By month nine, nobody remembers why.
01 · The question as it arrives
“We’re launching a new brand.” The date is usually set by something else — a funding round, an anniversary, a merger, a building.
The scope on the table is identity: name, logo, site, launch collateral. The work is described as a reveal, and the reveal has a date.
02 · Why the obvious answer fails
The launch is one day. The brand is operated for years, by people who were not in the room when it was decided.
Within a year the people running it are often different people. What they inherit is a folder of files and a set of rules with no reasoning attached. Rules without reasoning are overruled by whoever is most senior in the meeting.
The result is drift. Every piece is slightly off, nobody can say why, and everything routes back through the one person who remembers the original argument. That person becomes the bottleneck, and then they leave.
The cost is not a bad logo. It is an organisation that can no longer produce anything on brand without a meeting.
The reveal also sets the wrong test. On launch day the brand is judged by people who have seen the reasoning — the team, the board, the agency that made it. It is operated for the next three years by people who have not, and nobody is in the room to explain it to them.
None of this is a failure of the design. The identity can be correct and the handover still not happen, which is why it is rarely caught in the review that signs the work off.
03 · What we look at instead
What is being decided, against what is being designed. A launch contains a small number of decisions — who this is for, what is given up, what it will be called, what will never be said — and a large amount of execution. Only the decisions have to survive the year.
Who operates it afterwards, by name. If nobody owns it, the brand is a project with an end date rather than an asset with a custodian.
What gets written down, and where. The words that are used, the words that are retired, and the reason behind each. A vocabulary table settles more arguments in year two than a logo does in year one.
What the launch is allowed to break. Most launches are scoped to change everything visible at once. The things that already carry recognition — a mark customers navigate by, a product name the market uses, the sentence the sales team opens with — are worth listing before they are redrawn, so that replacing them is a decision someone made rather than a side effect nobody noticed.
The second-year test. Can someone hired next year produce something on brand without asking anyone? If the answer requires a meeting, the handover did not happen.
04 · Where to start
Before any design work, agree one page that sets out who the brand is for, what it gives up, and three things it will never say.
Share that page with the people who will run the brand day to day, not only the people signing it off.
List what customers already recognise you by — a mark, a product name, a familiar line — and decide deliberately whether each one stays.
If that page is difficult to write, the identity work is starting too early. It is far better to find that out in the first week than at launch.