Hiring Consultants

Why consulting is harder to buy

With a build, you know what done looks like: the thing exists or it does not. With advisory work, the output is a view, and a view can always be elaborated, extended and revisited. Nothing in the work itself says when to stop.

That is why consulting engagements drift more than builds do. Not because consultants are less honest, but because the format has no natural finish line unless you supply one.

Scope the question, not the activity

The most common scoping mistake is describing what the consultant will do rather than what you need to know. "Interview the team and review the data" describes activity that could run indefinitely. "Tell us why time to hire doubled in the last year, and what three changes would reverse it" describes a question that can be answered.

Once the question is written down, two useful things follow. The consultant can tell you whether it is answerable in the time you are offering, which is a conversation worth having before you start. And you get a test for completion: the engagement is over when the question is answered, not when the hours are used.

Insist on an artefact

Ask for something written. A recommendation document, a costed options paper, a decision memo, a prioritised list with reasoning attached.

This is not about distrust. Advice delivered only in meetings degrades: it is remembered differently by different people, and it is unavailable to whoever inherits the problem in six months. A written output also forces the consultant to commit to a position, which is what you are paying for and the hardest part of the work.

Two things make the artefact more useful. Ask for the reasoning alongside the recommendation, so you can tell later whether a changed circumstance invalidates it. And ask what they considered and rejected, which is often more informative than the recommendation itself.

Structuring the engagement

Short and fixed beats long and open. A two-week diagnostic with a defined output tells you a great deal, including whether this consultant is worth a longer engagement. It is a cheap test that an open-ended retainer never gives you.

Separate the recommendation from the implementation. They are different purchases with different skills, and a consultant who recommends work they will then be paid to do has an obvious conflict. Split them, and let the second decision be made after the first is delivered.

Name the audience and the decision. Who is this for, and what will they decide with it? A recommendation aimed at nobody in particular tends to be written for everybody, and lands with no one.