TPM Software Won't Fix Trade Spend You Can't Measure

Trade spend at Cinderhaven Provisions runs $5.3 million a year, all in, just over a fifth of its $25 million in wholesale revenue, and the brand cannot tell you which of its promotions made money. Trade promotion management software would give it a dashboard. It would not give it that answer, because $380,000 of that spend leaks into a gap the software does not reach.

Cinderhaven Provisions is invented and its ledger is synthetic; the way trade spend leaks between off-invoice, bill-back, and deduction is not. Most of that $5.3 million is structural (slotting, everyday allowances, manufacturer chargebacks) rather than a heavy promotional calendar. CPG companies put about 20 percent of revenue into trade promotion; even a light promoter carries the full line, and still cannot see the bottom of it.

McKinsey, working from Nielsen's promotion data, puts the share of US trade promotions that lose money at 72 percent. The reason most brands cannot name their own losers is timing: the analysis is not ready until the quarter closes, by which point the same tactics are already locked into next quarter's plan. Software is sold as the cure. It measures what it can reconcile, and reconciliation is the work the brand still has to do first.

TPM software measures the spend it can see

A trade promotion management system does something useful. It replaces a folder of spreadsheets with one planning surface, holds the accruals, and computes each promotion's return by netting the money committed to it against the incremental volume it drove over a baseline. Point it at a clean event and it will tell you whether that event paid back.

The number it produces is only as good as the three inputs behind it: what the brand committed, what actually ran at shelf, and what the retailer deducted. TPM software owns the first. It automates the trade-promotion process where the money is planned and approved. The other two live in scan data and remittance files it reads secondhand, if at all.

A tool that plans the spend does not reconcile the spend.

The gap is between three numbers that never meet

Trade spend is one of the largest line items on a CPG P&L, and it is paid in ways designed not to line up. An off-invoice allowance comes straight off the invoice. A bill-back or scan deduction arrives weeks later, based on units the retailer says it sold. A promotion planned in March is deducted against in June, under a code that never names the promotion.

When those streams do not reconcile, money leaks in three directions. The first is the double-dip: a deduction for an event that was already paid as a bill-back, the same promotion funded twice, paid once as a bill-back or scan claim and deducted again against the off-invoice allowance, because the claim arrives in one inbox, the deduction in another, and nobody reads both. Cinderhaven runs about 20 deals a year where they overlap, at an average of $4,750 paid twice, for $95,000.

The second leak is deductions that tie to no promotion at all: roughly $150,000 a year in trade codes that reconcile to nothing anyone planned, written off because chasing them costs more than they return, which is one of the ways small food brands lose control of trade spend. The third is money spent on promotions that never lifted: about $135,000 in off-invoice allowances funded for features and displays that did not run, or ran and did not move units. Add them up and the reconciliation gap is $380,000.

Three numbers, three timelines, and the software owns one of them. The other two are where the $380,000 goes.

A dashboard over unreconciled data is a confident wrong answer

Put a TPM system on top of that and it computes ROI on the roughly $4.9 million it can see. The $380,000 in the gap never enters the calculation, so the dashboard returns a number that is precise, current, and wrong by nearly $400,000. It will show a promotion as profitable while the double-dip that erased its margin sits in a separate deduction the tool never tied back.

| Leak the software can't net out | Mechanism | Annual | |---|---|---| | Double-dips | same deal claimed off-invoice and again as a scan or bill-back deduction | $95,000 | | Unmatched deductions | trade codes that reconcile to no planned promotion, then written off | $150,000 | | Funded, not executed | off-invoice money on promotions that never lifted at shelf | $135,000 | | Trade spend the dashboard cannot see | | $380,000 |

The result is the same 15 to 25 cents on the dollar that disappears between the invoice and the bank, now rendered as a chart. The 72 percent traces to measurement, not to the promotions themselves, and a faster dashboard only makes the losing promotions easier to renew.

Speed does not correct the number. It just renews the losing promotion on schedule.

Buy the software after you can measure, not to measure

None of this is an argument against TPM software. It is an argument about order. The software is a multiplier, and a multiplier applied to unreconciled data returns unreconciled answers faster. Reconcile first, then automate what the reconciliation proved out.

Reconciling is hand-work: matching every deduction to the promotion that justifies it, every scan claim to its off-invoice allowance, and every planned event to what actually ran. Do that for one quarter and two things happen: the double-dips surface, and the promotions that never paid back stop getting renewed on faith. Then a TPM system has something true to automate, and its dashboard means what it says.

Buy the tool to scale a measurement you trust. Do not buy it to create one.

Reconcile one promotion

Choose the promotion your team is proudest of from last quarter. Its plan, its scan data, and the deductions the retailer took against it are three files that rarely agree, and I reconcile them into the one number a dashboard cannot show: what the promotion actually returned once every deduction is tied back. The proud ones are where the double-dips hide. Reconcile one promotion.