Counting What an Exit Actually Costs

Count the time as well as the money, because the two components nobody pays for — management hours and the months before a replacement is competent — are routinely a third of the total and are the reason nobody believes departures are expensive.

The pattern described in “Counting What an Exit Actually Costs” becomes useful only when the source data is consistent and interpreted with context. Teams exploring remote employee productivity monitoring through read the official overview can use time and project records to locate operational questions, while direct feedback, HR evidence and accountable review remain necessary to explain them.

The purpose of counting is not accounting. It is to make the case for doing the cheap things described throughout this collection, which fail to happen because they cost visible effort now against invisible cost later.

For an independent reference relevant to “Counting What an Exit Actually Costs”, consult the Microsoft data-loss-prevention documentation. Use it to test record quality, access, retention, fair process and exception handling against the organisation’s real departure workflow.

The seven components

Payments made directly: notice, accrued leave, any termination payment, and anything settled.

Replacement cost: advertising, agency fees, assessment time, and the recruiting manager's hours.

Internal time: management, HR, payroll and IT hours spent on the process and the exit itself.

External advice, where it was needed.

Cover during the vacancy: overtime, temporary staff, work not done.

Time to competence: the weeks or months before the replacement produces what the leaver produced.

Second-order effects: the colleagues who left afterwards, the client relationship that weakened, the knowledge that went.

Which ones appear anywhere

The first two, usually. Internal time appears nowhere, external advice appears in a legal budget detached from the departure, and the last three appear in nobody's figures at all.

That asymmetry explains a great deal of organisational behaviour. A process improvement with a visible cost is weighed against a problem with no visible cost, and loses every time.

Counting the time

It does not need a timesheet. For a representative sample of exits, ask the people involved to estimate hours: the manager, the HR adviser, the payroll officer, whoever handled access and property.

The totals surprise people. A contested process routinely consumes forty to eighty hours of management and HR time, and an uncontested one still takes ten to fifteen across four functions. Multiplied by the annual number of departures, it is a role or two.

Time to competence

The hardest to estimate and the largest for senior or specialised roles. The usable approach is to ask the manager how long it took the last person to become fully effective, and to apply a proportion of salary across that period.

Precision is not the point. An estimate that is roughly right changes the conversation; its absence means the cost is treated as zero, which is definitively wrong.

What to do with the number

Compare it with the cost of the things that prevent it: a checklist, a second checker on final payments, notes written at the time, a stay conversation, an exit review once a year.

Every one of those is cheap against a single avoidable departure or a single contested process. That comparison is the entire argument and it cannot be made without a number, which is why the counting is worth the afternoon.

The avoidable proportion

Not every exit is preventable and not every cost is avoidable. Some departures are right and some are overdue.

The useful question is narrower: of last year's departures, how many does the organisation now think it would rather have avoided, and what would that have taken? Even a rough answer to that identifies where the money actually is, which is usually in the first year of employment and in one or two teams.

Counting the ones that went well

A costing exercise that only looks at difficult departures produces a frightening number and no useful comparison. Cost two or three straightforward exits alongside them.

The difference between the two is the actual figure worth knowing, because it is what the process is capable of saving. In most organisations a clean exit costs a fraction of a contested one, and the gap is wider than anybody expects before they measure it.

Where the estimate is allowed to be rough

Every figure on this page can be argued with, and arguing about them is the usual way the exercise dies. Agree in advance that the estimates are estimates, that nobody will defend a decimal place, and that the purpose is a comparison rather than an accounting entry.

Then publish the assumptions alongside the number. An estimate with its workings attached invites correction of the workings, which is useful; one without them invites rejection of the whole thing, which is not.

Reporting it

Present it once a year, alongside the exit review, as a single figure with its components shown. Not as a criticism of anybody — as the cost of a process that mostly works and fails predictably in three places.

Framed that way it tends to produce action, because the three places are nameable and the fixes are small. Framed as a failure, it produces an argument about the assumptions in the estimate, which is a conversation that never ends and changes nothing.