No mystery.
A beginning, a middle, an end.

The shape of an advisory season

An advisory season typically runs twelve to sixteen weeks. It begins with adiagnostic fortnight, in which I read what your organisation actually writes — board papers, clientletters, the deck that keeps being rewritten — and sit, quietly, in the meetings where its judgementtravels. Nobody performs for me; I am there as an editor of the record. The output is a written map ofwhere voice is strong, where it leaks, and what the leaks are costing you in decisions delayed andpositions softened.

From week three the work becomes editorial in the oldest sense. We meet fortnightly as a working group —usually the executive team or the partner group whose documents matter most — and each session works livematerial: a funding case, a partner promotion paper, the opening of an investor letter, a conferencekeynote. Drafts are re-argued before they are re-worded. The discipline is simple and uncomfortable: everydocument must be able to argue without its author in the room. If it cannot, the thinking is not finished,and no amount of sentence-polish will save it.

Between sessions I work directly on your drafts — marked, restructured, and returned with the reasoningshown. Participants tell me the margin notes become the most forwarded internal reading in the firm; oneclient laminated one, which I have decided to take as a compliment. By week six the difference is usuallyvisible in three places: papers that open with a position, meetings that start with the ask, and a housevoice that new joiners absorb without ever seeing a manual.

The season closes with a written handover — what changed, what to protect, and what to revisit in ayear — plus a working session where the group rehearses defending the new standard without me in the room.That session matters most. The measure of the work is not how good your documents are while I am engaged;it is how good they remain after I have gone.

What it asks of you

Three things, stated plainly. Real documents: the drafts that actually matter, not showcase exampleswritten for the exercise. Senior presence: the people whose names go on the papers attend the workingsessions — voice work delegated downwards dies downwards. And honesty in the diagnostic: if theorganisation's writing problem is actually a decision-avoidance problem, we will name it, and the planwill say so.

A week in the season, described

Because rhythm is easier to judge than description, here is a typical middle-of-season week. On MondayI read: two board papers due for next month, the draft of a client letter that has been "nearly there"since April, and the notes from last week's session — with, in the margin of each, the question theauthor was really asking. Monday evening the marked drafts go back: not rewrites, arguments — everychange shown with its reasoning, so the author learns the move rather than inheriting the edit.

Tuesday is a working day for the client; I am quiet. Wednesday fortnightly session: ninety minutes,five people, one paper that matters. The first half hour is uncomfortable by design — we defend thedocument's position before we touch its prose, and half the time the position turns out to be acompromise wearing a recommendation's clothes. The second hour rebuilds it in the room, roles rotatingso the habit does not pool in one person. Thursday I draft: the one-page summary of the new paperstandard the group converged on, ready for Friday's circulation, written in the client's register ratherthan mine. Friday, silence — deliberately. A season that demands five days a week of its client is aseason that has mis-specified itself; the work is designed to make organisations better at their ownrhythm, not addicted to mine.

Multiply that week by fourteen and you have the season. What accumulates is not just better documentsbut a better argument culture: the group develops a shared ear for the evasive sentence, a sharedintolerance for the buried ask, and — the outcome I care about most — a shared willingness to defend aclear position out loud, in front of hierarchy, before the paper is even written. Several clients havetold me the papers were almost a side effect. I take that as the compliment it is.

What the diagnostic looks at

The diagnostic fortnight is not a formal audit; it is closer to a structural survey, and it movesquickly through four layers. First, the record: what the organisation actually writes, read in sequencerather than in isolation, because patterns live in repetition. Second, the meetings: where possible I sitin — silently — on the two or three gatherings where the important documents are discussed, watchingwhether the paper decides the meeting or the meeting re-decides the paper. Third, the people: shortconversations with the writers, the approvers and the readers of the same documents, mapping the gapbetween what was meant, what was written and what was understood. Fourth, the stakes: which documentscarry decisions, and what it cost the organisation the last time one failed to.

The written map that closes the fortnight is short by design — rarely more than six pages — because adiagnostic nobody finishes is a diagnostic that never happened. It names the patterns with evidence fromthe client's own documents, ranks them by cost rather than by visibility, and proposes the shape of workthat would fix the top of the list. It is written to be argued with: clients who disagree with it in thefirst read-through produce the most valuable sessions of the season, because the disagreement is alwaysabout something deeper than prose.

Measuring whether it worked

Clients rightly ask how a voice engagement is measured, and the honest answer has two layers. Thefast layer is mechanical and visible within weeks: the ask appears before page two; board packs shed athird of their length without losing an argument; the average document needs one rewriting instead offour. These are counts, and we take them — the scorecard in your handover starts here. The slow layer iscultural and matters more: disagreements surface earlier and cost less; positions are defended out loudbefore they are defended in minutes; the meeting where the firm used to negotiate with its own documentnow spends that hour deciding. Nobody has found a number for that layer, so the handover records it astestimony instead — quotes from your own people, dated, describing the difference. Clients revisitingthe handover a year later tend to reread the quotes first. So do I.