Reserve Analysis

Cost technique · See it on the map

Runnable here

Setting aside one budget cushion inside the cost baseline for known risks, and another outside the baseline for the unknown.

When to use it

During cost baselining, using the project's risk register to size contingency reserve against identified risks, and using organizational policy or judgment about overall project uncertainty to size management reserve. Revisit both any time the risk register or the project's uncertainty picture changes materially.

When to avoid it

Don't collapse the two reserves into one undifferentiated cushion — that's the single most common way this technique gets misapplied, and it destroys the thing that makes reserve analysis useful: knowing which reserve an overrun should draw from and being able to report that draw-down honestly instead of quietly absorbing it into the baseline.

Steps

1. Walk the risk register and estimate a cost impact and probability for each identified cost risk.
2. Size contingency reserve from that impact-and-probability analysis (e.g. expected monetary value, or a simulation) and add it inside the cost baseline, kept visible as its own line rather than folded into work package estimates.
3. Size management reserve separately, as a percentage or judgment call against overall project uncertainty, and hold it outside the baseline and outside the project manager's unilateral authority to spend.
4. When a risk in the register materializes and its cost impact is realized, draw against contingency reserve and report the draw-down explicitly — don't just let the actual absorb it silently.
5. When an unforeseen cost outside the risk register hits, escalate a management reserve request through whatever authority controls it, rather than treating it as routine.

What it produces

• A contingency reserve figure, inside the baseline, tied to specific identified risks.
• A management reserve figure, outside the baseline, for the unknown.
• A running record of what's been drawn from each reserve and why.

Common pitfalls

• Merging contingency and management reserve into one number, so an overrun's source — an identified risk that materialized versus a genuine surprise — can't be told apart later.
• Drawing down contingency reserve without reporting it, so the reserve looks intact right up until a status report shows it's gone.
• Sizing contingency reserve as an arbitrary percentage of the estimate instead of from the actual risk register, so it bears no relationship to what the project's real risk exposure is.
• Treating management reserve as routinely accessible to the project manager rather than as a genuinely separate authority to draw against.

Worked example

A data-center fit-out has a $2.1 million baseline before reserve. The risk register identifies a $60,000 cost impact if a specific HVAC vendor's lead time slips (35% probability) and a $25,000 impact if a permitting delay pushes work into a higher-cost season (20% probability); expected monetary value sizes contingency reserve at $26,000, added inside the baseline for a BAC of $2.126 million. Separately, the organization holds a 5% management reserve — $106,300 — outside the baseline for genuine unknowns. When the HVAC vendor's lead time does slip and the cost impact lands at $58,000, the project draws it from contingency and reports the reserve as $32,000 of the original $60,000 risk-linked amount still remaining, rather than letting the actual cost simply rise unremarked.

Where it comes from: this technique is named by the PMBOK Guide, 6th edition. No clause number is recorded for it here — a guessed citation would be worse than none.