How do you measure whether an in-store promotion lifted sales?
Measure promotion lift by comparing exposed shoppers or stores against a control group, so the sales difference can be attributed to the promotion rather than to normal buying behaviour. Incremental sales lift uses statistical techniques to show how many product sales were driven directly by the campaign and were net new to normal sales behaviour, which makes it a causal metric, unlike ROAS, which correlates exposure with later purchase 1.
Designs that isolate the promotion
- Compare purchase behaviour and basket size of loyalty programme members who received the targeted in-store promotion against members who did not 2.
- Compare sales and traffic in stores that ran the campaign against stores in other regions or demographics that did not 2.
- Run multivariate tests that assign geographies to test and control cells across media, such as in-store audio combined with onsite product listing ads, and read sales per household against the control 2.
- Model promo uplift by analysing customer behaviour during promotion and no-promotion periods 3.
What to read alongside the sales number
Sales data works best when paired with store-level proof-of-play reporting and datasets such as impressions, dwell time and engagement KPIs like attention time or product pick-ups 2. A promotional plinth can be assessed on sell-through over time, and dwell time can be compared against similar products to understand in-store performance 4. Post-campaign analysis should extend past the advertised product to the total category, share and source of share, so cannibalisation, new shoppers and trade up or down are visible 2. Confirm the conversion methodology with a third party or align it to industry best practices 2.