Critical Role Register

What this tool does

CM8-352 works out which roles your business genuinely cannot do without, and how long you would actually be exposed if one emptied. It scores each role for criticality, adds up the weeks between somebody leaving and a replacement being useful, subtracts what notice and internal cover really buy you, prices the remainder, and flags the single points of failure.

Everything runs inside this single file — no account, no upload, no network request of any kind.

Roles, not people

This is a register of roles. The name of the current holder is recorded so the register is usable, and retention risk is recorded because it changes how urgent an action is — but nothing here scores a person, and nothing here should ever be used to rank individuals, inform an appraisal, or decide who to keep.

The distinction matters practically as well as ethically. "Dagmar is brilliant" is not a finding. "Only one person can release work against the control system, and it takes nine months to learn" is, and it stays true after Dagmar has gone.

What one row is

One row is one role. Where three people do the same job, that is one row with cover recorded as full — the whole point is that the role is not exposed. Where one person holds two distinct roles, that is two rows, because they will be filled separately.

Assess ordinary roles as well as obviously critical ones. A register containing only the frightening roles cannot show that anything was considered, and the sample includes a role that scores low on purpose.

How criticality is scored

Multiplying rather than adding is deliberate. A role with enormous impact whose skill is on every street corner is a recruitment problem, not a risk. A rare skill nobody would miss for a month is not a risk either. Only the combination is.

Judge impact by what happens in the first weeks, not eventually. Everything is survivable eventually.

The gap, and why it is longer than you think

Almost every organisation underestimates this, and it is nearly always the second half that is underestimated. Recruiting is visible, planned and diarised. The months between a competent newcomer starting and that newcomer actually doing the job unaided are invisible, and during them the role is partly empty while also consuming somebody else's time.

Include notice periods in the time to recruit — a good candidate you cannot start for three months has not solved your problem in week six. The stacked chart splits the two so you can see which half you are fighting: recruiting faster does not help a role that takes nine months to learn.

What cover is actually worth

Internal cover does not make the gap disappear. It makes part of it survivable, and the tool treats it that way:

A successor "ready in more than two years" contributes nothing, and that is not a mistake. If the role empties tomorrow, a successor two years away is not cover; they are a plan. Recording them is still worth doing, because it shows the plan exists.

Notice is subtracted last, because notice is the only part of the gap you get to use before the role is actually empty. A role with no notice period — a contractor, a retirement nobody saw coming, an illness — is exposed from day one, and setting notice to zero is the right way to model that.

Putting a number on it

Set a default in Settings and override it per role where the answer is obviously different. Think about what the week actually costs: output not produced, cover paid at premium rates, decisions not taken, a customer relationship cooling, a certification lapsing.

This figure is an argument, not an accounting entry. Its job is to make a cross-training programme that costs a few thousand look like what it is next to an exposure of forty. Do not present it as a forecast, and do not add it to anything.

The positions the tool works out

• Single point of failure — critical, nobody can cover it, no successor. The rows that justify the register existing.
• Critical, and leaving — the exposure is not hypothetical any more.
• Critical, retention risk — high retention risk on a critical role. The window to act is now and it is short.
• Nothing written down — sole holder of access or authority, with no documentation. The cheapest thing on this list to fix and the most often left.
• Exposure above tolerance — covered and documented, but the gap is still longer than the business says it can absorb.
• Critical, managed — critical, and everything reasonable is in place.

What to do about it

Four moves, roughly in order of cost:

• Write it down. Procedures, credentials, contacts, the reasoning behind decisions. Costs days, cuts weeks off time to competence, and is the only action that helps in every scenario.
• Remove sole access. A second authoriser, a second named contact at the customer, credentials in escrow. Usually an afternoon.
• Cross-train. Real cover, built deliberately, before it is needed.
• Change the role. Sometimes the answer is that one job should be two, or that a scarce skill should be bought in rather than held.

This register is sensitive

It contains names, a judgement about how likely somebody is to leave, and in effect a statement about how hard they would be to replace. That is personal information and some of it is the kind people would be upset to read about themselves.

Keep access to it narrow, be able to say why you hold it, and be honest that it exists — a succession and key person register discovered by accident does far more damage than one people know about. This is also why the tool never uploads anything.

What this does not do

• It does not predict who will leave. Retention risk is your judgement, entered by hand.
• It does not assess anybody's performance and must not be used as if it did.
• It does not know your obligations. Where a role must exist by law or by certification, that requirement stands whatever this scores.
• It does not model two roles emptying at once, which is the scenario that actually hurts.
• The relief percentages are stated above and are a convention, not a measurement. Argue with them if your experience differs — but change them for every row, not only the inconvenient ones.

Printing and sharing

The Report tab prints the tiles, charts and both tables with a title block you fill in. Filter to one area first if you are reviewing a department, or to critical roles before taking it to a board.

Saving your work

The register is held in this browser, on this computer, and stays there between visits. Use the backup button to write a JSON file you control; given what this register contains, keep that file somewhere access-controlled.

Accuracy & disclaimer

Every score, week and cost here is your own judgement, multiplied out. The tool has no view about whether a role is genuinely critical, whether a successor is genuinely ready, or whether anybody is genuinely leaving, and none of its figures are a forecast.