The Trap
A forecast is a spreadsheet, and a spreadsheet will hold any number you put in it. The distributor said 140,000 units if the spring program goes well. The chain’s buyer said she liked the samples. The Q4 promotion is “likely.” You add it up, and the year is 400,000 units, and the plant’s 300,000-unit tier is right there on the price sheet. The Forecast Trap™ is optimism converted into inventory.
Law III — Growth Eats Cash First™ — is blunt about it: forecasting is a cash decision. The forecast tells operations what to make, and production tells finance how much cash to commit, and once the cash is in a box it is not coming back until a shopper takes the box off a shelf. Law V — Buy Information Before You Buy Scale™ — says the forecast was a set of assumptions of different quality, and you funded all of them at the same quality, which was the highest.
The founder in the trap is not a bad forecaster. He is an accurate reporter of what everyone told him. The error is adding a purchase order to a compliment and a maybe and calling the total demand.
The tell
- The forecast is one number, not four.
- The largest line in it is an account that has not sent a PO.
- The run size was set to the forecast, and the forecast was set after the tier was chosen.
- Last year’s forecast was off by more than 30 percent and this year’s method is the same.
- Nobody has written down what has to be true for the number to arrive, or what happens if it does not.
The arithmetic
Sort the 400,000 into the four classes, and weight each by what it is.
| Class | What it is | Units | Weight | Weighted |
|---|---|---|---|---|
| Committed | Purchase orders in hand | 60,000 | 100% | 60,000 |
| Probable | The distributor’s stated plan, on last year’s pattern | 140,000 | 70% | 98,000 |
| Possible | The chain, “in conversation,” no PO | 120,000 | 25% | 30,000 |
| Speculative | The Q4 promotion, not yet approved | 80,000 | 10% | 8,000 |
| Forecast | 400,000 | 196,000 |
Two runs are on the table. A 300,000-unit run at $1.30 is $390,000 to the plant. A 100,000-unit run at $1.52 is $152,000. The tier saves 22 cents a unit — if the units sell.
The year arrives. Committed ships. The distributor takes 98,000, which is what 70 percent of probable looks like. The chain takes 24,000 in a regional test. The promotion is not approved. Sold: 182,000 units — almost exactly the weighted number, and not by luck. The 300,000 run leaves 118,000 units in the warehouse, $153,400 of cash in boxes, and the next label change and the next price move waiting behind it. The 100,000 run sells out in the spring, the second run is 100,000 on evidence, and the year ends with about 18,000 units of cover, $86,000 less spent at the plant, and the second $152,000 wired in April — after the first run’s sell-through and most of its collections — instead of in January.
Now run the Working-Capital Stress Test™ on the big run at the time you chose it. At 50 percent of forecast, 200,000 units, the run leaves 100,000 in stock. At 75 percent, 300,000, it just clears — if every class delivers three-quarters, including the promotion nobody approved. At 125 percent, you are out of stock by early fall and the second run’s deposit lands before the first receivable, which is a different trap. The run only worked at exactly 100 percent, and 100 percent of a forecast is the one outcome forecasts never produce.
The fix
Law III’s rule is that committed, probable, possible, and speculative demand should not receive equal inventory commitments. Law V’s rule is Staged Commitment™: break the bet into gates and release capital only as thresholds are met. Together they say: build to committed plus weighted probable, with a second run scheduled the day the possible becomes a PO.
Put the forecast’s assumptions on the Assumption Register™ — the distributor’s spring program, the chain’s test, the promotion — each with its evidence, its confidence, the downside if wrong, the cheapest test, an owner, and a decision date. Then run the Cash Gate™ with the four cases in it: What is peak cash need? What happens at 75, 50, and 125 percent? When is the next production commitment required?
The Gate answer this usually produces is GO SMALL — the run that covers committed and weighted probable, at the worse unit price, with the tier negotiated as a price across two runs rather than a size. Principle 09 — Find the Kill Assumption™ — is a rule about run sizes.
Split the current forecast into the four classes this week and write the weight next to each — no line goes in at 100 percent without a PO number beside it.
Size the next run to committed plus weighted probable, and put the second run’s trigger — a PO from the possible class — on the calendar with a name on it.
Run the four-case stress test — 50, 75, 100, 125 percent — on every run over three months of proven demand before the deposit is wired.
The Seed File is the check for this door: the company six months after its first batch, with a distributor’s forecast and a second run in front of it, and the file an investor will read. The First Batch™ Workbook’s Template 8, 08_Cash_Cycle_and_12_Month_Plan.xlsx, puts every run and every collection in the month it lands and finds the minimum-cash month. Template 7, 07_Velocity_Tracker.xlsx, turns the distributor’s plan into a door-by-door number you can weight. Later, the Build Line™ Workbook’s Template 7, 07_Management_Report_Pack.xlsx, carries the 13-week cash the four cases go into. Start with the check at /seed-file.html.
RAISE — this trap belongs to The Seed File. Take the check at /seed-file.html.
Take the Seed Readiness Check →
