Column: Taking Measure-Modeling for Retailers
Retailers face an ongoing task of balancing efforts often aimed at driving foot traffic, like discounts and coupons, with initiatives designed to promote their brands. Merchandisers tend to favor promotional tactics such as discounts, store-wide sales and newspaper inserts, knowing these strategies deliver short-term returns. But brand marketers know discounts don’t build equity, and suspect that over-reliance on discounts erodes differentiation, trains consumers to buy based on price and reduces margins over time. As a result, balancing short-term sales and long-term brand equity often results in organizational tug of war.
Because retailers live in a world of “same store sales” and week-to-week comparison to year ago, the price-based merchandising view usually wins. Merchandisers pressure the organization to match the previous year’s sales and newspaper-insert support, because the volume response of these tactics is seen as immediate, and relatively predictable; in other words, it carries a low risk. The equity programs advocated by marketers are hard to measure in the short-term, and unpredictable in their long-term impact; that is, they carry a higher risk.
But looking at this challenge as a simple balance of short- and long-term sales response does not account for the dimension of risk associated with each set of tactics. A more holistic approach is to consider the merchandising and equity-driving programs as part of an overall portfolio that balances risk and reward with long-term and short-term response.
A portfolio of retail marketing programs can be strategically managed in the same way a financial portfolio is managed. By using mathematical models that represent risk, potential return and time, portfolio components are chosen to balance these characteristics and create the best probability of meeting the overall strategic objectives.
In order to construct a balanced portfolio, retail analysts need data, or at least assumptions, to estimate short- and long-term response rates, how pricing will affect consumers’ decisions and the risks of each program. Marketing econometric modeling is a powerful tool for providing some of these data points. Models can reliably measure equity programs’ short-term volume, and can also be combined with consumer tracking data to estimate their long-term volume impact.
Unfortunately, typical modeling studies don’t assess long-term impact on price elasticity, or how a change in price will affect demand or risk. Unlike financial models, marketers cannot take advantage of a securities marketplace that sets prices based on a balance of risk and return. Marketers have to make risk assumptions based on history, experience and judgment to factor into the portfolio model. By estimating elasticity impacts, and program-effectiveness risk, marketers learn the stress points in the portfolio.
Often, it becomes clear that even small changes consumers’ response to price can disproportionately impact the expected net present value of a program and turn the risk perceptions upside down. In other words, if consumers grow to expect discounts, the company risks potential long-term price degradation.
Payback typically varies by type of campaign. Short-term responses to branding campaigns on TV, direct campaigns on TV, newspaper inserts and magazine print ads can differ based on a variety of factors. With insight to cross-elasticities between departments, it is simple to reallocate investment on this basis. Retailers who analyze the various response rates, and then redeploy funds, should see an improvement in short-term business results.
Additionally, when retailers factor in the potential long-term benefits of branding campaigns, and the potential of price degradation associated with promotional TV and inserts, they can achieve optimum balance of risk and return for the long- and short-term.
In and of itself, this portfolio approach will not make the issue of merchandising/promo support vs. equity investment disappear, but at least the debate can be grounded by the math and models.
John Nardone is chief client officer for Marketing Management Analytics. (email@example.com)