When can you draw down project contingency?

Contingency is a control account

It is money set aside against identified risks that may or may not occur. That is all it is. It is not a buffer for optimistic estimating, and it is not a fund for things somebody later decided they wanted.

The test is simple and almost never applied: did a risk materialise, and is it documented? If the answer is no, the drawdown is not a contingency drawdown. It is a scope change wearing a disguise, and it should go through the process scope changes go through.

Scope creep is what contingency is actually spent on

And it is the spend contingency is least able to justify. A materialised risk has a paper trail — the risk was registered, it occurred, here is what it cost. A scope addition has an argument, and arguments are what get made when the money is convenient.

The discipline is to force the distinction at the point of drawdown, with tiered authorisation, because it cannot be recovered afterwards. By the time the contingency is gone, nobody can reconstruct which half of it was legitimate.

The burn curve is the early warning

Plot contingency consumed against percent-complete. If contingency is being drawn faster than the project is progressing, the project is in trouble — and it is in trouble now, not at the point the cost report eventually says so.

This is one of the very few genuinely leading indicators in project cost control. A contingency drained by month eight on a twelve-month project is not a budgeting problem. It is a forecast of the conversation you will be having in month eleven.

Terms

• Contingency