Shopper-marketing stakeholders talk a lot about the need to “quantify the path-to-purchase” — not just to measure the effectiveness of programs and tactics, but also to evaluate them against traditional advertising investments.

Sales response models might make a nice playing field on which to compare retail marketing to traditional advertising. While shopper-marketing advocates frequently measure program effectiveness through direct sales effects, the traditional advertising world has shunned these models, arguing that they only capture short-term effects and miss the long-term benefits of brand building.

Marketing actions, they argue, move consumers closer to the buying decision gradually, and therefore must be assessed by probing their hearts and minds, as measured by upticks in awareness and intended purchase.

Not satisfied with this reasoning and determined to find a a fair way to compare the effectiveness of marketing vehicles quantitatively, we began scouring academic research. One gem, “Mindset Metrics in Market Response Models: An Integrative Approach,” appeared in the August 2010 issue of Journal of Marketing Research.

The report’s three authors, professors at highly regarded business schools, analyzed 60 brands in four categories. They used seven years of French purchase data, along with brand and competitive marketing mix data (pricing, in-store promotion, advertising support data) and mindset variables (trackers measuring ad awareness, brand liking, and inclusion in the consideration set).

Their main finding is that adding mindset metric variables to their model did, in fact, improve the model’s ability to explain long-term brand sales variation, with 53 percent of a brand’s sales variation being explained by all factors. However, a brand’s marketing-mix actions explained far more of the variation (23 percent) than did its mindset metrics (8 percent).

Perhaps more interesting are their findings on cumulative (short- and long-term) sales elasticities — a concept that had us reaching back to Economics 101. Sales elasticities represent the response in sales volume driven by a substantive change, or “shock,” to specific variables. The French data indicates a cumulative sales-to-in-store-promotion elasticity of .277 and a cumulative sales-to-advertising elasticity of only .036.

In other words, a 10 percent increase in promotion yields a sales increase of 2.77 percent, versus a 10 percent increase in ad spend, which yields only .36 percent!

I was boastfully sharing these findings with colleagues, when someone pointed out that while the advertising support was measured by spend, promotions were measured by the percentage of the distribution on promotion (in-store communication, presence of in-store flyers, price promotions, and bonus buys) during a given period.

A fairer contest would compare the .036 sales-to-advertising elasticity to a sales-to-promotion elasticity figure calculated from a shock to promotional spend. We decided that we would begin constructing our own database of sales-to-promotions elasticities across many programs and tactics, using real campaign costs and store sales data adjusted for seasonality, and compared to control stores.

Obviously, it will take us time to analyze enough campaigns with our marketing-mix models to produce an estimate in which we have confidence. Our SmarTeam analytics group, working with Jeff Dotson, Assistant Professor of Marketing at Vanderbilt University, has so far analyzed a handful of sampling programs in the grocery channel. The figures, now with investment accounted for, corroborate the French promotional elasticity figure of .277.

If we ignore the cookie program (a potential outlier since samping is particularly effective for line extensions) — and even round down to .2 — our findings indicate that an investment in sampling at retail drives a superior cumulative sales response than one in traditional advertising. The degree of its superior performance depends on whether you use the French ad elasticity of .036 (more than five times stronger) or the more forgiving empirical generalization of sales-to-advertising elasticity of .1 (twice as strong).

But what about the “mindset” metrics the traditional advertising world emphasizes in lieu of sales? After all, the seven years of French data used in the academic study suggest that improvements in mindset variables deliver a nice sales impact (see the right side of the chart above). Are these variables and their sales effects the direct byproduct of traditional advertising investments?

Not necessarily, according to the data. All three mindset metrics were most influenced by distribution. Ad awareness was only slightly more responsive to ad support (mindset-to-marketing-mix elasticity of .05) than one to promotion (.06). What’s more, brand liking and consideration — which both have a stronger sales impact than ad awareness — are actually more responsive to promotion than to ad support.

Some may find it hard to believe that ad awareness could be nearly as responsive to promotion as ad spending. However, shopper-marketing enthusiasts are not among them. That’s because promotions can reinforce traditional advertising communications, triggering memory links at the moment-of-truth.

The vehicles work in tandem and it should not really be a question of one being “better” than the other. That said, we love how these findings lend further credibility to shopper marketing, which often sits in its big sister’s shadow. The bottom line is that with marketing mix and mindset metrics all considered, promotions at retail appear to have a stronger impact on short- and long-term sales than advertising support.

If brand liking and consideration are more responsive to the presence of in-store promotions than ad spend — and ad awareness is practically a draw — shopper marketing’s role in winning shoppers’ hearts and minds likely goes beyond simply reminding them of traditional ad campaigns.

Let’s again look at in-store sampling, the focus of our initial modeling exercise. Sampling’s impact on near- and long-term sales is most obviously linked to the fact that human interaction and a mid-store setup are highly disruptive and engaging — taking shoppers off “autopilot.” A retail event for a well-liked brand can go even further and may positively impact brand liking and consideration. An attention-grabbing event station and sign can have such an impact, as well.

Second, the invitation to try an item before buying it eradicates the initial purchase barrier, ultimately leading to repeat purchases from those with positive experiences.

However, the product-trial process can also influence mindset metrics such as brand liking, as a brand-evaluating experience happens right there on the spot. Similarly, importing brand equities from national advertising campaigns bolsters their impact on ad awareness.

Sampling works among repeat product users and older products as well, so there’s even more to it. Supplementing the product experience with creative ideas to derive more uses from the product gives the shopper more reasons to purchase or repurchase. This is especially true today when, according to Deloitte’s 2010 American Pantry Study, many shoppers have adopted new saving norms and are seeking ways to maximize their pantries.

Ideas and solutions may also drive mindset metrics such as consideration set. If a brand has supplied me with a new way to prepare meatballs, won’t I be more likely to consider the brand when shopping its category? I may like the brand more to boot, since it inspired me with a new idea. Distributing the meatball recipe with a recipe card or digital tactic makes the effect more enduring; when I need that solution the most, I’ll know where to find it and be reminded of the brand that brought it to me.

The vast amount of syndicated data today makes it possible to identify which products and solutions are most relevant to which shoppers — increasing their impact on sales and mind metrics alike.

And so it goes that the concept of elasticity, taught in first-year economics classes, provides a common metric with which to compare the effectiveness of retail tactics to traditional advertising. With this common metric, and robust academic research and modeling, we have quantitative evidence supporting what we already suspected from statistics about decisions made in-store: promotions have a stronger impact on near and cumulative sales than traditional advertising.

Standing on the shoulders of these research giants and their recent paper, our own supplementary analysis incorporating costs further suggests that sampling constitutes a more efficient spend than traditional methods. These findings should help us maintain shopper marketing’s share of marketing budgets, but also inspire us to analyze some of the most tried-and-true tactics like sampling and identify ways to make them operate on all cylinders — unleashing their power to drive consumer mindset and sales alike.