How TPM Software Forecasts Trade-Spend Liability and Measures Promotion ROI

TPM software makes two numbers measurable: what a scheme calendar is going to cost before the period closes, and what a promotion actually did cost once the claims settled. Both collapse when the data lives in spreadsheets — which is why measurement is the value story here.

Be clear about the boundary, because the two get conflated. Projecting liability is arithmetic on data you already hold. Predicting lift is demand modelling, and it is a different discipline with different inputs.

What promotion data can and cannot tell you

The distinction that matters is between baseline demand (what would have sold anyway) and incremental lift (what the promotion actually drove). It is the right analytical frame, and it is worth understanding — but separating the two credibly needs market and consumption data that a settlement system does not hold. A trade promotion management system knows what you agreed, what shipped, what was claimed and what you paid. It does not know what would have happened without the scheme.

So treat lift as something you measure carefully with your own market data, and treat cost — projected and settled — as the thing the promotion record can tell you precisely. That is the reliable half, and it is the half that decides whether a scheme is affordable. This builds on the trade promotion management pillar and the channel view in the CPG trade promotion guide.

How do you separate baseline from incremental lift?

This is analysis you run against your own sales and market data — not something a settlement system does for you. The method is simple to state and only workable with structured data: establish what the territory would have sold without the scheme, then attribute everything above it — minus borrowing from the future — to the promotion. A worked illustration for one SKU family in one territory, one scheme month:

Step	Measure	Illustrative value
1	Baseline (3-month pre-scheme average, seasonality-adjusted)	₹1.00 crore
2	Actual sales in the scheme month	₹1.35 crore
3	Gross lift (2 − 1)	₹35 lakh
4	Less: forward-buying (stocking that borrows from next month, visible as the post-scheme dip)	₹10 lakh
5	Net incremental lift	₹25 lakh

Two details decide whether the numbers mean anything. First, the baseline must be clean — a reference period free of other schemes, price changes and stock-outs, adjusted for seasonality (a festive-quarter baseline applied to a lean month flatters every scheme). Second, forward-buying must be netted: if the channel simply stocked up, sell-in spiked while sell-through did not, and next month's dip gives the borrowing away. That is why lift measured on primary sales alone is misleading and secondary data is the real signal — and why the ERP and channel feeds need to flow in continuously, per ERP integration for claims, rebate and TPM software.

How do you forecast liability from scheme calendars?

Demand is one forecast; the cost of the promotion calendar is the other, and it is the one that ambushes period-close. Every scheme on the calendar is a liability curve waiting to happen: expected attainment × payout rate × eligible base, accruing across the window. With the calendar, live sales and the accrual in one system, finance can project where scheme cost lands at period-end — per scheme and in total — while there is still time to act.

Illustratively: a quarter's calendar carries a volume scheme (projected ₹60 lakh at expected attainment), a festive display program (₹25 lakh committed) and a growth scheme whose live run-rate has drifted 20% above plan (₹36 lakh against a ₹30 lakh budget). The system's rolling liability forecast reads ₹1.21 crore against a ₹1.15 crore provision — a ₹6 lakh overrun flagged in week seven, not discovered in week fourteen. The planning rhythm that produces the calendar in the first place is walked through in the step-by-step trade promotion management guide.

ROI: measuring true promotion return

Promotion ROI = incremental revenue ÷ true promotion cost — and the "true cost" is the catch. It includes the settled claim value and the leakage a spreadsheet hides, so ROI built on an estimated cost is fiction. Clean claim and settlement data are the prerequisite; the transparent method is in the claims management ROI benchmark, and the cost side is quantified in revenue leakage in rebate programs.

Continuing the illustration above — net incremental lift of ₹25 lakh at a 30% contribution margin yields ₹7.5 lakh of incremental margin. Now build the cost side honestly:

Cost component	Illustrative value
Settled claim value (all tiers)	₹4.8 lakh
Settlement leakage (overpaid slabs, one duplicate claim, out-of-policy exceptions honoured)	₹0.6 lakh
Execution spend (displays, merchandising)	₹1.1 lakh
True promotion cost	₹6.5 lakh

Incremental margin ₹7.5 lakh − true cost ₹6.5 lakh = ₹1 lakh net return — a scheme that pays back, but barely, and only visible as such because leakage was counted. Run the same scheme on the planned budget of ₹4.5 lakh and it looks comfortably profitable; run it with typical spreadsheet leakage and it is quietly underwater. Notice that leakage alone flipped more than half the net return — which is why disciplined settlement (validation against the scheme rules, the documented settlement trail) is not just compliance hygiene; it is what makes the ROI number real. Comparing schemes on this basis also requires like mechanics — a slab scheme, a growth scheme and a display program have different cost shapes, catalogued in types of trade schemes in India.

Why spreadsheets can't do either

Need	Spreadsheet reality	With TPM software
Baseline vs lift	Guessed	Modelled from data
Liability forecast	Quarter-end surprise	Live accrual
True promotion cost	Estimated	Settled + reconciled
Next-scheme decision	Gut feel	What-if simulation

The analytics discipline mirrors rebate analytics on the rebate side.

What-if simulation improves the next scheme

Once the data is clean, you can price a draft scheme before funding it — running its rate table across several attainment assumptions to see what it would cost. This is deterministic recomputation, not prediction: you supply the attainment assumptions, and the model tells you the payout each one produces. Present any figures as your own measured results, not industry benchmarks.

In practice a what-if run prices a draft scheme across three volume scenarios before the circular goes out. Illustratively, for a proposed slab scheme:

Scenario	Assumed attainment	Projected payout	Versus a ₹6 lakh budget
Low	70% of partners hit slab 1 only	₹3.2 lakh	Well under
Expected	Mix across slabs matching last quarter's distribution	₹5.1 lakh	Under
Stretch	Heavy slab-2/3 attainment	₹8.4 lakh	Over

The stretch row is the one that saves money: it shows the scheme going over budget precisely when partners perform best — the moment to add a cap or reshape the slabs, while the scheme is still a draft. Note what the table does not claim: it prices attainment assumptions you supply, and says nothing about whether the scheme will generate that attainment. The same simulation discipline applied on the rebate side is described in rebate analytics; this forward pricing is also the single strongest affordability argument for mid-market teams weighing tooling — see affordable TPM for Indian SMBs.

Where ClaimDS fits

ClaimDS gives finance the live accrual, the settled claim data and the scheme calendar in one place, so projected liability and actual promotion cost are both readable from the same record. <!-- TODO: confirm capability wording with founder --> It does not predict demand or promotional lift, and does not optimise promotional mix — those are a different category of product. It sits alongside the CPG trade promotion guide and secondary scheme settlement.