Overspend gets attention immediately. Underspend gets a comment about prudent management and no questions, which is why it is the more dangerous of the two: a project spending less than planned is frequently a project that has done less than planned, and the difference is not visible in a spend report.
Section 2.4.2 of the PMBOK® Guide Eighth Edition covers monitoring and controlling a project's finances, and the practical work is interpretation. Any variance has three possible meanings and they call for completely different responses.
Timing. The work happened and the money has not moved, or the reverse. An invoice held in a supplier's approval queue, an accrual that was not posted, a payment made ahead of the work. This is the most common explanation for a month's variance and the least interesting, and it is settled by asking the finance team rather than by managing anything.
A rate problem. The work is genuinely costing more or less per unit than planned: productivity below the estimate, a rate that was wrong at tender, a supplier charging at a higher grade than assumed. A rate problem continues until something changes, so it should be carried into the forecast for everything remaining, which is a large adjustment and an uncomfortable one.
Scope that arrived without a decision. Spend rising while the work list has quietly grown. This is the variance that matters most and looks least dramatic in any single month, because it appears as a modest overspend each period rather than as an event.
Spend without progress is not information. Half the budget consumed means nothing until it is compared with how much of the work is done, which is why every useful cost control conversation begins by establishing progress and not by reading the ledger. A project at sixty per cent spend and thirty per cent complete is in a completely different position from one at sixty and fifty-five.
Commitments are invisible in spend. Orders placed and not yet invoiced are money already gone from the project's point of view, and a report showing only invoiced cost will understate the position, sometimes by a large margin at the point where a major supplier has been engaged. Tracking commitments alongside actuals removes most of the surprises in the final quarter.
A bank's change programme was replacing the platform used by its business lending teams, with a phased rollout to eight regional offices. Four months into delivery it was running about nine per cent under its spend profile, and the monthly report noted this as favourable each time.
The underspend had one cause. The training workstream had not started. Rooms had been booked, materials had been printed, and the sessions had been deferred twice because the platform's configuration was not stable enough to train on. No training meant no trainer costs, no backfill for attendees and no travel, which was most of the variance.
Nobody had connected the favourable number with the stalled workstream, because the finance report and the delivery report were produced by different people and reviewed in different halves of the same meeting. The training room at head office stayed laid out between the deferred sessions: chairs squared to the tables, an unopened pad and a wrapped pen at every place, jugs upturned on a tray at the side.
The position became visible in month five, when the rollout plan needed trained staff in the first region. The training had to be compressed into three weeks, which meant additional trainers at a premium, backfill at short notice and two sessions run at a weekend. The eventual training cost came to about a fifth more than the plan, and the rollout slipped by five weeks.
The change the programme made was to report spend and progress on the same page, by workstream, with a line explaining any variance beyond a threshold in either direction. The first version of that page, produced retrospectively for month four, would have shown a training workstream at zero spend and zero progress against a plan that expected both.
Match the control effort to the exposure. A project with a handful of large committed contracts needs commitment tracking and very little else; one with many small discretionary spends needs a different kind of attention. Applying the same monthly cycle and the same thresholds to every project in a portfolio wastes effort at one end and misses movement at the other.
Set thresholds in advance and in both directions. A stated variance beyond which an explanation is required, applied to underspend as well as overspend, is what stops favourable variances from travelling unexamined for four months. The threshold matters less than the fact that it exists and is applied symmetrically.
For a PMP® candidate, the useful reading is that a cost variance requires interpretation against progress, so a scenario describing a project comfortably under budget is asking what has been delivered for that spend. A response that treats underspend as a positive has read one number without its pair. A structured PMP exam preparation course works on situations where the financial position looks healthy and the delivery position does not.
Look at your own project's largest favourable variance this month and find out what caused it. If the answer is that something has not happened yet, that is a schedule position wearing a financial costume, and it belongs in the delivery conversation rather than the finance one.
Favourable variances are the least examined numbers in project reporting and quite often the most informative. Omega's PMP® Exam Preparation works through financial control as interpretation against delivery.
Monitoring and controlling project finances is covered in the PMBOK® Guide Eighth Edition.
Ad · Amazon affiliate link.
A123: The PMBOK 8 Finance Performance Domain: What It Really Covers
A124: Cost Is Not the Same as Value
A125: Budget vs Cost Baseline vs Funding
A126: Contingency Reserve vs Management Reserve
A127: CPI and SPI Explained Without Formula Memorisation
PMP and PMBOK are registered marks of the Project Management Institute, Inc.