A good no comes with two names and a reason.
We take six projects at a time, which means that most of the conversations we have end with a no. Here is how we try to make that useful.
The limit is simple arithmetic. Eleven people can know six buildings well. At seven, someone starts answering questions about a project they have not visited that month.
So when the sixth place is taken we say so in the first reply, and we say when the next one opens. If the project cannot wait, we name two practices we would trust with it, and we tell the client why.
Sometimes the no is about the project and not the calendar. If the feasibility stage shows that the site, the rules or the money will not carry the building, we write that down and hand it over as a finished piece of work.
About a third of those clients come back, often years later, with a different site. We would rather be the office that was right than the office that was available.