Approval Can Replace Judgment
Teams say they want people with taste, then review every decision until nothing feels risky. A design passes through references, meetings, and approvals before anyone commits to it. The result may be competent, but the person making it did not practice taste. They practiced predicting what the room would accept.
You cannot develop taste if every choice you make is already socially approved.
Approval is useful because it lowers uncertainty. You can point to an established pattern, a respected company, or a person with status and explain why the decision is safe. If the result fails, the choice still looks reasonable because other people already endorsed it.
Taste begins before that protection exists. It is the ability to choose without knowing whether the room will reward the choice.
Consensus Arrives After the Interesting Decision
Consensus is easiest to trust after a choice has already worked. By the time an idea becomes accepted, the original uncertainty has disappeared. The design has users. The product has revenue. The unpopular opinion has become a case study.
Copying the result can teach you a vocabulary. It cannot teach you how the original decision was made. You see the answer after the market has graded it, not the judgment that existed before the evidence.
This creates a strange failure mode. Someone can consume excellent work, repeat all the right principles, and still struggle when a new decision has no approved reference. They know what good looked like yesterday. They have not built a standard for choosing today.
Taste Is Not Automatic Disagreement
Rejecting consensus does not make a decision tasteful. Reflexive contrarianism is still controlled by the group. The group approves, so you reject. The group rejects, so you approve. Your choice remains dependent on the same social signal.
The difference is the source of the decision. A contrarian asks what other people will dislike. A person with taste asks what the work needs, then accepts that the answer may not be popular yet.
That answer also has to survive contact with reality. Taste is not a private feeling that cannot be questioned. It is a working standard that produces choices, receives feedback, and becomes more precise over time.
Make the Choice Before Asking Permission
Teams cannot turn every production decision into a personal experiment. Shared systems need constraints, and expensive mistakes deserve review. But people still need room to make reversible choices before consensus closes the decision.
Make a complete choice before asking for approval. Pick the interface detail and write the sentence. Remove the feature or change the sequence. Then state why the result is better before asking whether everyone likes it.
Then watch what happens. The choice may work for reasons you understood or reasons you missed. It may fail and expose a weak standard. Each outcome teaches you something because the decision was yours before the feedback arrived.
Approval before the choice removes that learning loop. Feedback after the choice strengthens it.
Social Risk Is Part of the Cost
Developing taste means being wrong in a way that belongs to you. That is uncomfortable. An approved mistake is easier to explain than an unapproved decision that failed.
Avoiding that discomfort has a cost. Your standards remain external. You become good at recognizing what has already earned status, but not at deciding what deserves it next.
Taste develops when you make a choice you can explain before you know whether the room will reward you for it.