Enterprise Intelligence Group
CPG Trade Promotion Evaluation: From Spend to Decisions
Review CPG trade promotion with an agreed baseline, funding and claims record, clear data limitations and a documented next decision.
A CPG trade-promotion review should connect an agreed commercial objective with the evidence available after the event. The question is not simply whether sales increased. It is what the business can reasonably conclude, which costs and claims remain unresolved, and what should change before the next commitment.
Start with one event and one decision. Keep recorded facts, estimates and unanswered questions separate.
Agree whose result you are evaluating
A manufacturer's trade-investment decision and a retailer's category decision are not interchangeable. Record the perspective, participating products and customers, event dates and objective before comparing results. Do not assume the manufacturer controls the retailer's shelf price or store execution.
Choose the decision the review must support: repeat the event, change its terms or mechanics, investigate an execution issue, or stop funding the same approach. Agree the measures and cost definitions with the accountable commercial and finance owners.
Build an event evidence record
Use this outline to assemble a review pack. It is a working structure, not evidence from an EIG client engagement.
| Record | What to capture | Check before interpreting it |
|---|---|---|
| Objective and agreement | Intended outcome, products, customer, dates and agreed support | Were the event scope and commercial terms changed? |
| Shipments and consumer sales | Sell-in and, where available, sell-through, with dates, units and coverage | Do the records describe the same products and period? Shipments alone do not demonstrate consumer demand. |
| Funding and claims | Agreed funding, accrued or settled amounts, deductions and disputed claims | Which amounts are confirmed, estimated or unresolved? Avoid counting the same discount or contribution twice. |
| Comparison baseline | Expected performance without the event, comparison method and assumptions | Are seasonality, distribution, prices, availability and overlapping activity reasonably comparable? |
| Execution and availability | Evidence of where the event ran, price/display implementation and stock constraints | Is the evidence complete, sampled, reported by another party or unavailable? |
| Decision and ownership | Conclusion, limitations, proposed change, approver and next review point | Which conclusions are supported, and which need more information before commitment? |
If retailer data are unavailable, mark the gap. Do not substitute a shipment increase for evidence of incremental consumer sales or fill missing execution information with an assumption that the plan was followed.
Challenge the baseline before calling uplift incremental
A baseline estimates what might have happened without the promotion. It is not an observed alternate outcome. Record how it was constructed and which conditions could make the comparison misleading.
Review the periods before and after the event, not just the promoted weeks. Examine possible purchases brought forward, changes in distribution, stock constraints, overlapping promotions and movements in related products. Treat those patterns as questions to investigate, not automatic proof of pull-forward, substitution or a halo effect.
When evidence permits, compare results under more than one reasonable baseline assumption. If the decision changes with the assumption, make that sensitivity visible instead of presenting one precise number as settled fact.
Keep financial reconciliation separate from causal conclusions
Reconcile the event's funding and claims with the agreed definitions. Ask finance to confirm which discounts, product costs, media, logistics, returns and other event-related items belong in the evaluation, and how they are treated.
State what the reported measure means. Revenue uplift, incremental contribution and return on trade investment answer different questions. If reporting ROI, identify the numerator, denominator, period and treatment of unresolved amounts; do not compare percentages calculated on different bases.
A reconciled spend record can explain what was committed or settled. It does not, by itself, establish which sales the promotion caused. Keep that distinction in the review and its recommendations.
Use technology with testable requirements
Existing planning and analytics tools may support the work. For example, RELEX describes promotion analysis covering baseline sales, vendor funding, switching, stockpiling and halo effects. This is a description of its published capabilities, not independent proof of an outcome or an EIG partnership or endorsement.
For any proposed software or AI-assisted workflow, test it against your data and agreed scenarios. Can reviewers trace an estimate to its inputs? Can they see missing records, changed assumptions and unresolved claims? Can an accountable person challenge a recommendation before it changes a commitment?
Define those requirements before selecting a tool. Do not assume an agent can establish incremental demand, reconcile every claim or approve commercial changes without validated data, controls and human accountability.
Carry the finding into the next decision
Close the review with a documented choice: repeat, adjust, investigate or stop. Name the owner, supporting evidence, unresolved conditions and approval required. Preserve the original plan so the next review can distinguish an approved change from an unexplained difference.
For the wider six-check framework and an illustrative next-event record, use the retail and CPG promotion review guide.
EIG can help define the evaluation requirements, map the supporting evidence and clarify how findings inform the next promotion decision. The initial conversation establishes whether a paid review is appropriate; scope, information access and confidentiality are agreed before work begins.
Explore the Pricing and Promotion Decision & Delivery Review, or discuss your trade-promotion review with EIG. Please do not send confidential trading terms or datasets through the general contact form.
The EIG Library · Insights · AI Maturity Assessment · Contact