Gaps
The same gaps, portfolio after portfolio.
You have sat in the meeting where the plan was renegotiated from scratch — again. Your portfolio owes two auditable answers: what each piece of work won its place against, and who signed for it.
If it cannot produce them, that is not a failing — it is the normal state, and it is fixable. Open any gap for how it shows up, what it costs, and how we close it.
In the demand tier
One cause: nothing stands between an ask and the plan.
01Demand arrives through many doors, and none of them is a gate.
How it shows up. A steering committee, a corridor conversation, a board paper someone has already promised, an escalation, an inherited plan. Each route is legitimate on its own terms. None of them compares the ask against anything else.
What it costs. Work enters the plan without ever being ranked against the work it displaces. Nobody can point to the moment it was chosen, because there wasn't one.
How we close it. One front door. Every ask — whichever door it arrived through — lands in the same register and is compared against what it displaces before anything is committed. The route stops deciding the outcome.
02Capacity is asserted rather than measured.
How it shows up. A number is agreed in a planning session and then quoted back all year. It is rarely reconciled against who is actually available, for how long, at what allocation.
What it costs. The plan is committed against capacity that does not exist. The overrun is discovered in delivery, where it is expensive, rather than in planning, where it is free.
How we close it. Capacity becomes a measured position — who is actually available, at what allocation — reconciled each cycle, so the plan commits against what exists rather than what was asserted.
03The plan is a negotiation, not a decision.
How it shows up. What survives is what its sponsor defended most effectively, not what scored best against declared criteria. Seniority substitutes for evidence.
What it costs. The portfolio optimises for internal influence rather than value. The same argument is had every quarter because nothing was ever actually settled.
How we close it. Declared criteria and a scored comparison, with the ranking defended in writing before commitment. Seniority can still argue — but it argues against a record, not into a vacuum.
04No record of what earned a place in the plan.
How it shows up. Ask why a programme is in this year's portfolio and you get a name, a date, or a shrug — rarely a business case anyone can produce.
What it costs. Nothing can be defended when it is challenged, and nothing can be stopped when it should be, because there is no baseline to measure against.
How we close it. Every entry in the plan carries what it won its place against: the comparison, the basis, the signature. When it is challenged, the defence is produced, not reconstructed.
05No record of what proved it deserved to stay.
How it shows up. Approval is a one-time event. Once in the plan, work is reviewed for progress, not for whether it still deserves the slot.
What it costs. Zombie programmes. Investment continues because stopping requires a decision nobody is positioned to make.
How we close it. Approval expires. Work re-earns its slot at gates, and stopping becomes a decision the record supports rather than a confrontation nobody is positioned to have.
The common thread
Every one of these is a missing record, not a missing process.
Organisations rarely lack governance. They lack a decision anyone can produce afterwards — which is the only thing that makes governance checkable.
What the gaps cost together
Each gap has its own price. They are not independent — they compound in one direction.
Read separately, they look like a list of fixes. Read in sequence, they are one cascade: the decision that was never made at the front door is paid for at the far end, in capital and in the change capacity it consumed.
In governance itself
About independence — who reviews, who signs, and whether either can produce an inconvenient answer.
06Governance exists as a document rather than a gate something can fail.
How it shows up. A framework is written, circulated and approved. Stages have names. In practice no piece of work has ever been stopped at one of them.
What it costs. The cost of governance is paid — the meetings, the packs, the templates — without the benefit. A control that cannot fail anything is a reporting overhead.
How we close it. Governance is rebuilt as gates something can fail — set where the work is still cheap to change, with the refusals recorded. A gate that has never stopped anything is renamed reporting.
07Sponsors mark their own homework.
How it shows up. The person who wants the work also chairs the review of the work, approves its business case, and reports its status.
What it costs. Independence is structurally impossible, however capable and well-intentioned everyone involved is. It is a conflict of position, not of character.
How we close it. The review is structurally separated from the sponsor: independent challenge before signature, and the challenge is kept with the decision.
08Designing, building and reviewing the same work is a structural conflict.
How it shows up. One organisation produces the design, implements it, and signs off that it was done correctly.
What it costs. The review cannot be adversarial, because the reviewer carries the cost of finding something. This applies to this practice too — which is why critical designs here are challenged from outside before they are signed, and the challenge is kept.
How we close it. Design, build and review are held apart — nothing is signed off by its author. We hold ourselves to the same line, on the record.
09Eight disciplines, seven handoffs, and the reasoning gets dropped in the gaps.
How it shows up. Strategy hands to business analysis, which hands to process engineering, which hands to development. Each handoff is a document. None of them carries the reasoning that produced it.
What it costs. By the time something is built, the constraint that shaped the original decision has been lost. The build is faithful to the spec and wrong about the problem.
How we close it. One record carries the reasoning from strategy to build. The decision trail travels with the work, so the constraint that shaped the decision survives every handoff.
In the market that sells the fix
Why the others persist. Their gate decides how. Not whether.
10The tooling can govern demand. It is rarely switched on.
How it shows up. Portfolio and demand modules ship with the major platforms. The capability exists and is largely unused.
What it costs. Organisations buy the means to say no and then decline to use it, because using it means telling a sponsor no. The gap is not capability — it is the willingness to run it.
How we close it. The model runs on the tooling already in place — we switch the capability on and stand behind it, so the no has a place to happen that is not personal to a sponsor.
11An integrator's gate governs work that is already chosen.
How it shows up. The major firms sell governed delivery, and it is real. But a firm engaged to deliver has no mandate to reopen what was chosen before it arrived.
What it costs. The most expensive decision in the portfolio — what to do at all — sits outside the scope of the governance you bought.
How we close it. This practice sits above the delivery flow, not inside it. The gate governs whether, before any firm is engaged to deliver how — which is exactly the decision an integrator cannot reopen.
Start here
Start with the portfolio you already have.
Bring the register, the capacity position and the last few quarters of decisions. The first conversation is a read of which of these are live in your portfolio.
The first read ends in writing: which of the gaps are live in your portfolio, and the first gate that would expose them.