I have hired outside help and been hired as it. The engagements that work and the ones that disappoint differ in ways that are visible before anything starts.
Define the decision, not the deliverable
Engagements that produce documents nobody uses usually began with a request for a document. Engagements that matter began with a decision the business needed to make and could not make confidently alone.
Ask for the decision. Should we expand to a second location. Should we take this contract. Where is the money leaking. The work then organizes around answering that, and you can tell whether it succeeded.
Establish who owns implementation before you start
The recommendation is the easy half. Most value shows up during execution, when reality complicates the plan and something has to be adjusted.
Decide upfront whether the outside help stays involved through implementation or hands off at the recommendation. Both can work. What does not work is assuming one and getting the other, which is how a business ends up with a sound plan and nobody to run it.
Insist on senior involvement, and verify it
A common frustration is meeting experienced people during the sale and working with junior people afterward. Ask directly who does the work and who is in the room during it. The answer should be specific.
This is a real advantage of small firms and independent advisors. The person you evaluated is the person you get, and that continuity is worth something on complex problems.
Ask what you are left able to do
A good engagement usually leaves the organization more capable, not more dependent. Sometimes that is documentation. Sometimes it is training.
When I rebuilt the website for a collegiate baseball league after its previous platform shut down, part of the work was teaching their people to maintain it. A nonprofit running on volunteers cannot call someone every time a schedule changes. If it needs outside help for routine updates, the problem was solved for the length of the invoice rather than for good.
Worth asking directly before you sign: when this ends, what can we do ourselves that we could not do before?
Watch for scope that only grows
Good outside help occasionally tells you a project is unnecessary, or smaller than you thought, or that you can do it yourself. If every conversation expands the engagement, that is a signal about incentives.
Give them the unflattering information
The most common way clients undermine their own engagement is withholding context: the failed initiative, the internal disagreement, the number that is worse than they admitted. Advice built on a partial picture is worth what it is built on.
Agree what changed
At the end, both sides should be able to name what is different. A decision made, a process running, a system in place, a specific result moving. If nobody can name it, the engagement was activity rather than work, and that is worth knowing before commissioning the next one.