The designer who claims they cannot work without research is asking to be told what to think. The designer who claims they never need it has stopped being honest.
Most product weeks land somewhere in between. You have a partial signal from a customer success ticket, three field visits from last quarter, a PM with an opinion, and a deadline that does not move. Calling this “no research” is wrong. There is research. It is just not the research you wanted, and it is not interview-clean.
The skill is reading the partial signal well. I keep a small mental checklist when there is no time for a study.
What does the customer success team fix every week without a ticket. That pattern is usually the second-most-important problem in the product, and nobody has named it because nobody has the bandwidth to escalate it.
What did the last field visit reveal that the deck did not mention. Field visits get summarised into bullet points and the bullet points get filed. The thing that surprised you on site, the moment you said “wait, what” out loud, that is the signal worth carrying into the design.
What is the noisiest disagreement on the PM’s side. PMs argue about the things the data does not settle. Whatever they keep coming back to is the part of the problem where judgment, not data, is going to ship the answer.
What would I bet money on. Not what would I prefer. What would I actually wager fifty dollars on if a customer were to behave a certain way. That bet is the hypothesis the design should test, not the feature the design should ship.
The honest limitation is that judgment built without research is judgment built on previous research. The designer who has never run a study is not exercising judgment when she skips one. She is guessing. The skip is earned by having done the work earlier in a career, not by having read about it.
The shift is to stop framing it as research-or-no-research. The frame is signal-to-judgment ratio. Both go up over a career. The skill is knowing which one is doing the work on any given Tuesday.