Skip to content
Log in
← Glossary

contingency reserve

Also called: contingency budget

Money or time set aside in the plan for risks you have already named, sized from their likely cost. The project lead can use it when such a risk happens, without asking for a new budget.

A contingency reserve is money or time set aside in the plan for risks you have already named. Each named risk gets a rough cost and a probability on a written scale, and the reserve is sized from those, not from a round percentage. The project lead owns it and can use it when one of those risks happens, without asking for a new budget.

It differs from a management reserve, which covers surprises nobody can name today. That one usually sits with the sponsor, outside the lead's plan, and is released on a written request that says what happened and what was tried first. Keeping the two apart keeps both honest: the lead can act fast on the risks everyone agreed on, and the sponsor sees every truly unexpected cost.

A contingency reserve is not a pot for new features or for a general cost overrun. When a named risk fires, the lead spends from it and records what was used and why; when a risk passes without happening, its share stays unspent and the forecast improves. If a progress check shows the remaining work will cost more than the reserve can absorb, that is a decision for the sponsor, not a quiet draw on the reserve.

The same logic applies to time: one shared buffer held by the lead at the end of the critical path, instead of a private margin inside every task.

Example

Klasna's physics launch has a 600,000 UAH ceiling: 510,000 for six lines of work, a 50,000 contingency reserve held by the project lead, and a 40,000 management reserve held by Leo, the sponsor. The lead controls 510,000 + 50,000 = 560,000.

The contingency is meant for named risks: a second round of demo lessons if fewer than eight tutors pass, extra days from the outside reviewer, diagrams redone, more ads if the waiting list is weak. When the risk register was cleaned up in December, the two risks scored in money had an expected cost of:

  • ad prices rise in January: likely (80 %) × 30,000 = 24,000 UAH;
  • the test places learners at the wrong level: possible (50 %) × 24,000 = 12,000 UAH.

Together 36,000 UAH, inside the 50,000.

What the reserve cannot cover showed up at the 30 November check: at the cost rate so far the work would cost about 510,000 ÷ 0.853 = 597,726 UAH, which is 87,726 over its budget. The contingency covers 50,000 of that, and the remaining 37,726 would take most of Leo's 40,000. That is a sponsor decision, so the lead reports it with a plan to re-estimate the diagnostic test, where most of the gap sits.

Common mistakes

  • Sizing it as a round percentage. Size it from named risks, each with a cost and a probability on a written scale.
  • Mixing it with the management reserve. One is the lead's for known risks, the other the sponsor's for surprises; keep them apart.
  • Spending it on new features. A new idea is a scope change for the sponsor, not a risk that happened.
  • Quietly absorbing an overrun. If the forecast exceeds what the reserve can cover, take it to the sponsor early.
  • Not recording what was used. Each draw needs the risk it covered and the amount, or nobody can tell what is left.

Learn it in the course