A driver tree for an equipment business with recurring revenue, the next year it forecasts, and the ten per cent test that ranks the drivers.
A driver-based forecast writes each revenue line as a volume the business controls times a rate it can measure, then forecasts the drivers instead of the line. An illustrative equipment business sells 4,800 instruments at 92,000 and earns 16,600 a year on each of the 24,000 instruments already installed: 840.0 million. Run forward, the same tree says recurring revenue grows 12.0 per cent next year, and that a 10 per cent move in the installed base is worth 24.18 million of gross profit, a quarter of EBIT.
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 top-down forecast says "consumables grow 5 per cent". A driver-based one says "consumables are the installed base times what each instrument consumes, and the base is last year's base less what was retired plus what was sold". The second takes longer to build once and is faster every month after, because when the number moves you know which owner to call.
| Driver | Value |
|---|---|
| Instruments placed in the year | 4,800 |
| Instrument price / standard cost | 92,000 / 62,560 |
| Installed base, start of year | 24,000 |
| Attrition of the base, a year | 8% |
| Consumables revenue per installed instrument | 11,333.33 |
| Service revenue per installed instrument | 5,266.67 |
| Gross margin: consumables / service | 68% / 45% |
| Operating expense | 290,000,000 |
The two rates are derived, not assumed: the budget's 272,000,000 of consumables and 126,400,000 of service divided by the 24,000 instruments in the base at the start of the year. That is the first discipline of a driver tree: every rate must reconcile to a number the business already reports.
=Base*(1-Attrition)+Placements.With placements and prices held, next year's revenue is 887,808,000, up 5.69 per cent. The growth sits entirely in the recurring lines and was decided by this year's placements, not next year's. A top-down plan that grows recurring revenue by 5 per cent would forecast 418,320,000 and miss 27,888,000 that the base already implies, as long as the rates and attrition hold.
Of the 4,800 placements, 1,920, or 40 per cent, only replace retirements. With no placements at all the base falls to 22,080 and recurring revenue falls 8.0 per cent. That figure is invisible in a forecast built on growth rates.
The tree earns its keep when each driver is moved by the same 10 per cent and the effects are ranked. The answer tells you which drivers deserve a named owner and a monthly review.
| Driver moved 10% | Revenue (m) | Gross profit (m) | Share of EBIT |
|---|---|---|---|
| Instrument price | 44.16 | 44.16 | 47.41% |
| Installed base | 39.84 | 24.18 | 25.96% |
| Consumables per instrument | 27.20 | 18.50 | 19.86% |
| Placements | 44.16 | 14.13 | 15.17% |
Price and placements move revenue by the same 44.16 million, but price falls straight to gross profit and placements carry only their 32 per cent margin. Placements also have a second effect the single-year table hides: 480 more instruments add 7,968,000 of recurring revenue and 4,836,800 of gross profit the following year, and every year after, less attrition. Each installed instrument is worth 10,076.67 of gross profit a year.
Attrition works the same way in reverse. At 10 per cent instead of 8, 2,400 instruments retire instead of 1,920, and the 480 lost take 7.968 million of revenue off next year. A two-point change in a rate nobody reports costs exactly what 10 per cent more placements would have added.
Rates that do not reconcile. If 24,000 × 16,600 does not equal the recurring revenue in the ledger, the tree is wrong before it forecasts anything. Rounded rates leak: 11,333.33 multiplied back gives 271,999,920, 80 short, which is fine for a forecast and fatal for a reconciliation check. Keep the unrounded rate in the cell and round only the display.
Too many drivers. A tree with forty leaves is a second ledger. Keep the drivers that pass the ten per cent test and that someone in the business owns, and hold the rest as fixed lines.
Treating the sensitivity as a forecast. The ten per cent rows hold every other rate constant. In practice a deeper discount buys more placements, so price and volume move together; the bridge that separates them after the fact is in how to split a gross margin change into mix and rate. Once actuals arrive, test the forecast itself with forecast bias and MAPE.
The driver tree, the attrition roll and the ten per cent rows are live in the free workbook for this case, with the base and the rates as inputs.
Top-down grows last year's line by a rate; driver-based rebuilds it from volumes and rates the business manages. Growing 398.4 million of recurring revenue by 5 per cent gives 418.32 million. The installed base, 26,880 instruments at 16,600 each, gives 446.21 million: 27.89 million more, because the base grew faster than the top-down rate assumed.
Only those that move the answer and that someone owns. Apply a 10 per cent move to each and keep those that matter: here price is worth 44.16 million of gross profit, the installed base 24.18 million, the consumables rate 18.50 million and placements 14.13 million. A driver worth less than a few per cent of EBIT belongs in a fixed line.
Because it decides how much of each year's sales only replaces what was lost. At 8 per cent, 1,920 of 4,800 placements, 40 per cent, just hold the base flat. Raise attrition to 10 per cent and 480 more instruments drop out, taking 7.968 million of revenue and 4.84 million of gross profit off the following year.
The driver tree of chapter 9 of Financial Planning and Analysis is worked in the segment contribution workbook on its companion page. 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.