Trouble is, data from Nielsen Co.'s new shopper-marketing measurement system, Prism (for Pioneering Research for an In-Store Metric) show fewer than 10% of Sunsilk's young women aged 18 to 24 actually go down the ice-cream aisle on a shopping trip. End-aisle and hair-care shelf ads reach far more of them.
Improved research is shedding light on what's long been a black hole of marketing budgets: the $25 billion TNS Sorenson estimates marketers will spend annually by 2010 to advertise or get special displays for products in stores. The biggest and best-funded example of that is Nielsen's Prism, set to expand in July from a pilot into a full national rollout based on data from thousands of stores.
A study by Deloitte Consulting for the Grocery Manufacturers of America last year found shopper marketing, formerly known as trade promotion, is growing faster than any other medium for package-goods marketers, including digital advertising. Deloitte estimated package-goods companies spend 8% of their marketing budgets now on shopper marketing, but that percentage could be well over 35% if all forms of trade promotion are included.
Understanding, however, has lagged behind spending. It's often impossible to separate the impact of what marketers buy, such as in-store TV and radio or shopper circulars, from what they really hope to get for their money: special displays that can pay off big but often aren't explicitly purchased.
In separate research recently on cereal promotions by Nielsen rival TNS Sorenson, some of the more-established mainstays of category promotion, such as in-store sampling and instant-coupon machines (from News Corp.'s SmartSource) resonated well with consumers in surveys. But some applications more like traditional media advertising, including in-store TV and floor ads, fared worse.
The study gave high marks to on-pack advertising because of the billboard-like nature of cereal boxes. But many of the results are similar to what TNS has found in other categories, said James Sorenson, exec VP-retail and shopper insights for TNS Sorenson.
TNS's research works differently than Prism; it is based on consumer surveys. The Nielsen system, based on behavioral data, uses electronic eyes and human auditors to conduct head counts of shoppers moving through individual aisles and store-scanner data to track impact on sales.
The TNS data come from a "marketing contact audit" similar to those used in communications planning by agencies and marketers to determine what types of media appeal to various consumer segments. But while broader audits often lump all shopper marketing into a single bucket, TNS delved into more detail across 19 specific in-store media and tactics for its study, conducted with 552 consumers in February.
The in-store programs found "most influential" by consumers in the study were, in order, in-store samples, package ads, coupon dispensers, in-store fliers and end-aisle displays. Those deemed least influential were ceiling ads, in-store audio and floor ads. Most of the newer media -- including cart ads, in-store TVs, video displays, checkout-lane ads and entryway ads -- ranked in the lower half of the 19 media considered.
Relevance is key to any media working in-store, Mr. Sorenson said, so the more closely aligned the ad is with the category and product, the better.
Logistically, he added, a lot of newer in-store media are out of place because the best place to get a shopper's attention is about 40 inches above the floor, not on the floor or dangling from the ceiling.
Just being influential doesn't necessarily make a medium effective, either. Sampling usually does well in consumer surveys, but it's so expensive that it's hard to deliver a positive return on investment, he said, adding that some media found less influential could in the end be more effective if they're less cluttered by competitive advertising.