A budget split into twelfths reports seasonality as variance. A seasonality index built from two years of history puts each month where it belongs.
Work out each month's share of annual revenue in each of the last two or three years, average the shares, and multiply the annual budget by them. On an illustrative 24,000 (thousand) revenue budget, that gives a first quarter of 5,133, not the 6,000 a flat twelfth would set. Against actual first-quarter revenue of 4,980, the flat phasing reports a 17.0 per cent shortfall; the seasonal phasing reports 3.0 per cent, which is the real problem to manage.
Worked in full in Financial Planning and Analysis by Julian R. Sterling, with every figure reproduced in a free workbook.See the book on Amazon →
A fictional distributor has approved a revenue budget of 24,000 for 2026, in thousands, 11.1 per cent above the 21,600 of 2025. Its sales are seasonal: slow in the first quarter and in the summer, strongest from September to December. Two full years of monthly history are available. All figures are illustrative.
Divide each month by its own year's total. Working in shares rather than amounts removes growth from the pattern: 2025 was 8.0 per cent bigger than 2024, and averaging the raw amounts would let the larger year weigh more.
Share = month / year total; January 2024 = 1,300 / 20,000 = 6.50%
Weight = average of the shares; January = (6.50% + 6.71%) / 2 = 6.61%
Seasonality index = weight × 12; January = 0.79, meaning 79 per cent of an average month
Phased budget = annual budget × weight; January = 24,000 × 6.61% = 1,586
Excel, history in B:C and totals in row 14: =AVERAGE(B2/B$14,C2/C$14)*Budget
| Month | 2024 | 2025 | Share 2024 | Share 2025 | Weight | Index | Budget 2026 |
|---|---|---|---|---|---|---|---|
| Jan | 1,300 | 1,450 | 6.50% | 6.71% | 6.61% | 0.79 | 1,586 |
| Feb | 1,350 | 1,420 | 6.75% | 6.57% | 6.66% | 0.80 | 1,599 |
| Mar | 1,600 | 1,780 | 8.00% | 8.24% | 8.12% | 0.97 | 1,949 |
| Apr | 1,650 | 1,800 | 8.25% | 8.33% | 8.29% | 0.99 | 1,990 |
| May | 1,700 | 1,830 | 8.50% | 8.47% | 8.49% | 1.02 | 2,037 |
| Jun | 1,750 | 1,900 | 8.75% | 8.80% | 8.77% | 1.05 | 2,106 |
| Jul | 1,500 | 1,580 | 7.50% | 7.31% | 7.41% | 0.89 | 1,778 |
| Aug | 1,450 | 1,620 | 7.25% | 7.50% | 7.38% | 0.89 | 1,770 |
| Sep | 1,800 | 1,950 | 9.00% | 9.03% | 9.01% | 1.08 | 2,163 |
| Oct | 1,900 | 2,050 | 9.50% | 9.49% | 9.50% | 1.14 | 2,279 |
| Nov | 2,000 | 2,150 | 10.00% | 9.95% | 9.98% | 1.20 | 2,394 |
| Dec | 2,000 | 2,070 | 10.00% | 9.58% | 9.79% | 1.18 | 2,350 |
| Total | 20,000 | 21,600 | 100.00% | 24,000 |
The weights sum to 100.00 per cent by construction, so the months sum back to the approved 24,000 without any balancing figure. The index runs from 0.79 in January to 1.20 in November: the best month is expected to bring in half as much again as the worst.
| Month | Actual | Seasonal budget | Variance | Flat budget | Variance |
|---|---|---|---|---|---|
| Jan | 1,560 | 1,586 | -26 | 2,000 | -440 |
| Feb | 1,540 | 1,599 | -59 | 2,000 | -460 |
| Mar | 1,880 | 1,949 | -69 | 2,000 | -120 |
| Q1 | 4,980 | 5,133 | -153 | 6,000 | -1,020 |
On the flat phasing the quarter is 1,020 short, 17.0 per cent, and the business has a crisis meeting about a pattern it has seen every January for years. On the seasonal phasing it is 153 short, 3.0 per cent, a modest miss that still deserves an answer, because a budget 11.1 per cent above last year leaves little room for a slow start. The two phasings agree on the year; they disagree on every month, and a monthly report is read month by month.
The flat phasing distorts the other end too. It asks for 6,000 in the fourth quarter; the seasonal phasing asks for 7,023, 29.26 per cent of the year. A business delivering exactly its seasonal plan would look 1,023 behind at the end of September on the flat phasing and would then appear to recover it in the last three months: a reported catch-up that is only the calendar, and one that can hide a genuine fourth-quarter shortfall.
| Phasing basis | Q1 budget | Q1 variance on 4,980 |
|---|---|---|
| Flat, one twelfth a month | 6,000 | -1,020 |
| 2024 shares only | 5,100 | -120 |
| 2025 shares only | 5,167 | -187 |
| Average of 2024 and 2025 shares | 5,133 | -153 |
Once seasonality is in, the choice of history moves the quarter by at most 67. The big error is leaving seasonality out. Two refinements are worth making when they apply: strip one-off months (a large single order, a lost week to a system migration) out of the history before taking shares, and phase on working days where the calendar shifts, Easter being the usual culprit between March and April.
Phasing revenue flatly and costs seasonally, or the reverse. If revenue is phased by a seasonality index but payroll and rent by twelfths, the budgeted margin swings month by month for no operating reason, and the first quarter shows a loss that was never a forecast. Phase each line on its own driver: revenue on its seasonality, cost of sales on revenue, headcount on hiring dates, and fixed costs by twelfths. The rest of the reading discipline, against a restated budget at actual volume, is in how to calculate a flexible budget variance.
The book reads one quarter against five comparators, with the quarterly phasing of the budget and the plan side by side, in the free workbook for this case. For the year-end view built on the same phasing, see how to calculate a full-year forecast from YTD actuals.
Divide each month by its year's total, average that share over two or three years, and multiply by 12. An index of 1.00 is an average month. In the illustrative case January's shares of 6.50 and 6.71 per cent average 6.61 per cent, an index of 0.79, and November's index is 1.20.
Percentages. Averaging amounts lets the larger year dominate and carries growth into the pattern. Taking each month as a share of its own year removes growth, so the weights sum to exactly 100 per cent and the phased months add back to the approved budget, 24,000 in the illustrative case, without a balancing figure.
Because it assumes every month is one twelfth of the year. For a business whose first quarter is 21.39 per cent of the year, a flat phasing asks for 6,000 instead of 5,133, so actual revenue of 4,980 shows a 1,020 shortfall when the real one is 153.
This article is one calculation from Financial Planning and Analysis. The book takes the same case from first principles to the decision, chapter by chapter, and every figure it prints is a live formula in the free companion workbooks.
Get the book on Amazon →Free companion files
Also on Amazon UK · Amazon Germany · Amazon France · Amazon Canada
Reading guide: corporate finance, valuation and markets → · All 1340 articles →
If this book helped, or didn’t, a few lines on Amazon are worth more than they look: they are what the next reader goes on. Write a review. The workbook stays free either way.