Open fifty startup homepages in fifty tabs. Squint. They blur into one page: a geometric sans-serif wordmark, a soft gradient, a rounded-rectangle screenshot floating on a faint grid, a headline promising to reimagine something mundane. The logos that took agencies six months and a quarter-million dollars are now indistinguishable from the ones a founder generated in an afternoon.
This is not a coincidence, and it is not laziness. It is the predictable output of a system that optimised away every source of friction that used to make things look different.
Friction was the feature
Distinctiveness is expensive. A custom typeface costs money. A weird grid takes arguing for. A colour that isn’t “accessible-blue” gets flagged in review. Every one of those decisions is a small fight, and every fight is a place where a brand could have become itself.
The tools got so good at removing friction that they removed the thing friction was protecting: the decision.
When the default is excellent, you stop choosing. The component library ships with a perfectly serviceable button, so you use the button. The template comes with a hero section, so you write a headline to fit the hero section. The font pairing is “tasteful,” so you never audition the font that would have been yours. A thousand reasonable defaults compound into a single aesthetic.
The homogenising forces
It helps to name them, because each one feels harmless alone:
- Design systems reward consistency within a product and, as a side effect, consistency across all products that copied the same system.
- The marketplace of templates means the global minimum-viable homepage is a click away — and the click is free.
- A/B testing punishes the unfamiliar. The variant that tests best is usually the one users have seen a hundred times elsewhere.
- Trend aggregation — the dribbblisation of taste — turns whatever is popular this quarter into next quarter’s baseline expectation.
None of these are villains. Together they form a current, and most teams are not swimming against a current for a button.
Sameness has a cost, just not an obvious one
The argument for the flattening is real: familiar interfaces are learnable, accessible patterns are accessible because they’re patterns, and a small team should not be reinventing the date picker. Agreed. Spend your weirdness budget where it counts, not on the form controls.
But brands are memory machines, and memory needs an edge to catch on. When the wordmark, the palette, and the layout are interchangeable, the only thing left to remember is the name — and names are cheap. You have traded the most durable asset a brand can own, recognition, for the comfort of looking correct.
Reintroduce friction on purpose
You don’t fix this with more taste. You fix it by deliberately re-inserting the decisions the tools tried to make for you:
- Pick one thing to be strange about. One typeface, one ritual interaction, one colour nobody else would dare. Defend it in every review.
- Audit your defaults. For each component you accepted, ask whether you chose it or merely allowed it. Allowing is not designing.
- Design the thing that doesn’t screenshot well. Motion, sound, copy voice,
the feel of a transition — the parts a competitor can’t lift in a
⌘⇧4.
The great flattening isn’t a failure of imagination. It’s what happens when imagination becomes optional. The work now is to make it mandatory again — to spend the friction, on purpose, in the one place it will be remembered.