Target store promotion: pick products, design the offer and measure incremental sales

Target wants more top-line revenue from its physical stores. Plan a promotion aimed at that goal. Explain how you would run it, how you would decide which products or categories to put on promotion, and how you would set up an experiment to learn whether it worked. ### Clarifying Questions - Why the focus on in-store revenue specifically: to offset shoppers moving to online orders, or to get more out of fixed store costs such as rent and staff? If the promotion pulls purchases from Target's website into stores, does that count as a win? - Is the goal purely top-line, or must the promotion stay above a margin floor? - Does the promotion run in every US store or only in particular regions or store formats, and over what time window and budget? - Can we target offers to identified shoppers through Target's app and loyalty program, or must the promotion be open to everyone who walks in? ### Part 1 — Design the promotion Describe the promotion: its mechanics, who it targets, how shoppers hear about it, and why it should raise total store revenue rather than mainly discount purchases shoppers would have made anyway. ```hint Break revenue into its drivers Store revenue is the number of shopping trips times what each trip spends. Decide which of those your promotion is meant to move, and let that choice shape the offer. ``` #### What This Part Should Cover - Mechanics tied to a specific revenue driver: more trips, bigger baskets, or both. - Targeting and channels that bring shoppers into stores rather than to the website. - The cost of discounting sales that would have happened without the promotion. ### Part 2 — Choose what to promote Explain how you would decide which products or categories go on promotion, and what data you would use to make that decision. ```hint Look beyond the promoted item The best candidate is not necessarily the item whose own sales jump the most on discount. Consider what else ends up in the cart, and what happens to similar products and to online sales. ``` #### What This Part Should Cover - Selection criteria such as price sensitivity, pull on store traffic, basket attachment, margin and stock availability. - The data and analysis behind the ranking of candidates. - Tailoring by season and local store demand, and the risk of cannibalization. ### Part 3 — Run the experiment Design the experiment that tells you whether the promotion raised in-store revenue and whether to roll it out more widely. ```hint Who can be randomized? Every shopper who walks into a store sees its signs and displays. Choose the unit you randomize with that in mind. ``` #### What This Part Should Cover - The unit of randomization, the control group and how spillover is handled. - The primary metric and its guardrails. - Test duration, statistical power, effects after the promotion ends, and the decision rule. ### What a Strong Answer Covers - A promotion built around a stated revenue driver for physical stores, not a generic discount. - Product selection judged by the effect on total store sales, not by the promoted item's own sales. - An experiment design that fits a store-based promotion and measures incremental revenue. - Guardrails for margin, stock-outs and online cannibalization held alongside the top-line goal. ### Follow-up Questions - Test stores beat control stores during the promotion, then trail them for the two weeks after it ends. How do you judge the result? - Store revenue rises in the test areas, but online orders in the same areas fall by almost the same amount. Was the promotion a success? - You are allowed to run the test in only a small number of stores. How would you change the design?