Filtering can be reduced but not removed. What remains is the need for a small number of people whose account of things is not adjusted for your position at all.
Almost nobody has this by accident, and the people who do usually built it before they needed it.
The requirement is structural
The person has to be able to be wrong about you at no cost to themselves.
This rules out almost everyone you work with, however much you trust them and however sincerely they intend to be straight. A person whose compensation, hours, projects, or prospects you influence is making a calculation whether or not they are aware of it. They can be honest, and frequently are, and the honesty passes through a filter they cannot switch off.
Which means the useful people are mostly outside your organization. Peers running comparable operations elsewhere, somebody senior who has no stake in your outcomes, a former colleague now somewhere unrelated, occasionally a friend from before the role who is unimpressed by it.
Three or four is the working number. Fewer and one person’s blind spot becomes yours. More and it stops being a set of relationships and becomes a survey.
What they need besides independence
Independence alone produces contrarians, and a contrarian is not the same thing as a source of accurate information.
The person needs enough context to understand your situation, which usually means someone who has held a comparable seat rather than someone bright and uninvolved. They need to be willing to say the thing plainly rather than gesturing at it, since a hint will be absorbed as agreement. And they need to be right often enough that you will not learn to dismiss them, which is a real risk with people who disagree reflexively.
What they do not need is agreement with your general approach. The most useful person in this set is frequently somebody who runs things differently from you and thinks some of your instincts are mistaken.
Asking a question they can answer
Even with the right people, a badly framed question returns reassurance.
Do you think this is a good idea invites a verdict, and a verdict from a friendly party is usually favourable. Am I being unreasonable about this invites the answer no, which is what people say.
The productive versions do not ask for a judgment about you. What is the strongest case against this, and who would make it. What would you be worried about if this were yours. What am I not asking about. That last question is the most useful one in this entry, and it is the one nobody thinks to ask.
Present the situation without also presenting your conclusion. The moment your preferred answer is visible, the response adjusts toward it, and the adjustment happens even in people with no stake at all.
You can destroy it in one exchange
These relationships are more fragile than they appear, because the other person is doing you a favour that costs them effort and social risk and returns nothing.
Defend your position vigorously once and they will conclude the exercise is not what it was described as. They will not say so. They will simply become less specific next time, and the relationship will continue in a diminished form that looks identical from your side.
The disciplines are narrow. Do not argue in the moment; take it away and consider it. Come back and say what you did with it, including when you rejected it, since a person whose input disappears without trace stops offering any. And do not ask only when you are in trouble, because a relationship activated exclusively during crises will not be available during one.
The reciprocal version is the durable one
The arrangement that survives is symmetrical. You do the same for them.
A one-way arrangement in which somebody advises you indefinitely will fade, because it is unpaid work with no return and social obligation only carries it so far. A mutual one persists for decades and improves, since each party accumulates context on the other’s situation and the quality of the counsel rises accordingly.
In Idaho this is more available than people assume and less used than it should be. The number of people running organizations of comparable scale in a comparable market is small enough that they can be identified and large enough that the useful ones are not competitors. Most of them have the same problem you do, and most of them have not solved it either.
Build it before the role requires it
The moment you need this most is the moment it is hardest to construct.
A leader in genuine difficulty who begins reaching out to peers is asking strangers to do something demanding on no established basis, and the request itself signals the difficulty. Approaching the same people in an ordinary period, over something not urgent, costs nothing and establishes the thing that will be available later.
Which makes this one of the few items in this subchapter with a clear instruction attached. Identify two or three people now, while nothing is wrong, and ask them something you do not particularly need answered.
Edited by Patrick J. Wolf, PhD