Budget variance analysis

Budget variance analysis is the process of measuring the difference between what was budgeted and what actually happened, then explaining why the gap occurred. If a team budgeted 100,000 pounds of cost and spent 115,000 pounds, the variance is 15,000 pounds, or 15 percent over budget. The analysis then asks which line items caused it and whether the cause was price, volume or timing.

Variance can be favourable or unfavourable. Spending less than budget or earning more than budget is favourable. Spending more or earning less is unfavourable. The label depends on the line: lower cost is favourable, lower revenue is not. A complete analysis reports the size, the direction and the cause, not just the number.

The base calculation is actual minus budget, reported in both absolute terms and as a percentage of budget. The percentage matters because a 15,000 pound variance is trivial on a 5 million pound line and severe on a 50,000 pound line. The deeper work is breaking the total variance into the price, volume and timing effects that produced it.

1. 1

   Set the budget baseline

   Lock the budgeted figure for the period. Use the approved plan, not a later revised forecast, or the variance loses its meaning.

2. 2

   Capture the actual figure

   Record the actual result for the same period and the same scope, so the two figures are genuinely comparable.

3. 3

   Compute absolute and percentage variance

   Subtract budget from actual for the absolute gap, then divide by budget for the percentage. Note whether each line is favourable or unfavourable.

4. 4

   Split the variance into causes

   Separate the gap into price effect, volume effect and timing effect so the explanation points to a specific lever rather than a vague overspend.

A total variance is rarely caused by one thing. A cost overrun might be part higher unit price, part higher volume, and part spend that slipped from last period. Reported as a single number it hides the cause. Laid out as a tree, the headline variance decomposes into the price, volume and timing branches that produced it, and each branch points to a line and an owner.

KPI Tree models the variance this way by connecting each branch to the team that controls it. With RACI ownership on every node, an unfavourable variance does not sit in a spreadsheet waiting for the month end review. The system pushes the gap to the accountable owner when it moves, and the verified impact loop checks whether the corrective action actually closed it rather than assuming it did.

There is no universal target variance, because a tight operational budget tolerates less drift than an early stage plan built on assumptions. The ranges below are common tolerance bands used to flag variances for review. The threshold for action should scale with the size and predictability of the line.

Better variance analysis is less about smaller variances and more about faster, clearer explanation. The aim is to know the cause of a gap while there is still time to act on it. The practices below shorten the path from a number being off plan to someone owning the fix.

1. 1

   Comparing actual to a revised forecast

   Quietly swapping the baseline for a later forecast makes every variance vanish. Hold the approved budget as the reference.

2. 2

   Reporting absolute only

   A large absolute variance on a large line can be immaterial. Always pair the absolute figure with the percentage of budget.

3. 3

   Treating favourable as always good

   Underspend can mean work that did not happen. Trace favourable variances to the outcomes they affected before celebrating them.

4. 4

   Stopping at the total

   A headline variance with no breakdown gives no one an action. Split it into causes or it cannot drive a decision.