard to get a preemptive OK. The alternative would be to form a Working Group with a clear charter to execute a redesign, but the downside there is that’s probably a month or more of “paperwork” getting the group formed, writing the charter, getting a Board vote on the WG, etc. That’s some serious stop energy and I don’t want to tamp down the enthusiasm building here.

(Though: if you want to go the WG direction, now or in the future, I’m happy to support that, let me know.)

One final suggestion: I’m of the firm belief that design by committee doesn’t work. The parts of Django that have the best design were all designed by one or two, maaaybe 3 people. (I was thinking about aesthetics as I wrote this, but in retrospect the same actually applies to a lot of the best API designs in Django too!). You’re going to get lots of feedback, and while I do hope you’ll listen to it, ultimately I think you’ll be the most successful if there’s a very small core group make clear and consistent decisions.

Good luck! I’m super excited to see what comes out of this.

9 Likes

I really appreciate this response, your sentiments, and especially the link to prior art from 20tab (that I had never seen before).

I’m asking that you put in work (possibly lots of work) with a possibility of it going to waste

Personally, I’m ok with that as long since there is a possibility it won’t be a complete waste of time. I’ll keep this thread updated with where this ends up one way or the other. :grinning_face:

3 Likes

I think this is great @adamghill ! Already it’s definitely an upgrade over the existing one, which is old, so nothing against its design that has held up for so long.

I really like how you have a quick start on there to emphasize how quick it is to get started. I think I’m probably too close to this, since I worked with @carltongibson and @pauloxnet to help with the 20tab work @jacobian mentioned, but I’ll just add here my own list of things to consider, as well:

  • Sponsor logos vs companies using it? Every other framework does this. Something to help the fundraising groups etc since companies, beyond the goodness in their hearts, their marketing department wants to know, if we donate $100k does that mean XYZ million views from the homepage?
  • Related and something @jacobian and I discussed with others at DjangoCon US was that we need some kind of anonymized, non-US-based analytics for djangoproject.com so we can both give numbers to sponsors and also have some internal visibility into which pages get traffic and which do not. For example, the new Ecosystem page is wonderful, but hard to find.
  • Love how you have APIs on the homepage. I think we could also add CRM (Wagtail, Django-CMS). And in my perfect world, “something” around AI and how Django fits with it, though that’s a reach and more than likely want to chew off.

Anyway, apologies for the brain dump. I find this super exciting. And I think the way forward is for you and a small (very small) group to come up with your best version and then we all can go from there!

5 Likes

Sponsor logos vs companies using it

I kind of think we need both, although I’m open to prioritizing one over the other. One is social proof and the other is a reason to sponsor the project – both are important and they fulfill different needs.

anonymized, non-US-based analytics

I currently use Plausible which would fit the bill, I think. The other major player in this space is Fathom. There are self-hosted platforms like Matomo or umami, too if someone really wanted to deal with the infrastructure, and/or parse nginx logs to at least get some insights. But, if it were up to me, I would reach out to both Plausible and Fathom and at least get a quote. Maybe there is a discount for major, important OSS projects. :grinning_face:

I think we could also add CRM

Agreed! CRMs are already mentioned further down the page, but it wasn’t included in the screenshot unfortunately.

“something” around AI and how Django fits with it,

I did not think about this at all, but you are completely correct. Will add it to the GH issues. :smiley:

1 Like

I agree, I think there’s an important distinction between the two, and both are important to showcase.

there are a lot more concerns to think about here! @thibaudcolas and i had a long chat about this at DCUS25. there has already been ux research and accessibility audits. also brand is extremely important here. great to see the conversation being started though, and content strategy at the forefront as well.

id be concerned about a homepage-only redesign feeling out of place with the rest of the site.

1 Like

focusing on updating and adding content with the existing site will go a long way, and help alleviate concerns about how the framework is presented and understood without needing a redesign

I think it’s especially nice to highlight things we can’t really expect funding from. It’s nice for everyone to se NASA here I think (wow, NASA uses Django it’s quite powerful I think).

Great initiative. It’s been on mine and others’ minds for some time already, so it’s nice to see something a little more concrete. I think we definitely need to think holistically about structure, design across the whole site, messaging, relative importance and hierarchies of information. I think at the moment we might have a bit too much “stuff” and not great navigation to this stuff. A bit of simplification, consolidation, making sure the important things are front and foremost. This will go a long way. But it’s not just UI design, it’s the whole user journey type stuff and this is not only pretty hard but requires a lot of work.

Sorry, I don’t have many concrete thoughts right now or anything that’s really immediately useful. Definitely interested in pushing this forwards as I don’t think we really seel the framework well enough, either to developers or to prospective sponsors.

I think the UX research is linked above from @jacobian – I’ve been reading through that and will try to incorporate as much of it as I can.

id be concerned about a homepage-only redesign feeling out of place with the rest of the site.

I’m also a little concerned about this. I’m hoping if at least the top-nav matches it might be ok, but definitely something to keep in mind.

Let me know if you want access repo where I’m actually experimenting with different approaches!

I think updating/adding/editing content will also be very useful. But, in my opinion, it doesn’t alleviate the need to also update the homepage. The homepage is the first interaction that newcomers have with Django. Some have no idea what it can do, its capabilities, the scale it handles, its stability, etc., etc. I think it’s crucial that their first impression provides them quick hits of information and lets them know how great the framework is. Then, link them to the fantastic documentation where they can get more details.

coming from many years of ui/ux/front-end experience: the actual css/visuals of the page don’t matter until you have a solid foundation for content on the page/content strategy.

I think it’s crucial that their first impression provides them quick hits of information and lets them know how great the framework is.

This can and should be done outside of/before any visual work, ie: a collaborative document

3 Likes

it’s the whole user journey type stuff and this is not only pretty hard but requires a lot of work.

Agreed and to be honest, that is not really my skillset. I’d love help figuring that out.

However, I have seen a variant of “It’s been on mine and others’ minds for some time already” multiple times the past few days. Lots of people are thinking about this which is great! It’d be a shame to not figure out a way to make these types of improvements even when it seems a lot of people think they are necessary.

There are a constant stream of improvements and PRs to the framework – I think we should also be iterating, improving, and enhancing the “marketing” of the framework as well.

1 Like

Happy to read any previous research, collaborate on, or participate in a process. Let me know what the next steps would be. :+1:

:waving_hand: just for context, echoing what @sodevious mentioned, we’ve had a few months’ worth of discussions about setting up a marketing working group (proposal as a private draft currently), and doing a website redesign. It’s been a topic for years, but internally we’re working on a brief based on the needs of our fundraising WG, the 20tab user research, the reports of our accessibility team. We have a website team up and running as of a few months ago that is gearing up to ship this new design once it exists. And we have draft content for specific areas of the site (fundraising related).

@adamghill if you’re doing competitors’ research by reviewing other frameworks’ sites, I think that would be most useful for us at this stage, so we can use that when creating our brief? Share a list of (relevant) competitors, why they’re relevant, what factors matter in researching them, and results of that research.

And if you’d like to work on this longer-term I’d recommend helping us get this marketing group started? I can share our draft proposal for your feedback?

Also don’t want to tamper enthusiasm but keep in mind there are valid reasons why those things take time! Here are the gaps we’ve identified so far in actually making a redesign happen:

  • Marketing strategy
  • (Possibly renewing the vision for the framework a bit)
  • (Brand guidelines)
  • :white_check_mark: User research
  • :hourglass_done: Redesign brief
  • (Website content strategy)
  • (Website moodboard)
  • User journey mapping / information architecture for the site
  • (Data on current site usage)
  • Design mockups, possibly low-fidelity prototypes
  • Visual design

Some might look at this list and think it’s overkill. I would personally say for an organization of our scale and ambitions it’s a must to go about this the proper way. With almost none of the above having been done over the last 10 years, we’re really suffering from this lack of shared direction. The homepage is a tempting target but it’s really just a symptom of overall lack of marketing / design governance. And I’d say lack of a shared vision for what Django is about in 2025+. Not like the current vision is bad but it’s definitely dated, and hard to interpret for marketing purposes.

5 Likes

Sure, where would you like me to share my current research?

I wouldn’t call it overkill, but (for better or worse) my bias is toward action. If this information was already collected and ready, I’d say let’s use it! Since it’s not, I’m going to make my best guess at what I think solves real pain points.

As mentioned multiple times now, I’m happy to participate in a process if there is one. But, I don’t think that means I shouldn’t work on things (on my own time) I’m passionate about. :slight_smile:

really just a symptom of overall lack of marketing / design governance. And I’d say lack of a shared vision for what Django is about in 2025+

Agreed. I’d love to help with this.

In the meantime, I think there are obvious changes to improve the website (which can then be iterated on!), however the homepage has basically been stuck with the same design for 10+ years (since at least 2014: https://web.archive.org/web/20141231140629/https://www.djangoproject.com/).

I just think it’s way past time for a refresh and for Django to make a better first impression.

2 Likes

@thibaudcolas why can’t we do both? Ship something smaller sooner and use that to galvanize the rest of the team(s) to do the larger effort behind it. I don’t think it’d be horrible to have three different home pages over three years.

Nobody is going to argue there aren’t biggest issues at play here, but they aren’t well known / well communicated.

As a community member, for this topic, I’m more on the side of “let’s try something” rather than “let’s do everything properly”

4 Likes

People not understanding what the framework can fully do is a user experience problem. Not a tailwind/css/html problem.

Ask any UX person, and they will tell you some form of the following; The very first first first things that need to be done before any code or Figma touched is to:

  • Identify goals marketing the framework
  • Outline what the homepage will say, how it will speak to those goals
  • Prioritize goals/content. What is the #1 thing you want people to see/know?
  • Determine what works about the existing homepage/site and what doesn’t

Want a refresh? Start with that list.

Like @CodenameTim said, ship small. Make content tweaks and solve some obvious problems. For example, there is no clear link or way to get information on contributing to the djangoproject.com site. Making repairs should not involve an entirely new visual UI/web design.

And to @thibaudcolas point, yes that is absolutely the correct process. The website has existed for decades and will continue on for more decades. It’s an investment.

Both “let’s try something” and “let’s do everything properly” can exist at the same time. Motivated folks can focus on identifying quick wins, identifying pain points, creating issues/tasks from the existing UX research, etc. while the “proper” workflow is being executed behind the scenes. Results/learnings from both types of tasks can inform the each other.

6 Likes

yeah go for it if you want to try quick homepage wins! I’m not saying no to that, just trying to make you all aware of our current plans, challenges, and all the aspects that will make it difficult.

there are obvious changes to improve the website

As a community member, for this topic, I’m more on the side of “let’s try something” rather than “let’s do everything properly”

Yes there are obvious changes needed :slight_smile: And they’ve been obvious for years for people who do marketing. But please don’t get the impression that nothing is happening or that we’re trying to do things properly at the expense of progress. We’ve been working on this behind the scenes for a while now, but fact of the matter is our priorities as far as the website currently are:

  1. Fundraising to $300-350k to sustain current needs of Django
  2. Fundraising to $500k so we can afford our short-term ambitions for the project
  3. Information architecture / UX / accessibility improvements (new IA, new color theme)

So the plans we’ve had for iterative improvements to the site are in those areas, via content and restructuring, not a new coat of paint just for the homepage? If we can do all of that at the same time great. Personally I’d rather our contributors helped us with the above goals that have been agreed by our Board. If we can meet those goals, and we can redesign the homepage in that direction at the same time, great.

Anyway, tying things together, if you want to proceed with iterative homepage improvements I’d recommend to:

  • Integrate some of the content (language, design elements, graphics) from our 2024 annual impact report. They’ve all been extensively reviewed by our fundraising / marketing / Board people, and are well suited for reuse across the site. Next step is we’re working on a “sponsorship prospectus” with a similar web-first approach (HTML, Tailwind) and content that’s even more refined.
  • Make room for a “Django 6.0 features highlights” marketing video on the homepage. That’s been tentatively approved by the Board (working on budget & procurement).

From my side, @sodevious has kindly offered to help us with the process in the “proper way”, and we really need more people with UX / design / content strategy in the long run, so will keep working on that

2 Likes

I agree that there is definitely a need for redesign, and I remember this was one of the goals of the Website WG when we reach the implementation phase. I have had discussion with @thibaudcolas regarding the user research and accessibility issues that need to be fixed.

People not understanding what the framework can fully do is a user experience problem. Not a tailwind/css/html problem.

@sodevious I 100% agree with you. As someone who is active in Open Source design communities, a design should always be guided by usability improvement and not just changing the visual look. And the actual implementation of the design comes much later.

As a community member, for this topic, I’m more on the side of “let’s try something” rather than “let’s do everything properly”

@CodenameTim The reason I agree with @thibaudcolas on doing this the proper way are few different factors:

  • We need the redesign to be guided by the findings of user research instead of just trying to change the visual look of the website. Hence, I think we should start at a design mockup and roadmap stage
  • We need some branding/color schemes updates. Although this might not be the reason of making the website look outdated, it’s a reason for a lot of pain points such as dark modes. And if not done carefully, we risk having the same issues again after redesign.
  • We need the website to be accessible. I think it won’t be acceptable if in the process of “let’s try something” we end up introducing accessibility regressions instead of solving them.

All this is to say that the process needs to be iterative for sure, but maybe it needs to start at a content architecture, and UX mockup/roadmap stage.

For smaller improvements/fixes, we can always make smaller fixes and not wait for an entire redesign.

From my side, @sodevious has kindly offered to help us with the process in the “proper way”, and we really need more people with UX / design / content strategy in the long run, so will keep working on that

Glad to hear that @thibaudcolas . Looking forward to it! We definitely need more UX/design persons.

1 Like