In practice · 01 of 09
The people who understand this are leaving.
Two people understand a system and one retires in March. Every conventional answer to that problem needs the person walking out of the door.
Why the usual answer does not work
The standard response to key person risk is documentation. Write it down before they go. It fails for a reason that is structural rather than lazy: the person who holds the knowledge is the only person who can write it down, they are the busiest person in the team, and they are being asked to spend their last months producing a document whose quality nobody can check. The people qualified to review it are the same one or two people.
The second answer is shadowing. Put a successor alongside them. That works where the knowledge is procedural and fails where it is judgement, and it assumes you already know which systems need it. Most organisations do not. They know which individuals feel indispensable, which is a different and much less reliable list.
The third answer is to wait, and it is more common than either of the others. The retirement is known about for two years and nothing is done until the notice period, because there is no forcing event and no budget line called succession.
Start with evidence, not with a list of names
We evidence what genuinely depends on fewer than three people. Not who feels important, and not who has been there longest, but which parts of the estate have a contribution history, a support record and an operational footprint that narrows to a very small number of individuals. That is a factual question about the systems and the records around them, and it can be answered without asking anyone to nominate themselves.
The result is usually uncomfortable in a useful way. Some of the people everyone worries about turn out to be replaceable, because what they hold is documented elsewhere or genuinely simple. Some systems nobody was worried about turn out to have a single point of understanding, often because they are boring, stable and therefore never discussed.
Then design the succession, the transfer and the run model
Once you know where the concentration actually is, three decisions follow, and they are decisions rather than documents.
- What is transferred, and what is retired instead. Some of what the leaver holds is knowledge about logic that should not survive at all. Transferring it is paying twice.
- What the receiving team has to look like. A transfer into a team structured by skill fails differently from a transfer into a team that owns an outcome end to end. The second absorbs judgement. The first absorbs procedure and then discovers the gap during an incident.
- What has to be true before the handover is accepted. This is the gate most programmes never define, which is why so many transfers are declared complete on a date rather than on a condition.
Where this shows up in the money
Dual-running is the line item most likely to break an offshore or capability-centre business case. The plan assumes a knowledge transfer period, the transfer overruns because nobody defined what completion meant, and the organisation pays for both teams for longer than the model allowed. The saving does not disappear, it arrives late, and arriving late is often enough to make the original case look wrong.
Getting the concentration evidenced early is what closes that line item on schedule. It is also the part of the work that a delivery partner has the least incentive to do honestly, because a longer transfer is revenue to them and cost to you.
Common questions
Is this the same as a knowledge management or documentation programme?
No. A documentation programme produces a document and treats the document as the outcome. This treats the transfer as the outcome and the evidence as the way you prove it happened. Documentation is a by-product, and quite a lot of it is deliberately not produced, because the underlying logic is being retired rather than carried forward.
Our people are willing to help. Is that not enough?
Willingness is necessary and not sufficient. The constraint is rarely goodwill. It is that nobody can describe an estate they have not read, including the people who built it, because the parts they built are the parts they remember and the accumulated accidents around them are the parts that cause incidents.
When is it too late to start?
The honest answer is that the day the notice is given is late but not hopeless, and two years out is when it is cheap. The variable is not the leaver's calendar, it is whether you know where the concentration is. That part can be established quickly and does not depend on the leaver's availability.
When this comes up. Comes up with a named retirement date, a capability-centre move, or a vendor exit.
How it is delivered
Compass to establish the concentration, Vault to hold the evidence, Watch to keep the transfer honest after the programme has moved on. Each module is a fixed deliverable behind a go or no-go gate, and the baseline earns the design. The full set of modules is here.
Related situations
- Capability centre and GCC designClassify the capabilities before anything moves, and design the receiving model around what does not commoditise.
- Regulatory evidence and auditLocate where control logic actually sits against where policy says it sits.
- Independent estate assessmentBuild the estate picture independently, and separate what is locked in from what is portable.
Tell us what you are trying to land.
A short conversation about your situation and whether an independent accountable role is the right instrument. If it is not, we will say so. No deck follows automatically.