The Delivery Craft, Written Down — Why This Library Exists
The scandal is that we have systematically chosen not to teach it, on the grounds that it cannot be written down.
We Wrote Down Everything Except the Thing That Mattered
For twenty-five years we have been writing down the wrong half of the job.
Open any bookcase in any delivery function and count the spines. Method after method. Body of knowledge after body of knowledge. Certification guides, maturity models, operating-model playbooks, scaled-agile handbooks thick enough to stop a door. Since the turn of the century the profession of delivering change has produced more codified material than any comparable discipline, and we have been diligent students of it. We can recite the stage gates. We can draw the governance tiers from memory. We can argue the relative merits of a dozen estimation techniques.
And yet the programmes still fail, at roughly the rate they failed in 2000, for reasons that appear in none of those books.
That is the charge this manifesto lays, and the reason this library exists. We have spent a quarter of a century documenting method — the repeatable, teachable, certifiable mechanics of delivery — while leaving craft — the judgement that decides whether any of that mechanics is applied well — almost entirely unwritten. The first half is now over-supplied to the point of parody. The second half is passed down, if it is passed down at all, by standing next to someone who has it, and it dies with them when they leave.
“We are fluent in method. We are barely literate in judgement.”
The Orthodoxy We Are Refusing
The prevailing view in the profession is comfortable and rarely stated aloud, so let us state it plainly: delivery competence is method competence. Know the framework, apply the framework, improve the framework. When a programme fails, the orthodoxy looks for the method that was skipped — the risk register not maintained, the benefits not baselined, the retrospective not held — and prescribes more method as the cure. Every major failure of the last two decades has produced a new standard, a new assurance regime, a new checklist, and the failures have carried on regardless.
We refuse this view. Not because method is worthless — it is the floor beneath everything, and nobody should be on a programme without it — but because it explains almost nothing about why good programmes succeed and bad ones founder. Consider how the difference actually shows up.
Two programme directors run two programmes of comparable size, both on the same methodology, both with the same governance stack, both reporting into the same portfolio office. The first spends the Tuesday before a steering committee rewriting the paper three times until the one decision the board must take is in the first paragraph, and then spends Wednesday walking the two sponsors most likely to object through it in private. The second sends the standard template, forty pages, amber on three criteria, decision buried on page twenty-two, and is surprised when the committee defers. Both followed the method. Both held the gate. Only one of them knows what a steering committee is for — and that knowledge is in no handbook, because the handbook describes the meeting and not the craft of using it.
That gap — between knowing the meeting exists and knowing what it is for — is the entire territory this library is built to occupy.
What Craft Actually Is
Craft is not charisma and it is not instinct. Those are the lazy words we reach for when we cannot be bothered to analyse what a good practitioner is doing. Craft is accumulated, pattern-level judgement: the capacity to recognise a situation you have seen before, to know which of the hundred things the method permits is the one thing that matters this week, and to act on it before the situation has finished declaring itself.
It has recognisable components, and every one of them can be described:
- Pattern recognition. Knowing that a programme reporting green for nine consecutive months while its change-request log doubles is not a healthy programme but a quiet one — and knowing why the two signals diverge.
- Sequencing judgement. Knowing that the order in which you secure commitments matters more than the commitments themselves; that the sponsor who signs last is the one who will own the outcome, and that this can be arranged.
- Proportion. Knowing when the method’s full weight is warranted and when it is ceremony; which gate to hold hard and which to wave through with a conversation, and being able to defend the difference.
- Reading the room, then writing it down. Knowing that the real decision was taken in a corridor, and that the craft lies in getting it minuted before it evaporates.
- Timing of truth. Knowing the moment at which an uncomfortable fact becomes cheaper to surface than to carry — and that this moment is almost always earlier than it feels.
None of this is mystical. All of it is learnable. The scandal is that we have systematically chosen not to teach it, on the grounds that it cannot be written down.
The Strongest Objection, Taken Seriously
The serious version of the objection deserves a serious answer, because it is not stupid. It runs: craft is tacit knowledge. It lives in the doing, not in the text. The moment you write it down you flatten it into rules, and rules are exactly the method you claim to be rising above. You will produce one more handbook, and it will fail for the same reasons the others did.
Three things are true here, and one thing is false.
It is true that craft cannot be fully transmitted in prose. Nobody learns to run a difficult programme by reading about it, any more than anyone learns to sail from a book. It is true that the worst possible outcome would be to reduce judgement to a checklist — “Step four: exercise proportion” — and we should be vigilant against it. And it is true that a written account of craft is always a simplification of the lived thing.
What is false is the conclusion. The alternative to writing craft down badly is not leaving it unwritten; it is writing it down properly — as the master practitioners of every other discipline have always done. Surgeons write case reflections. Barristers write opinions that reason through judgement, not just law. Architects publish the thinking behind the building, not only the drawings. These texts do not replace apprenticeship. They make apprenticeship faster, because the apprentice arrives with a vocabulary for what they are about to see, and can recognise the pattern on the second encounter instead of the tenth.
Writing craft down does not turn it into method. Method describes what to do. Written craft describes what a situation means and why a seasoned practitioner would read it that way. The first is a procedure; the second is a reflection. The failure of the last twenty-five years was not that we wrote too much — it was that we only ever wrote the first kind.
The cost of the orthodoxy’s position, meanwhile, is concrete and measurable. Every practitioner who leaves a delivery function takes with them ten, fifteen, twenty years of pattern recognition that was never captured anywhere, and the people who remain start again from the handbook. We would consider it negligent to let a complex system run with no documentation of how it actually behaves under load. We have tolerated exactly that for the most expensive, most failure-prone activity most organisations undertake.
What This Library Is For
So this is the library’s purpose, stated as a commitment rather than an ambition.
It exists to write the craft down — the judgement layer of delivery, above the method layer, which the profession has left blank for a generation. It sets itself a discipline in doing so: every claim must show its mechanism, not merely assert conviction; every pattern must be illustrated with the concrete texture of a recognisable situation, never the safe abstraction of a maturity level; and where an argument is being made, the strongest opposing view must be engaged, because a craft that cannot survive contradiction is a superstition.
It will take several forms, because craft does not have one shape. Some of it argues a position. Some of it sets out a method — complete, usable, with its decision points and its trade-offs visible. Some of it offers a new lens on an old problem. Some of it works through a composite case from the situation to the transferable lesson. Some of it looks forward, honestly unsure. The form varies; the discipline behind it does not.
And it is written in the practitioner’s voice — from the pattern, not the podium — because craft is owned by those who do the work, and it should be written by them, in their register, for their successors.
The Testament
We have been handed a profession with an enormous, well-organised written inheritance of method and an almost empty written inheritance of craft. We are not obliged to leave it that way.
The people entering delivery now will spend their careers inside frameworks we have already perfected, facing situations we have already seen, making mistakes we have already made and never recorded. That is not a tradition. It is a failure of stewardship dressed up as humility about what can be written.
So we will write it. Imperfectly, as all craft is written. Contestably, as all judgement should be. But written — so that the next practitioner who stands in front of a green report and a doubling change log, or a steering committee that will not decide, or a sponsor who has quietly withdrawn, does not have to learn from nothing what we already know.
That is why this library exists. It is the half of the job we never wrote down. We are writing it down now.