Brand drift usually isn't a design failure — it's a systems one. Why consistency comes down to workflow and governance, not another redesign.
A home has upwards of 30 billion problems, give or take. I know this because we're currently researching a kitchen update, and part of that research is taking a long look at every nook and cranny of our existing kitchen. It has since extended into the rest of the house — the back patio, the front porch lights, even down to the toilet seats. And while our home is (for the most part!) holding up rather well, I can tell you there still seem to be roughly 30 billion problems remaining. (Give or take.)
A lot of these "problems" aren't really "problems" at all — they're just ... differences. When this house was built, everything was standard — the contractors used the same brands of paint, installed similar fixtures, and generally created a cohesive home because it was built all at once, to one plan. And for a long time, that plan held. As far as we can tell, the cupboard and flooring and paint decisions all stayed relatively close to the original. We've seen pictures of the house before we bought it, and it looks pretty close to what we have today.
But that's the far-away view. Look closer and the seams start to show: patches of paint that are the right color but the wrong sheen, a hole here and there from now-removed surround sound speakers, a bare bulb where a standard fixture used to live. Even the light bulbs themselves aren't standardized anymore.
This isn't a major problem. It's just reality! Contractors and suppliers don't stay in business at the whim of any one house — places close, or get expensive, and a general replacement goes in where the original used to be. From across the room, the house still feels cohesive. But when you look close — and when you're living in the middle of it, day after day — you start to see where the original design has fallen into a patchwork.
Anyway, let's talk about design systems.
Drift happens over hundreds of defensible choices.
At some point in the past, your site was the freshly built house. You got everyone hyped up for a rebrand — a new identity, with a refreshed site and hot new design language. The homepage, services, and articles all looked like a small part of a cohesive whole. It looked like one company built it, because (usually) one team did build it, all at once, to one plan.
Then it started to sprawl. To be clear: no one decided to make it sprawl. It just, like, sort of happened one justifiable and reasonable decision at a time.
Your experience might vary, but it could have gone like this: a campaign needs a landing page by Friday, and the person building it needs a button, and the original button wasn’t easy to find, or doesn't quite do what this page needs — so they built a new one. It's a good button! It ships on time. The campaign works.
Then, three months later, someone else needs a landing page. They look at the most recent one for reference — the one with the new button — and match it. Now there are multiple copies of TWO versions of that button in the wild. What’s more, the newer one is winning, because it's the one people keep copying.
Multiply that across a couple of years, then complicate it with a few staff changes. Make the deadlines a little shorter, and maybe swap out the original design partner. Every one of these decisions was defensible in the moment — the same way a general replacement fixture makes sense when the original is discontinued. But the overall sum is a site that now looks like it’s been pieced together using three very similar sites. Our design system has become less of a system and more of a collection of exceptions.
Why a rebrand won't fix design system drift.
Design system drift is commonly positioned as a design issue, which means the most logical solution is to redesign the site to bring everything back together. And, yes — this is often what needs to be done, in some sense.
But a rebrand is a gut renovation. It replaces what the site looks like today, top to bottom, and for a while everything matches again. Yet, it doesn't change the forces that produced the drift in the first place. It does not change the deadlines, and it does not change the staff turnover. While the new identity will be cleaner and more current than the old one, it will eventually start eroding, just like last time, for the same reasons, until several years later you find yourself in exactly the same position as before.L
In other words, a new design system isn’t a fix as much as it’s an extension on the original problem, which is most likely tied to either a workflow system or a governance system. It’s systems all the way down, and the challenge is figuring out which one needs tweaking.
A good design system needs elements on-hand.
Let’s start with the workflow system.
When noticing all of the flaws in our house, we realized that several of the updates were not a taste issue or a quality issue: they were supply issues. The previous owner couldn’t find the right paint, or couldn’t locate a matching drawer pull. If the right paint had still been sitting in the garage, or a few extra pulls in a drawer, we’d never know there was a replacement.
Of course, this is unrealistic — no homeowner actually purchases a “back-up” of every part in the house. But with a design system, that’s EXACTLY what we’re going for: it’s not a guidelines document as much as it’s a set of actual parts, already built and tested, sitting where you’d naturally reach them.
Take the button. In a site experiencing design system drift, the "right" button might live in a brand guidelines Figma doc with hex codes and pixel values — helpful, to a point. But it puts the onus of reproducing that button correctly on the front-end developer, every single time. Reproducing it by hand is work; building a fresh one that looks close enough is often less work, so that's what happens. In a systematized site, the right button is a component — already built and already correct — so using the approved button is less work than making a new one.
The workflow is key to putting these pieces in the right spot and making them available at the right time.
A good design system has an owner and champion.
A garage full of paint cans with labels is only as good as the ongoing organization of those paint cans. A garage does not stay organized on its own; it barely stays organized by those who love organizing garages. It takes a ton of work to keep up, and at any moment someone might drop a paint can on the wrong shelf, or toss two similar colors next to each other, and now you’re hunting through the clutter again. Someone has to keep it true.
This would sound like an exhausted metaphor if it wasn’t the actual truth. Getting organized is not the same as being organized. This is governance: someone owning the system so it stays current, and championing it so people actually use it.
Change is going to happen. Someone’s gonna need a new button, even if your current buttons are really, really great. And this is okay! Good systems flex and adapt to the needs of the project. BUT, if that new button never comes back to be folded into the larger design system, you're back to two versions in the wild.
That's governance’s job. It’s not a committee that meets twice a year, but an actual commitment (of time, of budget, of attention) to folding in new patterns when a real need shows up, so people have a way to add to the system instead of building around it. It might be a single person, or it might be a design steering group; it does not matter who, as long as the actual ownership is intact.
Add to this one more thing: ownership is only half of the battle. A governed system means nothing if no one uses it, which means the owner must also be a champion. Someone has to make the case for keeping the buttons in order, and someone has to ensure the deadline-driven designers are using the in-order buttons, and someone has to teach the new hire what all of the buttons even mean.
A design system is only the path of least resistance if people know it's there and trust it'll have what they need. Where your workflow system makes the right choice easy, governance keeps that choice viable now and into the future.
What to ask when design drifts again.
Do not worry. Actual creative design still matters! It’s just that design does not necessarily always mean reinvention. The raw act of blue-sky design is not the only answer to a design problem.
The next time your design system starts to sprawl — and it will, because sites are living things and drift is the natural state — resist the urge to start over. Instead, ask a more useful question: what caused it to end up like this? What guardrails can we put in place so the right choice is easier and more palatable than free-range design?
And remember that the places we live don’t stay beautiful because they were built or designed well, but because someone maintains them. Design systems are no different, which makes it so important to keep the garage in shape and make sure someone’s in charge of keeping it that way.
