Ask a leadership team who the organization could least afford to lose and you will get an answer in about four seconds. Ask what has been done about it and the room goes quiet.
That gap is not negligence. It is that the risk never presents itself as urgent. The person is there. The work gets done. Addressing it means slowing down the most capable person you have in order to prepare for a day that may not come. There is always something more pressing.
Then they take a new job, go on leave, or burn out, and the organization discovers what it had actually been depending on.
Skill is not the risk
The instinct is to think of this as a talent problem: they are very good and hard to replace. But skill is usually replaceable, given time and budget. What is not replaceable is everything that person knows and holds that was never captured anywhere.
The useful distinction is between what only they can do and what only they know. The first is a hiring problem. The second is a continuity risk, and it is the one that hurts.
Four kinds of concentration
Knowledge
Why the process has that odd extra step. Which client cannot be invoiced on the first of the month. What was tried three years ago and failed. This is the layer that no SOP captures, because nobody thought to write down the exception, only the rule.
Relationships
The funder who takes their call. The supplier who does the rush order as a favour. The long-standing client who stays because of a person, not a contract. These transfer badly and slowly, and often the organization does not know which relationships are personal until they are tested.
Authority
Decisions that formally belong elsewhere but in practice route to them, because they are trusted and it is faster. This one is invisible on the org chart and obvious in the calendar.
Access
Accounts, systems, vendor portals, the spreadsheet on their drive that half the department depends on. This is the most mundane category and it is the one that causes the most immediate damage, because it stops work the same week.
Why documentation alone does not fix it
The standard response is to ask the person to document what they do. It rarely works, for three reasons.
First, people cannot reliably describe expertise they have automated. Ask a fifteen-year veteran how they handle a difficult account and you will get a summary, not the judgment. The exceptions, which are the valuable part, do not surface because they do not feel like steps.
Second, documentation written but never used is not a transfer. It is a file. The only proof that knowledge moved is somebody else doing the work from the record and it going fine.
Third, asking a key person to document themselves is asking them to spend real time reducing their own indispensability, on top of a full workload. Some do it generously. Most produce something thin, and nobody wants to push.
What actually reduces it
- Find it before you fix it. Concentration is usually assumed to sit with the obvious person and turns out to sit with two others as well, in quieter functions. Finance, scheduling and client records are common surprises.
- Move access first. It is the fastest category to fix and the most damaging to leave. Shared ownership of accounts and records costs an afternoon.
- Transfer by doing, not by writing. Have a second person run the work from whatever record exists, with the expert available but not intervening. Every place they get stuck is a gap in the record, discovered cheaply.
- Name decision thresholds. Much of what routes to a key person routes there because no one else is allowed to decide. Writing the threshold moves the work without moving the person.
- Introduce relationships deliberately. A second name on the account, early, while everything is calm.
None of this is about distrusting the person. Most of the people at the centre of these concentrations are exactly who you want there. The point is that an organization should not be one resignation away from losing capability it took years to build.
How concentrated is your operation?
Five questions about continuity. Answer for the operation as it runs today.
Where this goes next
A short diagnostic surfaces a pattern. An operational audit tests it against how the organization actually runs, across leadership, the team and the records, then sequences what to address first. What follows the audit depends on what the findings justify, which is why the service ladder starts after the diagnosis rather than before it.
If replication is the reason you are asking, franchise readiness is the version of the audit built for that decision.
Writing about what the work reveals before an organization asks itself to carry more.