Novelty is cheap. Restraint is expensive.
Every season the web grows a new set of tricks. Scroll-jacking. Cursor followers. Auto-playing video walls. Parallax stacked on parallax. They are easy to add and easy to admire in a portfolio reel. What they rarely are is necessary. Novelty is cheap because it asks nothing of you except the willingness to bolt one more thing onto the page.
Restraint is the opposite. It asks you to decide what the page is actually for, and then to remove everything that doesn't serve that purpose - including things you are proud of. That is a far harder discipline, and it is the one that separates work that photographs well from work that performs for a decade.
What restraint actually costs
The reason most websites are overbuilt is not bad taste. It is that addition is the path of least resistance. A stakeholder wants their feature seen, so it goes above the fold. A designer falls for an effect, so it ships. Each decision is locally reasonable and collectively ruinous. The page becomes a committee made visible.
Removing something requires a defensible point of view. You have to be able to say, in a room, why this carousel earns its place and that animation does not - and to hold that line when someone with authority disagrees. Restraint is expensive because it is paid for in conviction, not pixels.
We have learned to budget for that conversation. Every engagement begins with a single question we return to obsessively: what is the one thing this page must do? Everything that cannot defend itself against that question is cut, no matter how much we like it.
The performance dividend
Restraint is not only an aesthetic. It is a performance strategy with a measurable return. Every element you remove is a request not made, a script not parsed, a frame not dropped. The quiet page is almost always the fast page, and on the modern web speed is a feature users feel before they can name it.
A site that loads in under a second and responds to input without hesitation reads as expensive, even when the design itself is plain. A cluttered, janky page reads as cheap, even when the production budget was enormous. Users do not separate how a site looks from how it feels. Restraint is the rare decision that improves both at once.
This is why our web development practice treats the performance budget as a design constraint, not an afterthought. The frame rate is part of the brief.
Restraint is not minimalism
It is worth drawing a line here. Minimalism is a style - a particular vocabulary of white space, thin type, and muted palettes. Restraint is a method, and it can produce work that is loud, saturated, and maximal when the brief genuinely calls for it. The discipline is not in the result looking sparse. It is in every element being load-bearing.
A baroque page can be restrained if nothing in it is decorative for its own sake. A stark page can be self-indulgent if its emptiness is a pose. The question is never how much is on the screen. The question is whether each thing on the screen has a job.
How we practice it
Restraint is a habit, and habits need structure. We design in grayscale first, so that hierarchy has to work before colour is allowed to rescue it. We set a hard ceiling on the number of typefaces, weights, and motion patterns a project may use, and we spend that budget deliberately. We review every interaction against a simple test: if we removed it, would anyone be worse off? If the honest answer is no, it goes.
None of this is glamorous. It is the slow, unphotogenic work of subtraction, and it almost never makes it into the case study. But it is the reason our work still feels considered years after launch, while louder sites from the same season already look dated.
If you are weighing a project and suspect it has been quietly overbuilt, that instinct is usually right. Start a conversation and we will tell you what we would take away.