Most identity work does not fail at the concept stage. It fails six weeks later, in a trademark search nobody scheduled, after the name has already been shown to the board.
The sequence is almost always the same. A studio presents three names. One is loved. It goes into decks, into a domain purchase, into an internal announcement. Then counsel runs a clearance search and finds a registration in the same class, in a market the client plans to enter next year. The work is not bad. The order was wrong.
Clear before you present
We run a preliminary knockout search on every candidate before it reaches a client. It is not a legal opinion and we say so, but it removes the obvious collisions early, when replacing a name costs an afternoon rather than a quarter.
A name you cannot register is a sketch, not a decision.
Present in threes, with evidence
Each candidate arrives with its search notes, its available domains and the classes it would need to cover. The conversation shifts from taste to risk, which is the conversation a board can actually have.
On the transit rebrand this cost four days at the start and saved a launch date at the end. Two of the three shortlisted names were dropped before anyone outside the working group saw them, and nobody had to explain why the favourite disappeared.
Write the order down
Search, shortlist, clear, present, file, then design. Put it in the scope with dates attached. The phase that gets skipped is always the one nobody owns.