September 6, 2026 · 9 min read
TL;DR: Status questions test what you do with a number, not whether you can compute it. Analyze the cause before acting: escalating, adding resources, or logging a risk are all keyed wrong while the cause is unknown. Variance analysis explains the past, forecasting predicts the end state, and velocity never compares two teams.
A schedule variance of twelve percent lands on your desk. Four options: escalate to the sponsor, add resources to recover, log it in the risk register, or analyze the variance before deciding on anything. Three of those feel like action and one feels like delay, which is exactly why the third is the one the exam wants.
Work out what caused it. A variance says something moved; it says nothing about why, and the right response is different for every why.
Of the 75 status-tagged questions in this bank, 18 turn on a variance and 20 on earned value, against just three mentioning a status report. Status is not a reporting topic here, it is an interpretation topic. The distractors are consistent across at least five records: act now, escalate now, or record it somewhere and move on, each offered before anybody has established the cause.
Common trap: treating urgency as diligence. Adding resources to a schedule variance caused by rework makes the rework worse. Escalating a variance caused by a one-off supplier delay spends sponsor attention on something already resolved. The bank's explanations name this directly, and the exception proves the rule: when a stem describes a committed date that has genuinely changed, immediate notification is correct, because that is no longer a variance to diagnose but a fact stakeholders need.
Status says where the project stands now. A forecast says where it will end up, and it is built from the project's own measured performance rather than optimism.
Variance analysis belongs firmly on the status side. It compares actual results against a baseline that already exists, which makes it an explanation of past and present performance and explicitly not a forecasting method. Forecasting takes a measured index, such as the schedule performance index this project has actually been achieving, and projects it forward.
That distinction powers one of the better questions in the bank. A post-mortem shows a cost performance index of 0.8 and a schedule performance index of 1.25, and asks what happened. The keyed answer is that the project was terminated early, over budget and ahead of schedule.
The reasoning is worth internalising. A project that runs to completion ends with a schedule performance index of exactly 1.0, because every unit of planned value is eventually earned. An index still sitting at 1.25 after the fact means the project never reached completion, so it stopped early. The cost index of 0.8 says it was over budget while it ran. Candidates who read 1.25 as simply "ahead of schedule" pick the option saying it finished ahead of schedule, and miss that a finished project cannot show that number at all.
Because they usually contain the right number attached to the wrong operation, sign, or unit. The arithmetic is not what is being tested; knowing which operation the question asked for is.
The pattern appears in at least five records here and is mechanically exploitable once you see it:
| Distractor style | What it actually is | How to catch it |
|---|---|---|
| Right magnitude, missing minus sign | The variance without its direction | A negative variance means behind or over; check the sign says what you mean |
| A ratio where a currency figure was asked | An index reported as a variance | Variances are money or time, indexes are dimensionless |
| A currency figure where a ratio was asked | A variance reported as an index | Same check, run the other way |
| The inverted ratio | Divided in the wrong order | Earned value goes on top for both CPI and SPI |
Run the unit check before the arithmetic check. If the question asks for a cost performance index and an option is denominated in dollars, it is wrong regardless of how right the number looks, and this bank flags that pairing by name in its explanations.
The same discipline applies, with different instruments. Eleven of these records turn on burndown or burn-up charts, and the exam is careful about which chart answers which question.
Executives asking what has been completed and what remains across ongoing work are asking two questions at once, and the keyed instrument is a burn-up chart. It plots completed work rising from the bottom and total scope as its own line above it, so the gap between them is the remaining work and any movement in the scope line is visible rather than hidden. A burndown chart shows a single descending line, which conceals scope changes inside the same slope, and efficiency directives tend to change scope.
Velocity carries the sharpest trap in this group. When supervisors complain that Team A underperforms because its velocity is always lower than Team B's, the keyed answer is to explain that velocity cannot be used to compare different teams. Story points are assigned by each team during its own estimating, so the unit itself differs; comparing the numbers measures the calibration of two yardsticks rather than the productivity of two teams. Moving strong performers, pushing Team A to raise its number, and setting up a reward scheme are all offered, and all three act on a comparison that was never valid.
What should a project manager do first when a variance appears? Analyze the cause. Escalating, adding resources, and logging a risk are all keyed wrong while the cause is unknown.
Is variance analysis a forecasting technique? No. It compares actuals against an existing baseline, which explains past and present performance rather than predicting the end state.
Can you compare two agile teams by velocity? No. Story points are calibrated per team, so comparing velocities compares yardsticks rather than productivity.
When does a variance become a change request? Once the cause is understood and a baseline is genuinely affected. The trigger is the diagnosis, not the size of the number.
The 75 status questions in PMP Practice sit inside a bank of 2,141 re-certified against PMBOK 8 and the July 2026 Exam Content Outline, and every wrong answer carries the reasoning that makes it wrong, explained rather than just marked. Start the free 20-question sample with no card and no signup.
Related: Earned Value Management · Critical Path Method and float · Process domain study guide
Sources: PMI — PMP Examination Content Outline, 2026 (PDF) · PMI — PMBOK Guide standards
What should a project manager do first when a variance appears?
Analyze the cause before choosing a corrective action. In this bank, escalating immediately, adding resources, or logging the variance as a risk are all offered as options and all keyed wrong when the root cause is still unknown. A number tells you something moved; it does not tell you why, and the response depends entirely on why.
Is variance analysis a forecasting technique?
No. Variance analysis compares actual results against the baseline that already exists, so it explains past and present performance rather than predicting the future. Forecasting uses the project's own measured performance, such as a schedule performance index, to project where the work will end up.
Can you compare two agile teams by velocity?
No, and questions built on that mistake are common. Velocity is counted in story points a single team assigns during its own estimating, so the unit differs between teams. Comparing two velocities measures the calibration of two yardsticks, not the productivity of two teams.
When does a variance become a change request?
Once the cause is understood and the gap genuinely affects a baseline. The trigger is not a percentage. This bank keys a two-step gate: analyze the variance for its cause first, then, if a baseline is affected, raise a change request that sets out the gap, its cause, and the options, routed to the body that owns the baseline.