Commissioning Judgement: What a Good Brief Should Specify, and What It Should Leave Alone

Most briefs for professional services arrive with the answers already written in. The document names the workshops, the reporting cadence, the deliverable formats and, on a difficult day, the number of slides. What it frequently omits is the problem itself, stated plainly enough that a new joiner could understand it.

That inversion is understandable. A buyer who has been disappointed before reaches for specificity as a form of insurance. The logic runs: if I describe the work precisely enough, I cannot be surprised. The difficulty is that an over-specified brief does not purchase quality; it purchases compliance. The two are not interchangeable, and the gap between them is where most disappointing engagements live.

When the brief answers its own questions

A brief that prescribes method has quietly removed the supplier's judgement from the engagement. If the buyer has already decided which interviews to run, in what order, and what the output must look like, then the supplier's value narrows to execution capacity. That may be exactly what is wanted for a defined piece of delivery work. It is rarely what is wanted when an organisation is buying analysis, challenge or advice.

The practical consequence is that the supplier stops telling the client what it thinks. Observations that fall outside the specification are set aside as out of scope, and the document that lands at the end is a competent answer to the wrong question.

Compliance is not the same as quality

Compliance is measurable and therefore comforting. A brief can list twelve deliverables and a buyer can tick twelve boxes. But a tick tells you the work arrived, not that it was any good.

Mid-market organisations feel this pressure more sharply than most. A 200-person business buying a finance review, an ERP selection or an HR restructure often lacks a permanent procurement function, so the brief becomes the whole of the control mechanism. Over-specification is a substitute for the governance that a larger buyer would have in place. The answer is not to loosen the brief and hope; it is to put the control somewhere more useful.

A four-part commissioning model

A better structure holds four elements in balance. It is deliberately shorter than a specification and considerably more demanding to write.

Outcomes

Describe the destination, not the route. What should be true at the end of the engagement that is not true now? A manufacturer in the West Midlands commissioning a costing review might write: we need to know which three product lines destroy margin and have a defensible plan to address them before the next budget round. That sentence constrains the work far more tightly than a list of fourteen interview targets, because it defines success rather than activity.

Constraints

Name only those constraints that genuinely bind. Budget, timing, regulatory obligations, systems that cannot be changed, sites that cannot be visited, an industrial relations sensitivity. A good constraint list is short and slightly uncomfortable, because each entry removes an option the supplier might otherwise have proposed. If the list runs to a page, it is a specification in disguise.

Success measures

Say how you will know, not how you will score. Measures should be observable by both parties and agreed before work begins. For a piece of advisory work, that might be: the board can take a decision at its February meeting without asking for further analysis. Such a measure is honest about the fact that the real test of professional services is what the client is subsequently able to do.

Decision rights

This is the element most often missing and most often the cause of late-stage friction. Who decides what, and by when? Who approves a change of scope? Who signs off a recommendation that the engagement team is divided on? Who can stop the work?

Setting decision rights in the brief protects both sides. The supplier knows whose judgement prevails on which questions, and the buyer knows where its own accountability sits. It also removes the ambiguity that lets a project drift into a series of informal approvals that nobody has recorded.

What to leave out

Leave out the slide count. Leave out the workshop agenda. Leave out any requirement whose only purpose is to demonstrate that the buyer has been thorough.

Most importantly, leave out the instruction to confirm what the buyer already believes. A brief that implicitly rewards agreement will receive agreement, and the fee will have bought a mirror.

A worked example

Consider a professional services firm of 120 people commissioning support on a succession plan. The over-specified version asks for a competency framework, two workshops with the partnership, a nine-box grid and a written report. The supplier complies. Everyone is satisfied and nothing changes.

The commissioning version reads: we need a named successor for each of four partner roles, agreed by the partnership, with the development gap for each documented and resourced. Constraints: no external recruitment before 2027; two partners retire within eighteen months. Measures: the partnership votes on a slate by the end of the first quarter. Decision rights: the managing partner decides on process; the partnership decides on candidates; the supplier may escalate disagreement to the board without prejudice.

That second brief is half the length and roughly twice as hard to write. It also leaves the supplier free to say something the client had not thought of, which is the point of buying expertise in the first place.

Putting it into practice

A useful discipline is to draft the brief in two columns: what we are buying, and how we expect it to be done. Then delete the second column and read what remains. If the document still describes a coherent piece of work, the brief is sound. If it collapses without its instructions, the buyer is not commissioning a supplier; it is recruiting temporary staff by another name.

Clear commissioning is not a relaxation of standards. It is a relocation of them, from the method to the outcome, where they can actually be tested.