Trade Promotion Analysis: Measuring If a Promotion Paid Back

Trade promotion analysis is the work of answering one question after an event runs: did this promotion make more profit than it cost? Sounds simple. It is also the single most-botched calculation in CPG, because the obvious version of it, promoted-week sales minus normal-week sales, is wrong in three specific ways. Each of those three makes a losing promotion look like a win.

This guide is the analysis method itself. For where it sits in the wider process, see the trade promotion management overview; for the forward-looking version, see How to Forecast Trade Spend ROI for Promotions.

Trade promotion analysis: the core calculation

Trade promotion analysis comes down to four numbers. Get them right and the ROI falls out on its own. Get the first one wrong and nothing downstream is salvageable.

Error one: no real baseline

The baseline is the volume the SKU would have sold anyway. It is not last week's number, and it is not a flat average. It has to account for seasonality, trend, and the SKU's own distribution changes. Use "the four weeks before" as your baseline and you will misjudge any promotion that ran into or out of a seasonal swing. Lift measured against a wrong baseline is not a small error. It can flip the sign of the ROI, turning a loss into a reported win.

Error two: ignoring forward-buy

When shoppers, or the retailer, stock up at the promoted price, some of that promoted-week "lift" is just volume borrowed from the weeks after. A promotion that shows plus 40% in its week and minus 15% for the next four weeks did not sell 40% more. It sold a fraction of that and rearranged the calendar. Honest analysis measures a window wide enough to catch the post-period dip, then nets it out.

Error three: undercounting trade spend

Trade spend is not just the per-unit price reduction. It includes the feature fee, the display allowance, the ad scan: lump-sum payments that arrive later as deductions. An analysis that divides incremental profit by only the scan-down cost reports an ROI that is too generous, often badly so. Count every dollar the event cost, including the ones that surface months later on a remittance.

From a number to a decision

The point of trade promotion analysis is not the recap deck. It is the next calendar. Each analyzed event should land in one of three buckets: repeat as-is, repeat with a change (shallower discount, add a display, move the week), or kill. A brand that produces ROI numbers but never moves an event between buckets is doing analysis as a ritual, not as management.

Where Scout fits

Scout exists to get the hard number right: the baseline. It builds a no-promotion baseline for each SKU at each retailer from syndicated movement data, accounting for seasonality and trend, then measures actual movement against it over a window wide enough to catch forward-buy. It loads the full trade cost into the ROI, lump-sum fees included. What comes out is a per-event ROI you can defend to finance and act on in the next plan. For the framework view, see A Guide to Trade Promotions Effectiveness Analysis.