Shopper Insights for CPG Brands

Consumer insights are about the person who eats the product. Shopper insights are about the trip on which somebody buys it. The same human is both, and they behave differently enough that treating them as one is the most common reason a shopper marketing program underperforms.

A parent who prefers your brand and buys the competitor because it was on the endcap is not a consumer problem. It is a shopper problem, and the fix is in the store rather than in the advertising.

Key takeaways

• The unit of analysis is the trip. Trip type predicts behavior better than any demographic cut.
• Grocery shopper segmentation should be behavioral. Demographic segments in CPG rarely behave differently enough to act on.
• Most category decisions happen in seconds at the shelf, which is what makes pack, position, and price do so much work.
• Shopper insights are the currency of a buyer conversation, because they are about the retailer's shopper rather than your brand.

Trip type is the useful segmentation

The same household shops differently depending on what the trip is for, and the category's behavior changes with it.

The practical consequence: an item that skews to quick trips is badly served by a price promotion, because the shopper is not comparing. An item that skews to stock-up is badly served by a display in a high-traffic aisle, because the shopper is working a list. Knowing which one you are is worth more than knowing your buyers' median age.

Grocery shopper segmentation that holds up

Demographic segmentation is popular because the data is easy and because it produces a slide. It usually fails the actionability test: two demographic segments that buy the same items at the same rate do not need different plans.

Segment on behavior instead. Four cuts do most of the work: purchase frequency in the category, deal sensitivity, brand repertoire size, and channel mix. A shopper who buys the category weekly, never on deal, from a repertoire of two brands is a completely different target from one who buys monthly, only on deal, from a repertoire of six, and those two are frequently the same age with the same household income.

One caution: segments have to be reachable. A behavioral segment you cannot target through any retailer program, retail media platform, or shelf decision is an interesting finding with no delivery mechanism attached.

Cross-shopping is the shopper behavior a sales report hides

Two kinds of cross-shopping matter here and they get confused constantly. Brand cross-shopping is one household buying you and buying your competitors inside the same period, which most category buyers do; a repertoire of four to six brands is ordinary in a fragmented set, and a repertoire of one is rare enough to be worth investigating when you find it. Banner cross-shopping is the same household buying the category at more than one retailer, and that is the version a buyer cares about, because it means their shopper is finishing the trip somewhere else.

Neither is visible in a sales report, which is the honest limit on this whole subject. Store-level POS records transactions and not people, so it cannot tell you that one household bought your 16 oz at one banner and a competitor at another. Cross-shopping reads come from a household panel or from a retailer's own loyalty program, and the useful ones are usually the retailer's, because they can see their shopper and you cannot.

What POS does support is the inference that earns the question. If your item runs 1.8 units per store per week at one banner against 3.1 at a comparable banner on the same price and similar distribution, something about that store's set or that store's shopper is different, and cross-shopping is one of a short list of candidates. Bring the velocity gap and ask the buyer for the shopper behavior behind it. That conversation goes better than presenting a panel number about their shopper that they can dispute with data you cannot see.

The four questions worth answering

• Who is buying the category at this retailer, and how does that differ from the market? This is the question buyers care about most and brands answer least.
• What else is in the basket when we are in it? Drives adjacency arguments, cross-promotion, and secondary placement.
• Where do we lose the trip? Out of stock, wrong pack, position, or price. Each has a different fix and they are frequently confused.

The third question deserves its own method, because the four causes look identical in a sales report and cost very different amounts to fix. Out of stock shows up as zero-unit weeks in stores with prior sales. Wrong pack shows up as healthy category velocity with weak share in one size. Position shows up as velocity varying widely across stores with similar demographics and similar distribution. Price shows up as velocity tracking the price index across the set. Run those four checks in that order, cheapest first, before commissioning any shopper research to answer a question the POS data already answers.

• What does a new buyer look like in their first three purchases? The best early signal of whether a launch will hold.

Using shopper insights in a buyer meeting

This is where shopper insights pay for themselves. A buyer is largely indifferent to your brand's performance and highly interested in their category and their shopper. An insight framed the second way gets a different meeting.

The pattern that works: a fact about their shopper, a gap it implies in their set, and a specific recommendation with a number attached. "Your fermented set over-indexes on stock-up trips, and you carry no item under $5 for the fill-in trip that is 41 percent of category occasions. A 12 oz at $4.99 addresses it, and in the two banners already carrying one it does 2.1 units per store per week without cannibalizing the 16 oz." That is a shopper insight. "Consumers care about gut health" is not.

For the discipline this sits inside, see shopper marketing and category assessment.

How Scout fits

Scout supplies the store-level half of the shopper picture: what sold in which stores, at what price, in which weeks, against the rest of the set. That is what turns a panel-based shopper hypothesis into a claim a buyer will accept, because it is measured in their stores rather than projected from a national sample.