Restraint, Practiced Quietly, in the Overlooked Corners of Contemporary Design

There is a moment in almost every design review when someone asks for one more thing. A slightly bigger logo. A brighter accent colour. A line of copy that explains what the layout is already showing. The request is rarely unreasonable on its own. It is the fortieth version of the same request, arriving from forty different people over the life of a project, that turns a clean piece of work into something apologetic.

I have watched this happen from both sides of the table — first as a studio director in Lisbon, later as a writer who visits other people's studios to ask how they made their peace with it. The pattern I keep finding is not that confident designers say no more often. It is that they have a clearer sense of what a "yes" costs, and they spend that currency deliberately.

One practice worth borrowing: a running list, kept somewhere private rather than posted where a client might see it, of every element added to a project against a designer's recommendation. Next to each entry, a short note on what it cost — a broken grid, a colour that had to be repeated elsewhere to justify its presence, a week spent making an unnecessary feature look intentional. The list is not there to shame anyone. It exists because the pattern is invisible unless someone writes it down. Restraint erodes one small compromise at a time, and each individual compromise always sounds reasonable in the room where it gets approved.

The discipline nobody applauds

Restraint does not photograph well. A crowded homepage, a loud campaign, a poster with six typefaces fighting for attention — these get shared, screenshotted, held up as evidence of ambition. A page that says exactly what it needs to and stops does not travel the same way. Nobody reposts a caption for being appropriately short.

This creates a quiet asymmetry in how design work gets judged, especially early in a career. Add something and the addition is visible; you can point to it in a portfolio review. Remove something and the removal is invisible by design — the whole point was that nobody would notice it was ever there. Junior designers learn to pad decisions with justifications, because a bare "I left this out" sounds like an excuse rather than a decision.

The designers who eventually get comfortable with restraint tend to describe a specific turning point: the first project where a client complimented the simplicity of something they had fought hard to keep simple, without knowing what it had cost. That is usually the moment restraint stops feeling like an absence and starts feeling like a skill.

It also tends to be the moment a designer stops apologising for a decision in the room where it is presented. Early in a career, the instinct is to over-explain a restrained choice — to walk a client through every alternative that was considered and rejected, as if the emptiness of the final page needs to be earned in front of an audience. Designers who have made peace with restraint present the decision once, plainly, and let the work carry the argument. If it needs three more sentences to justify why something isn't there, the something probably needed to be there after all.

Restraint is not the absence of a decision. It is a decision that refuses to explain itself twice.

What restraint removes

It helps to be specific about what gets cut, because "less" is a word that promises more than it delivers. In practice, restraint tends to remove three things, and they are not interchangeable.

The first is redundancy — the same message said in a headline, a subheading, and an icon, as if the reader might miss it twice. The second is decoration that exists to fill a gap rather than to do work — a background pattern added because the layout felt empty, not because the pattern meant anything. The third, and the hardest to give up, is the safety net: the extra button, the additional navigation item, the fallback explanation, all included because someone worried the main path might not be enough on its own.

Cutting the first two is mostly a matter of taste. Cutting the third requires trust — trust that the primary path works, and that if it doesn't, the fix is to strengthen it rather than to build a detour around it. A studio that keeps its safety nets has usually stopped believing in its own main road.

Consider a fairly ordinary, representative example: a booking form for a small studio's client portal, with a single clear button to confirm an appointment. Somewhere along the way, someone worried that first-time visitors might not trust a single button, so a secondary "or call us instead" line was added beneath it, in smaller type, as reassurance. The phone line went on fielding calls from people who could plainly have booked online, because the smaller line had quietly told them the button might not be trustworthy. The safety net had created the very doubt it was meant to prevent. Removing it, and instead making the button impossible to misread, was less a simplification than a correction — the form had been undermining itself the whole time.

Where it tends to hide

Restraint shows up most clearly in the parts of a design system nobody browses on purpose: the spacing scale, the line-height defaults, the handful of type sizes that get reused across a hundred templates instead of a hundred bespoke choices.

A spacing scale with too many values is a symptom, not a feature. It usually means several people made slightly different decisions at different times and nobody went back to reconcile them. A restrained scale — four or five steps, used consistently — looks boring in a style guide and produces pages that feel considered without anyone being able to say exactly why.

Typography carries a similar tell. A page set in one well-chosen typeface, at a small number of weights and sizes, will almost always read more confidently than a page that reaches for a second or third face to signal variety. The variety was never the point. Contrast — of size, of weight, of spacing — does the same job with a fraction of the visual noise, and it ages better, because it depends on relationships rather than on a specific pair of fonts staying fashionable.

There is a version of this argument that turns into minimalism as an aesthetic, all white space and monochrome, and that is a different thing entirely. Restraint is not a look. A maximalist poster can be perfectly restrained if every colour and shape on it is doing something the poster genuinely needs; a minimalist one can be undisciplined if the emptiness is unexamined. The test is not how much is on the page. It is whether anything could be removed without the piece getting weaker.

Colour palettes tell a similar story, and they are usually where a restrained system first starts to fray. A brand guideline might specify three core colours and two accents, a sensible number for almost any use case. Then a product launch needs a colour that "feels different," and a partnership needs another, and within a year the palette has quietly doubled, with nobody able to say when the sixth or seventh colour stopped being an exception and became a fixture. Systems rarely collapse in one dramatic decision. They loosen the way a drawer full of cables loosens — one reasonable addition at a time, until retrieving anything specific means digging through everything else first.

The cost of getting it right

None of this is free, and the people who talk about restraint as though it were a shortcut are usually the ones who have never had to defend it in a meeting. Every element left out has to be left out on purpose, which means someone has to be able to explain why it isn't there — a harder conversation than explaining why something was added.

There is also a slower cost, paid over the life of a project rather than in a single meeting. Restrained systems require more maintenance discipline than loose ones, because the whole system depends on everyone continuing to work within the same small set of choices. One exception, quietly approved because a stakeholder insisted, becomes the precedent for the next exception. A system with five type sizes can survive a handful of these; a system that already has fifteen has nowhere further to fall, and nobody notices the difference.

This is why restraint tends to be easier to establish in a young studio than to introduce into an old one. It is not that established teams lack the taste for it. It is that undoing years of accumulated exceptions is slower and more political than starting with a clean rule and defending it from day one.

There is a practical middle path worth adopting deliberately, rather than waiting for a crisis to force the question. Instead of auditing an entire system in one exhausting pass, schedule a recurring, modest review — a few hours every quarter — whose only purpose is to look for exceptions that have quietly become the norm. Nothing needs to be removed on the spot. The point of the review is simply to notice, and noticing turns out to do most of the work. A team that knows its own exceptions are being counted tends to make fewer of them, in the same way a household spends differently once it starts keeping receipts.

None of this requires a formal design system or a large team to matter. A single freelancer working alone can keep the same discipline with nothing more than a personal rule — say, a fixed set of three type sizes for every project regardless of the brief — and the rule does the same job a committee would otherwise be needed for. Restraint scales down as easily as it scales up, which is part of why it is worth learning early rather than treating it as something only large studios can afford.

A decision that holds

The honest version of this argument has to admit that restraint can be taken too far. A page that removes everything eventually removes the thing the reader actually needed — the label that would have prevented confusion, the extra beat of explanation that turned a feature from confusing to obvious. Restraint practiced as a reflex, without judgment attached to it, becomes its own kind of noise: a studio style applied regardless of what the content requires.

The designers worth learning from are not the ones who cut the most. They are the ones who can tell you, specifically, why each remaining element earned its place — and who can tell you just as specifically what they left out, and why that particular thing was safe to lose. That second list is usually longer, and it is the one that never makes it into the portfolio.

It is worth saying plainly that restraint is not a virtue that applies equally to every brief. A festival poster and a hospital signage system are not asking for the same thing, and a designer who applies one fixed idea of "enough" to both will eventually get one of them wrong. What travels between the two is not a specific look but a habit of asking, before anything is added, what it would cost to take away later — and asking the same question, just as seriously, before anything is left out. The habit is unglamorous, mostly invisible, and it is very nearly the whole job.