The Development Audit

An hour, once a year, done alone. It answers three questions most organizations have never asked and cannot answer from memory.

Part 1 — Who has actually been tested

List everybody in the organization. Against each name, write the last decision of consequence they owned — one they made, where being wrong would have cost something, and where the result was theirs.

Not projects managed. Not tasks completed well. Decisions.

Reading it: the blanks are the finding. Expect them to cluster among people who are geographically distant, on other shifts, in functions that never present upward, or simply quiet. That distribution was not chosen by anybody, which is exactly why it persists.

The uncomfortable version: count how many distinct people appear. Most leaders find the number is two or three, and believe they are developing a team.

Part 2 — Ninety-day cover

List the roles the organization cannot operate without. Usually between four and eight, and rarely the ones on a succession chart.

Against each, name the person who could hold it within ninety days if the incumbent were unavailable. A real name, not a role. If two names, write both. If none, write none.

Reading it: every “none” is a development priority with a deadline attached, and it is more urgent than any question about eventual succession to the top job.

The check: for each name you wrote, ask whether that person knows. Cover that exists only in the leader’s head is not cover.

Part 3 — What leaves with whom

Name everybody likely to retire or move on within five years. Against each, write what they hold that exists nowhere else.

Not their job description. The specific knowledge: why the arrangement with that supplier is structured that way, what the county actually does when a permit is contested, which reassurances from a particular lender turned out to be worthless, how the seasonal pattern really behaves in a dry year.

Reading it: this is the transfer problem in one page, with dates. Everything on this list is currently scheduled to leave the organization permanently.

The action: an afternoon with each person, recording a dozen cases in their own words. What was the situation, what did you know, what did you decide, what happened. It is the highest-return hour available to most institutions and almost nobody spends it.

What to do with the result

Not a programme. Three specific commitments for the coming year, each with a name attached.

One person from the blanks in Part 1 who will be given a real decision, and what it will be.

One role from Part 2 with no cover, and who will be developed into it.

One person from Part 3 whose cases will be recorded, and by when.

Three is the right number for the same reason it is elsewhere in this Library. Eleven commitments produce two that happen and nine that teach the organization that stated intentions here are decorative.

Notes on running it

Keep it private. An audit that becomes a reported exercise generates presentable answers within one cycle, and the blanks are the entire value.

Do not turn Part 1 into a list of designated people. The distinction is covered elsewhere in this chapter and it matters: this is a record of what the organization has done, not a prediction about who will succeed.

Repeat annually and compare against last year. The comparison is more informative than either year alone, because it shows whether the distribution is widening or whether the same two names keep appearing.

For an organization of a dozen this takes twenty minutes and is more useful rather than less, because at that size a single unfilled “none” in Part 2 is an existential exposure rather than an inconvenience.