We forgot what design was for.

For most of the last century, design served a question: what would make this object better to live with? Then somewhere between Helvetica and the App Store, the question quietly changed. This is a working theory about what changed, and what we might do about it.

Mudit Lal

Edited by Mridula Iyer · Published 23 May 2026

9 min read

he new question, the one that quietly replaced the old one, is a question of conversion. Every pixel earns its keep by moving a metric. The button gets bigger because the metric demands it; the colour gets louder because the metric demands it; the copy gets shorter because attention is short and the metric is even shorter. The thinking happens at the metric, not at the user, and the object that results is not designed so much as optimised.

Optimisation and design look similar from the outside. Both involve careful people making careful decisions about size, weight, colour, position. Both produce artefacts that are measurably better than the artefacts they replace. The difference is what they are better at. Optimisation makes things better at the metric. Design makes things better to live with. These are not the same thing, and the longer we pretend they are, the worse our objects get.

“The most useful design questions are still the oldest ones. They just stopped being asked, because they stopped being measurable.”

From the cover essay, Vol. iv

What the old question used to do

The old question, the one about living-with, was wider than the new question. It swept up considerations the metric will never see: how the object feels on the third day, how it ages, how it sits in a room beside other objects, how a stranger reads it, how a tired person reads it, how a kind person would have made it. None of these are anti-metric. Most of them, given enough time, do produce metrics. They are just not the metrics you can ship a release against.

So they fall out of the brief, quietly, the way an unloved photograph falls out of an album. Nobody removes them. They are simply not invited to the next meeting, and the next meeting, and the next. After a while the brief is just the metric, and the metric is just the brief, and the people in the room have forgotten there was ever a difference.

What this means, in practice

Write the question down before you write the metric down. Keep the question visible during reviews. If the metric ever contradicts the question, the question wins, and the metric goes back to being a tool. This sounds obvious; it is not, in any team I have worked on, common.

I do not think this is anybody’s fault, exactly. Optimisation is honest work. The people doing it are usually kinder, smarter, and more diligent than the systems they are inside. But systems are heavier than people, and a system that only measures one thing will, given enough time, only produce the one thing. The corrective is not to throw the metric out. The corrective is to put the older question back where it can be seen, in the room, on the wall, in the brief, in the review.

That is what Patrika is, for me: an attempt to keep the older question visible. The whole issue lives at patrika.devalok.in. Take it apart in the margins.