The Work Before the Work

A brief usually arrives as a request for a thing: a workshop, a strategy, a redesign, a growth plan, a coaching engagement. The request is useful, but it is rarely the whole truth.

Under the request there is usually a pressure: a team cannot align, a product is not converting, a founder is carrying too much ambiguity, or a market story has become too broad to act on.

The shape of an incomplete brief

An incomplete brief has a recognisable shape. It names a deliverable, a date, an owner, and a budget line. It is precise about format and vague about outcome. It describes what should be produced without describing what should be different once the thing exists. The distance between those two descriptions is where an engagement either holds or drifts.

What the brief leaves out

What is missing is usually the same short list.

• Who decided this was the problem, and on what evidence.
• What has already been tried, and what happened when it was.
• Which constraint is fixed and which one is inherited from a decision nobody has revisited.
• Who can authorise a change to the thing that would have to change.

None of this is withheld out of caution. It is absent because it was settled inside the organisation before anyone wrote the brief, and the reasoning stopped being visible the moment it became consensus. A company can recover most of it without help, by running its own diagnostic before the shortlist.

A conversion drop is the cleanest example, because it arrives with evidence attached. A team watching a checkout flow sees completion fall across a quarter and reads it as an interface problem. The brief that follows asks for a redesign: fewer steps, clearer buttons, a shorter form.

Sometimes that reading is correct. Often the drop sits somewhere the interface cannot reach. In one payment flow, the fall began at the point where total cost became visible for the first time, including a fee introduced upstream by a commercial decision made in another part of the business. Customers were not confused. They were declining. A redesign would have moved the moment of refusal one screen earlier and left the refusal intact. Price sat outside the brief, so the brief could not see it.

The same shape in onboarding

Digital banking onboarding produces the same pattern in a different register. Abandonment concentrates at identity verification, and the obvious reading is friction: too many documents, too many retries, a camera step that fails on older devices. That friction is real and worth fixing. Abandonment also concentrates where the process asks for something the customer does not yet trust the institution to hold. The same step performs differently depending on what was said before it, which makes it a positioning problem wearing the costume of a usability problem.

A skills gap that is not one

The training version is harder to say out loud. A request arrives for sales enablement, because a team is not converting pipeline and the internal diagnosis is capability. Sitting with the team, the calls are competent. What is absent is permission. Nobody on that team can move a price, change a term, or commit to a delivery date without an approval chain that outlasts the buyer's interest. That is not a skills gap. It is an authority gap, and training a team to sell harder inside it produces frustration with better vocabulary.

Stakeholder interviews are the standard instrument, and they are reliable for one thing: what people are willing to say. They are weaker as a record of what an organisation does. People describe intentions accurately and priorities generously. Everyone can name the strategic priority. Fewer can name what was given up for it.

Where attention and money actually go

So the interviews get read against two other documents.

• The calendar shows where attention goes: which meetings are protected, which are cancelled first, who is in the room when a trade-off is settled and who is briefed afterwards.
• The budget shows what has been committed against reversal: what is staffed permanently, what is funded by a project line that expires, what is described as important and resourced as optional.

Neither document is written to be read this way, which is part of why they hold up under reading.

The gaps between the three sources are the finding. When a leadership team names retention as the year's priority, and retention has no recurring meeting and no owner holding a budget, the priority is a preference. That is not hypocrisy. It is what happens when a stated intention has never been converted into a decision that costs something, a pattern set out at more length in Decisions That Shape Outcomes.

What people say still matters, and the disagreements matter most. Two directors describing the same problem in incompatible terms is evidence that the organisation's model of itself has split, and the split usually runs along the line where the real cause sits. The point is not to catch anyone out. It is to locate the distance between the account an organisation gives of itself and the behaviour it funds, because the brief was written from the account.

The work before the work is a short period of disciplined attention. What is being asked underneath the request? What has already been tried? Which constraint is fixed, and which one is only assumed?

This protects the engagement from becoming performative. It keeps a project from solving the most visible problem while leaving the important one intact.

The loss that does not look like failure

The cost of getting it wrong is easy to miss, because nothing about it looks like failure. The redesign ships. The training runs and reviews well. The workshop is well attended and the team is glad it happened. Effort was spent, craft was applied, and the number that prompted the brief does not move. The effort is real. The result is not.

The second loss is larger than the first. It is not the fee or the quarter, but the standing of the method inside the company. A team that has solved the wrong problem well becomes sceptical of the next diagnosis, including a correct one. Budget holders learn that work of this kind does not change anything, and the next attempt is harder to fund and shorter than it should be. This is why we treat framing as part of the deliverable rather than a preamble to it, and why the diagnosis before solution pillar sits underneath most of what we publish.

A better question does not make the work heavier. It makes it lighter. It gives the team a shared object, a sharper definition of progress, and fewer loops of explanation.

This phase is often unbillable. It happens in a first conversation, a second one, and a document nobody commissioned. There is a version of practice that turns it into a paid discovery phase, and for some problems that is the right structure. The honest reason it often stays unbilled is that it decides whether we take the engagement at all. That judgment belongs to us before a client pays for it. A brief we cannot reframe is a brief we cannot deliver against.

Scope is fixed before the invoice

Unbilled does not mean unimportant. By the time a proposal is written, the framing has already fixed the scope, the sequence, the measure of success, and the people who need to be in the room. Everything invoiced afterwards is executed inside a definition that was set for free. Scope written from a symptom stays written from a symptom, however well the work is later run.

An engagement rarely fails at the delivery stage. It fails at the sentence that described the problem.

This is why we begin quietly. Not to delay action, but to make action worth taking. Where the thinking becomes work, it takes two forms: consulting that carries a reframed problem through product, growth, and commercial systems, and training that leaves a team able to do this framing without us.