When a product feels heavy, the instinct is to redesign it. Better hierarchy, cleaner components, a calmer palette. That work is real, but it’s usually the second job. The first is to find out how much of the product needs to exist at all.
I start every project the same way: list every question the product asks the user, and try to delete each one.
Three questions per field
For anything a product asks for, there are only three useful questions.
Do we already know this? Most businesses hold far more about a customer than the form admits. Anything already on file should be shown for confirmation, not typed again.
Can we derive it? A surprising number of fields are calculations wearing a costume. If two other answers determine the third, don’t ask for the third.
What breaks if we don’t ask? Sometimes the honest answer is “a report someone stopped reading in 2019.” That field can go.
What survives all three is the actual product. It’s normally a fraction of what was there.
Do the deleting before the design
The order matters more than it sounds. If you design first and cut later, you spend the cutting phase defending layouts instead of examining need — every removal becomes a loss of work rather than a gain in clarity.
Cut first, and the visual work gets easier, because there’s less to arrange and the hierarchy is obvious. Fewer things on a screen means fewer decisions about how they relate.
What this looks like in practice
On one application flow, the version people used ran across several screens and asked for things the firm already held. Cutting it to a single screen wasn’t a visual exercise — it was pre-filling what was known, deriving what could be derived, and leaving exactly one real decision on the page.
The design that followed took a fraction of the time it would have taken to make the original version look good.
The uncomfortable part
This approach makes the first two weeks look unproductive. There are no screens to show. Someone will ask what design has been doing.
The answer is that the team is deciding what the product is, which is a design decision even though it doesn’t arrive as a picture. It’s worth learning to defend that, because it’s where most of the value is.