Pricing & Promotion
Pricing and promotion analytics
Five questions a brand team has to answer every period, the four measurement errors that make the answers wrong, and what Scout computes from your own POS and syndicated data.
The five questions
What did this promotion actually return?
Incremental units against a baseline that accounts for seasonality and competitor activity, costed with the trade spend and the deductions the retailer took, not the ones you planned for.
Where should the everyday price sit?
Base price elasticity measured on non-promoted weeks with distribution held constant. A different number from promoted elasticity, and the two get confused constantly.
How expensive do we look on this shelf?
Price index against an explicitly defined competitive set, weighted and unweighted, trended so a mix shift is visible as a mix shift rather than a competitive win.
Did the retailer take the price?
Which stores actually rang the promoted retail, and for how many weeks. Promotions that underperform for this reason are read as demand failures more often than not.
What did we steal from ourselves?
Cross-price effects inside your own lineup. A promotion that moves volume from your 32 oz to your 16 oz at a lower margin has grown nothing and reports a lift.
Four errors that make the answers wrong
A baseline that absorbs the lift
Baselines fit on a window that includes the promotion pull the promoted weeks upward, so measured lift shrinks toward zero. Every promotion then looks mediocre and the pattern is invisible because it is uniform.
Distribution change read as price response
An item that gains 74 doors during a promoted period shows a unit lift that has nothing to do with price. Velocity per store per week is the only honest unit, and total-unit lift is the most common reporting default.
Deductions left out of the margin line
Planned promotional cost and what the retailer actually deducted are different numbers, sometimes by 20 percent or more. A promotion ROI computed on the planned figure is a forecast being reported as a result.
One elasticity per brand
Elasticity is per item, per retailer, per price zone. A brand-level average blends an inelastic hero SKU with an elastic secondary size and describes neither, then gets used to set both.
The second and third are worth a number. A 16 oz fermented salsa promoted from $6.49 to $5.99 across 412 Sprouts doors read as a 30.3 percent unit lift and an elasticity of -3.94. Holding distribution constant and stripping the two weeks a competitor ran a $3.99 endcap put the price-only elasticity near -1.6. The promotion did clear its hurdle, with contribution up 6.4 percent against a 22.5 percent break-even, but at -1.6 the price cut itself only accounts for about 12.5 percent of the volume. It made money on a competitor’s traffic, and the same $5.99 as a permanent shelf price would lose money on every unit. The full reconciliation is in price elasticity in CPG.
Where Scout fits
Scout computes base and promoted elasticity separately, per item per retailer, from your own POS and syndicated feeds. Distribution is held constant, so a door-count change never reads as price response, and competitor promotions overlapping your window are flagged rather than silently absorbed into your result.
Promotion results carry the deduction cost through to the contribution line, so the number the team reviews is what the promotion returned after the retailer took what it took. Price index runs against a reference set you define explicitly and is trended alongside distribution, which is what separates a genuine competitive move from a mix shift.
Product pricing analytics is the item-level half of this. Scout carries price, cost and margin per item per retailer, so a recommendation arrives with the contribution it changes attached rather than as an elasticity coefficient on its own. Pack architecture reads the same way. A 16 oz at $3.49 is $0.218 an ounce and a 32 oz at $5.99 is $0.187, a 14 percent per-ounce gap the 16 oz buyer can see on the shelf, and the cross-price effect between the two sizes is measured rather than assumed.
One boundary, stated plainly. Nobody sets the shelf price at a retailer they do not own, and brand-side work is no exception: Scout models and recommends the pricing and promotion decision, and it does not manage trade-promotion execution or claim settlement. Those stay in your TPM and your systems of record; Scout is the analytics layer that tells them what to do and what the last one was worth. (Retailers running their own stores are a different case: there Scout can hold the pricebook itself and push the price file. That purchase is a price optimization software decision, not this one, and it sits alongside retail operations software.)
Related: Retail price index · Trade promotion optimization · Sales cannibalization · Retail execution software
Tell us what you’re working on
A 30-minute conversation to scope fit. Pick a time that works for you.