Zone Home
0

Healthy Evolution Is Design Drift Somebody Decided On

Put two website changes side by side, one that made a site better and one that started it drifting, and most of the time you couldn’t tell which was which. Both show up as a new color or a section that doesn’t match anything around it.

In January I put a pixelated, retro hero section at the top of my own homepage, and nothing else on classiccity.com looked anything like it. It pulls from old Mario and Zelda games (on purpose, I promise), and if a client’s marketing team asked me about doing something like that, I would have tossed up some red flags.

It was fine on mine because I meant it. I called it out on the homepage as a glitch in the system, I wrote about it, and I could give the reason in one sentence: give people a reason to click over from a LinkedIn post and see what the weird thing is.

There’s a line I want to draw in this post: I’m all about consistency, but when does it make sense to break the mold? The difference between healthy website evolution and design drift is intention. A change counts as evolution when someone chose it on purpose, can name the reason, and folds it back into the block library so the next page can use it (or pulls it back out when it doesn’t work). And a site that changes that way can keep changing for years without ever hitting the redesign everybody budgets for every two to three years.

The same change can go either way

Drift can happen on any platform (WordPress, Squarespace, Wix, Webflow, it doesn’t matter), because every one of them lets you change font colors and sizes across the board. It almost always starts small. Somebody wants a little extra emphasis on one line and bumps the font size up a notch. Somebody else needs a section to stand out and gives it a slightly different color. A new page gets built by someone who doesn’t know how the image cutouts were made on every other page, and every image on that one ends up a plain square.

Every one of those changes has a sensible story behind it. The people making them aren’t doing anything wrong (managing a website shouldn’t require web experience at all), but nobody chose the result. The phrase I listen for is “Is that good enough?” When someone looks at a page and asks that, it usually means the page has wandered away from the brand and they can feel it. Give that 12 to 24 months and you can tell which pages the professionals built and which ones the admins built, until people don’t even know what page they’re on anymore. (I wrote about why that’s an operations problem more than a design one in design drift is an operations problem.)

Evolution works from the exact same raw material. I’m 100% open to trying something new on a site. Go for it, try it out, and if it works, add it to the block library so everybody can use it. I push back when the new idea goes against the brand standards because it’s what the person building the page felt like making that day. If there isn’t an intentional reason to be wildly different, it’s just noise, and a style guide sitting in a shared drive won’t catch it for you (more on that in why your style guide didn’t stop the drift).

A lot of brand advice treats every difference from the guidelines as a violation. By that standard my homepage fails, and so does any ad campaign built to surprise people. The test I use is simpler and a lot more useful: did somebody decide this?

The test I use is simpler and a lot more useful: did somebody decide this?

Why my retro hero passes

When I wrote why I’m breaking my own website, the plan was simple: “I am going to run 24 micro-experiments on this website for the next year.” The retro hero was experiment number one, and on the homepage I point out that there’s a glitch happening to explain why the top of the page looks so different from the rest.

The hard part (and it’s the part I think about most) is keeping it from being so weird that a bystander who wanders in gets freaked out. I have to tie it back to everything else somehow. And the experiment only stays healthy if it goes somewhere. A few months later I wrote my website is a video game, and the first line about that hero is this: “The aesthetic is now everywhere. The retro hero from Experiment #1 used to be stand alone (nothing else on the website looked like it).” The glitch earned its way into the system, and now it’s part of the brand (heck, it IS the brand).

One caveat, because I don’t want anyone reading this as permission to redesign one random spot and pray it converts. For most companies that’s the wrong move. Personally, I treat my site like a playground and its a big part of my marketing, so I can afford to experiment on it in public. You probably can’t afford to take a big swing on that service page that is driving a bulk of your good leads. Where the same thinking fits almost everyone is an ad campaign or an event, when you want people’s first thought to be “Oh, that doesn’t look like anything they’ve done before, that’s interesting.” I’m a big fan of that. The surprise helps, and if you can keep enough of your core brand in there that people still know it’s you, even better.

Four questions that sort any change

I’d run every new section or page style through these before it goes live:

1. Did someone choose this?

If the change showed up because a page had to get built by Friday, nobody did.

2. Can you name the reason in a sentence?

“It’s for the spring ad campaign” counts. “It looked a little flat” doesn’t.

3. If it worked, does it go into the block library?

Try it out, and if it works, it gets a name and a spot in the library so the next person can grab it. If it didn’t, it comes back out. Some one-offs are meant to stay one-offs (a campaign page, or my own retro hero), and that’s fine as long as somebody chose it.

4. Did the structure survive?

Your page layouts and your primary call-to-action color should come through the experiment the same, even if the imagery, the mood or the palette shifts a little.

That last one is easier to see with an example. We’re designing for a client with about five different product lines, and each one targets a very different customer (one is very hardware oriented, one is very beachy, one is all about the mountains). On their existing site, each product’s pages use a different font to feel more on brand for that customer, and I get the reasoning. What we’re building lets the imagery, the iconography and even the base color shift a little by product line, inside the brand guidelines, while the layouts and structures stay exactly the same, and orange is the CTA color on every single page. Because the variation is designed into the system, every product page that uses it is evolution from day one.

There’s a tradeoff sitting underneath all of this. Lock a site down so tightly that all anyone can do is fill out a form, and people start wanting more flexibility. Open it all the way up, and the more flexibility you give, the less consistency you’re likely to keep. Intention is what lets a site sit further toward the flexible end of that line without paying for it in drift.

Somebody has to own the call

Intention needs an owner. In a small company, I’d have one person act as the gatekeeper for the website, making sure the brand, the messaging structure and the design principles are followed. That person also knows when it’s intentionally a good time to break the rules. (I get into who that person should be in who should edit your company website, and into what stays locked for everyone else in how to let your marketing team edit the site without breaking it.)

You can see what happens when those calls get made page by page on a university site we’ve worked on for years, one with more than 3,000 pages. Different departments kept asking for one-off pages that looked different from everything else, for small groups of users that didn’t matter much to the goal of getting new students in the door, and the feedback on the on-brand versions kept coming back as “not high quality enough.” Meanwhile, some team members building new pages kept pulling blocks from the archive that were 5+ years old. Every one of those decisions made sense to the person making it, and none of them came with a reason anyone could defend to the rest of the site. That’s the drift pattern, and it happens right under people’s noses no matter how well a site is built.

Nine years on one site

The best proof I have that intention changes the lifespan of a website is a client we built for in 2015.

This was back before we had a team, and before most of the principles and structures we use now existed. The phrase “design drift” wasn’t even in my vocabulary yet. WordPress didn’t give us saved patterns the way it does now, so when the site launched we taught their team the basics: how to add the buttons, how to add a two-column block, how to add that image texture to the side.

That website lasted through 2024. Over those years we kept adding building blocks to their library as they needed them, and sure, some of the newer ones looked better than the older ones, but none of them looked disconnected. We kept handing them more tools for the toolbox, and the site kept working for them. Same brand, same look, different people managing it along the way.

What finally forced a rebuild was speed. They had so much content and a big marketing push coming, and they needed to make a lot of pages quickly, and the old system just took too long. It had also picked up a fair amount of duct tape around an e-commerce integration with a custom system they ran. The site was about 100 pages, and at that size rebuilding the whole thing at once made sense.

When we redesigned it, we didn’t scrap much. We rebuilt every block from scratch in WordPress’s Gutenberg editor and gave them a new block library with saved patterns, but the design moved only about 20% from their last version of the old site to the first version of the new one. Their customers didn’t have to relearn anything, because everything was pretty much where they left it, just a little nicer. They didn’t need a huge overhaul, and they didn’t get one.

Compare that with the answer you’ll get almost anywhere else. Google’s AI Overview for “how often should you redesign your website” opens with it: “Most companies redesign their websites every 2 to 3 years.” That matches what I see for companies using their site hard without a system behind it. They use the site, they don’t use it well, it drifts, and it becomes a pain in the butt to manage. The 2015 site kept changing for nine years without drifting, because the changes came in as new blocks somebody decided to add.

If drift has already taken over your whole site, a redesign is usually the honest answer (patching a site that disconnected rarely pays off), and I’ve laid out what that costs on our website redesign cost page. Our project pricing is public too.

If you’re not there yet, there’s a small exercise you can run this week. Open your site and find the one section that looks the least like everything else. Run it through the four questions. If it passes, give it a name and add it to your block library (if you don’t have one yet, the order of operations that keeps a new page on brand walks through starting one). If it doesn’t pass, pull it. And if you’ve got a weird section on your site that you put there on purpose, with a reason, good for you. Keep it. When AI has everything online looking buttoned up in the exact same way, being a little bit unbuttoned shows there’s a human behind the site.

If you’d like a second set of eyes on whether your site has been evolving or drifting, book a call and we’ll look at it together.

Ready to chat? Let's plan your next website. Grab 30 minutes with Chris LaFay of Classic City.

    Want to dive deep into your site? I’d love to chat.