How do retailers test assortment, pricing and promotions with shoppers?
TL;DR
- Retail research is about baskets: what gets added, what gets swapped, and where shoppers would go instead.
- Read online and in-store shoppers separately. Online is a large and growing share of retail, and the two groups shop differently.
- Test promotion depth on a ladder, not one discount level, so you find where response flattens.
- Neon Blue returns basket, switching and promotion response by shopper segment, labeled as simulated. Run your first study free
Neon Blue builds simulations of human behavior. Retailers test changes by showing shoppers the shelf, page or offer they would actually see, and reading what goes in the basket, what gets swapped, and whether they would shop somewhere else. The questions are practical: which items to carry, how deep a promotion has to be to move people, whether shoppers will try the store brand, and which loyalty perk brings them back. Neon Blue runs these reads on a simulated audience of your shoppers so a change can be screened before it hits every store and the site.
Why should retailers read online and in-store shoppers separately?
The U.S. Census Bureau's quarterly e-commerce report put retail e-commerce at 17.1% of total retail sales for April through June 2026, seasonally adjusted, with e-commerce sales up 12.2% from a year earlier against 6.7% for total retail. Online is a big and fast-growing slice of retail, and it doesn't behave like the store.
Online shoppers compare prices in a tab, read reviews and see endless assortment. In-store shoppers face a fixed shelf, see end caps and buy on impulse. A promotion or assortment change can work in one channel and fail in the other, so every retail study should read the two groups apart.
Which retail decisions can you test with shoppers?
| Decision | What to read | Shopper cut that matters most |
|---|---|---|
| Assortment change | Would they still find what they need, or shop elsewhere | Shoppers of the item being cut |
| Promotion depth | Response at each discount level | Deal-seekers vs regular-price shoppers |
| Private label launch | Willingness to try, what it replaces | Brand-loyal vs value shoppers |
| Loyalty perk | Which perk changes visit frequency | Members vs non-members |
| Delivery or pickup fee | Keep using, switch to store, go elsewhere | Online-heavy shoppers |
The pricing methods behind promotion and fee tests are in how to measure willingness to pay.
What questions should a retail shopper study ask?
- How often do you shop at [retailer], in store and online? [Never to weekly or more]
- If [item] were no longer available, what would you do? [Buy a different item here, buy it elsewhere, skip it]
- At [discount level], how likely would you be to buy [item]? [Shown at three levels, random order]
- Would you try [retailer]'s own brand of [item] at $[price]? Why or why not?
- Which of these perks would make you shop here more often? [Perk list]
- If a delivery fee of $[amount] were added, what would you do?
Question 2 is the most useful assortment question. "Buy it elsewhere" from shoppers of an item tells you how much traffic that item holds. Question 3 shows where extra discount stops adding buyers.
What does a retail study look like on a simulated audience?
Hypothetical: a home improvement retailer is deciding whether to launch a store-brand line of paint and how deep the launch promotion should be.
- Audience: US adults who bought paint for a home project in the past 12 months, split by DIY homeowners and those who hire contractors, by online versus in-store buyers, and by brand-loyal versus value shoppers.
- Design: private label trial question, what it would replace, then a three-level promotion ladder.
- Readout: willingness to try by segment, brands it would replace, and response at each promotion level.
Example output layout (illustrative, not a real result)
| Segment | Would try store brand | At lowest discount | At middle discount | At deepest discount |
|---|---|---|---|---|
| DIY homeowners | __% | __% | __% | __% |
| Online buyers | __% | __% | __% | __% |
| Brand-loyal shoppers | __% | __% | __% | __% |
If response barely changes between the middle and deepest discount, the deeper promotion gives away margin for little extra trial. Brand-loyal shoppers who won't try at any level are better reached with a message about quality, tested with shoppers before it runs.
What should retail marketers keep in mind?
- Baskets, not items. A discount that pulls in shoppers who then buy only the promoted item can lose money. Ask what else they'd buy.
- Holidays change everything. Read seasonal promotions with seasonal shoppers, not an average audience.
- Pages count too. For product pages and landing pages, see landing page testing.
- Check with real shoppers. For a big change, run a small store test and keep simulated reads separate from live results.
Retailers also run brand trackers; the setup is in brand tracking. Rolling out simulation across merchant and marketing teams is covered in implementing AI in a marketing research team.
Questions about retail shopper research
How do retailers test promotions before running them?
They show shoppers the offer at several discount levels, read response at each, and look at what else shoppers would buy, before a live test in a few stores or on the site.
How do you research whether shoppers will buy a store brand?
Ask brand buyers whether they would try the store brand at a set price, what it would replace, and why they would or wouldn't. Read the answers by value and brand-loyal shoppers.
Should online and in-store shoppers be researched separately?
Yes. They see different prices, assortments and cues, so the same change can land differently. Read them as separate segments in every study.
How does Neon Blue help retail marketing teams?
Neon Blue builds a simulated audience of your shoppers and reads basket, switching and promotion response by segment, labeled as simulated. Start a free first study.