Trade schemes are the largest discretionary spend in most FMCG P&Ls after the product itself — commonly several percent of revenue — and the least controlled. The money moves through discounts, free quantity and claims across thousands of invoices, which is why the honest question is rarely "which scheme?" but "where did the scheme money actually go?"
This guide covers the scheme types, the leakage arithmetic, and the settlement discipline. For the definition, see what is trade scheme management; for the product surface, schemes & promotions.
1. The five scheme types
Nearly every trade scheme in general trade is one of five shapes, or a combination:
- Flat (on-invoice) discount. A percentage or amount off, for a period or a product set. Simple, visible, and the easiest to stack accidentally.
- Slab schemes. The discount steps with quantity or value — buy 50 cases for 3%, 100 for 5%. The workhorse for driving order size; the slab boundaries are where games are played.
- Quantity purchase schemes (QPS) / free quantity. Buy 10, get 1 free. Moves the benefit from price to stock, which protects the price ladder but makes the true cost easy to mis-book.
- Combo / basket schemes. The benefit requires a mix — a new SKU alongside the runner. The standard tool for forcing distribution of launches.
- Display and visibility schemes. Money for shelf conditions — a paid display, a planogram commitment — settled against execution rather than purchase. These are claims by nature, and the ones that most need evidence.
Layered on top: who funds it (brand, distributor, or shared) and who benefits (retailer, distributor, or the rep through incentives). A scheme is not designed until all three are answered.
2. Where the leakage happens
Scheme leakage is rarely theft; it is arithmetic done in too many places. The recurring failures:
- Stacking. Two overlapping schemes both apply because nothing enforces exclusivity. A 3% and a 4% scheme on the same invoice is a silent 7% programme.
- Slab gaming. Orders split or merged to cross a boundary — 100 cases as two fifties, or two retailers pooled into one bill.
- Manual application. The rep or the distributor's biller applies the scheme from memory. Some eligible outlets never get it (goodwill cost), some ineligible ones do (money cost), and no one can say which afterwards.
- Claims without evidence. The distributor claims reimbursement for scheme cost passed on; the brand cannot line-verify it and settles a negotiated fraction months later. Both sides book the difference as mistrust.
The arithmetic worth running: a brand spending 4% of revenue on schemes, with a fifth of it misapplied or unverifiable, is leaking 0.8% of revenue — usually more than the net margin gained by the schemes' volume effect. Leakage is measured per scheme as claimed cost − evidenced cost; if that number cannot be produced, the control does not exist.
3. Design rules that survive the field
- The server computes the benefit. Eligibility, slab, and free quantity resolve at order capture from rules — never from a PDF circular interpreted at a billing desk. In xMatix the scheme engine prices the order at save, offline included.
- Exclusivity is explicit. Every scheme declares what it stacks with. Default: nothing.
- Windows are dates, not months. Scheme periods with precise start and end stop the quiet back-dating that inflates the last week of a quarter.
- The benefit lands on the document. Discount lines and free-quantity lines appear on the invoice itself, so the scheme's cost is a ledger fact, not a quarter-end estimate.
- Display schemes settle on evidence. The payout references scored visits and shelf photographs, not a claimed compliance percentage.
4. How a claim should settle
Where the distributor funds the scheme up front and claims it back, settlement discipline is the whole game: the claim arrives line-referenced to the invoices that carried the benefit; the system matches claimed lines against the scheme's own computed applications; mismatches surface as named exceptions, not a lump-sum argument; and the approved claim posts as a credit note against the distributor's account. Month-end becomes reconciliation, not negotiation — the same evidence-first flow as collections. The machinery that generates, nets and settles these claims is described in trade claims for FMCG.
5. Measure the scheme, not just the spend
A scheme is a hypothesis — this benefit produces that behaviour — and it should be read that way: uplift during the window against the outlet's own baseline, and against sell-through, because a scheme that loads the distributor without moving secondary sales has bought inventory, not demand. Spend per incremental case, by scheme and geography, is the number that decides next quarter's programme; with schemes computed and settled in the system, it is a report rather than a project.
Where to start
Sequence for getting control: put current schemes into rules (most businesses find overlaps in the transcription itself); turn off manual scheme entry at billing; run one cycle with claims settling line-to-line; then start measuring uplift per scheme. The taxonomy stays the same — what changes is that every rupee of it becomes attributable.