In short: An evaluation that finds a program isn't working is still useful, and there are honest ways to report it that protect the program and the funder relationship at once.
Sometimes the data say the program is not working. These pages cover what to do with a hard finding, how to tell a funder, and how a difficult result becomes useful.
The choice between an outside evaluator and doing it internally turns on independence, capacity, and what the funder requires, and sometimes a full evaluation is more than you need.
Vetting an evaluator comes down to specific questions about fit, method, and availability. Credentials alone won't tell you who's right for the work.
A report gets read when it answers the question the reader arrived with, and a credible impact story carries evidence and context alongside the human account.
Boards, communities, and accreditors each need something different from the same results, and sharing results with a community carries obligations a board briefing doesn't.