MarketingProfs B2B Forum is going virtual... with a twist. Don’t miss it.

As retailers strive to touch consumers at every step of the purchase cycle, retail marketing has evolved to become a mosaic of mass-media branding, tactics for driving store traffic, in-store experience, and loyalty programs.

Brand-equity TV campaigns are overlaid with local newspaper, radio, and direct mail to get consumers into the store. In-store merchandising programs enhance the customer experience at the point of sale and make the register ring. Frequent-shopper cards and branded credit cards create long-term preference and repeat visits.

And for those in considered purchase categories, the mosaic becomes even more intricate as digital marketing and ecommerce are woven into the picture. With all of these marketing tactics working on the consumer in different ways, understanding and improving marketing effectiveness is a particularly vexing challenge.

Like most other marketers, retailers tend to manage marketing effectiveness from within functional silos. Marketing efforts tend to be created, evaluated, and justified by functional experts in brand, DM, merchandising, and loyalty groups, each fighting to justify its efforts and build its budgets. As a result, management often has limited visibility to how much each effort is truly contributing to overall sales. To improve this visibility, many retailers are beginning to use marketing mix modeling.

As typically employed, mix modeling is a great tool for understanding the contribution of marketing spend at a macro level. It can reveal the overall contribution to sales of the marketing budget as a whole, and provide insight into the relative performance of the big buckets of tactical spend. In essence, it helps management answer critical big-picture questions about marketing effectiveness: how much to spend overall, what's working and what is not, and how to best allocate spend across tactics.

But while there is great benefit to be gained from such high-level accountability, continuous improvement of marketing effectiveness in a retail environment must account for the wildly different dynamics that occur at a store-by-store level.

The fact is, retail is the most local of businesses. Each store exists in a trading area with its own consumer characteristics and competitive set. Each store represents a unique set of merchandising characteristics: store size and format, floor layout, merchandise array, discounting, and signage, among other factors. And each store has local management and all that implies in terms of service experience, cleanliness, and overall quality of execution.

All of these factors are important influencers on sales, and all of them contribute to how marketing programs perform on a local basis. Ideally, retailers would like to understand, track, and quantify all of these factors as drivers of sales on a store by store basis, so that all can be managed for continuous ROI improvement. This ideal can be accomplished with a technique called store-level modeling.

Understanding Store-Level Dynamics of Marketing and Business Drivers

While modeling the impact of marketing investments at the national and regional level is becoming more common, modeling at the store level is a more recent development, made possible by innovative econometric modeling techniques, improved technology, and the availability of more in-depth data.

All of this allows national retailers to understand the dynamics of their business at a much more granular, dynamic, store-by-store level. By taking advantage of store-level modeling, industry leaders have been able to answer the following questions:

  • Does a store's location influence marketing response? Do mall stores perform differently from strip mall stores or freestanding stores?

  • How much do incremental staffing and staff incentives contribute to sales?

  • What is the impact of remodeling individual stores and what is the payback period?

  • What is the impact of non-marketing factors, such as gas prices, weather, and competitive store openings within a store's trading area?

When armed with the recognition that the consumer response to marketing differs by store, retailers are discovering that they can improve their return on marketing investment (ROMI ) by varying their mix of marketing tactics to fit a store's profile. That is, one type of store may be more responsive to TV advertising, another more to direct mail, and another to in-store promotions. For that reason, it is important that retailers segment their stores by type, such as mall, freestanding, and strip mall.

A major national retailer recently used store-level modeling to understand the impact of remodeling its stores. Through this effort, it learned that the payback period from remodeling standalone, strip mall, and mall stores varied significantly, allowing the retailer to manage its capital investments more profitably. This retailer also learned that different types of stores react differently to marketing and environmental factors:

  • Their marketing program had less impact on standalone stores than on mall or strip stores overall.

  • Holiday TV was more effective in mall stores, while targeted flyers had a much greater sales impact on strip-mall and standalone stores.

  • Consumers that shopped at mall stores were more sensitive to gas prices than those shopping at strip or standalone stores.

Another Fortune 500 retailer was able to improve its Hispanic marketing efforts by modeling individual stores. The retailer had categorized as "Hispanic" those stores that were located in communities with a majority Hispanic population. The modeling confirmed that Hispanic-designated stores were less likely to move sales by traditional mass marketing. It also indicated the importance of targeted Hispanic marketing in general and in particular that monthly flyers targeted at Hispanics were particularly effective. In addition, the modeling revealed that advertising on Hispanic TV had a significant impact on stores in areas that had large, but not majority, Hispanic populations. This led to an increase in overall spending on Hispanic marketing and a change in targeting strategy.

Another global company has used the same approach to understand how staffing changes over time, employee incentive programs, and employee training influence their business results. And on the cutting edge, another national retailer is beginning to use the store-level approach to get a better understanding of how to optimize its highly integrated digital marketing, ecommerce, and in-store experience efforts.

Pioneering retailers are using store-level models as tools to provide deeper, more actionable insights into the marketing and non-marketing drivers of the business. To be sure, developing models at the store level requires more effort than the typical marketing mix approach. More granular data is required, and the modeling is more complex. However, if the goal is more impact over more aspects of the business, there is no better approach.

Putting It All Together

Here is a checklist for retailers wanting to improve their ROI, five key ways towards a rosier bottom line:

  • Get local. Don't just rely on regional and national marketing plans. Segment your retail stores by type of store (i.e., mall, freestanding, strip mall) and analyze how each store type responds to different marketing tactics. Act on the information.

  • Knock down silos. Establish open communication and common metrics among functional departments, including brand, merchandising, and direct marketing, among others.

  • Strike a balance between brand building and promotion. Don't dilute brand equity by over-promoting.

  • Don't overlook non-marketing factors. Gas prices, weather, and competitive store openings all impact sales. Factor them into any ROI calculation.

  • Take advantage of synergies and interactions. Understand how your different marketing tactics work together (e.g., how TV impacts digital marketing).

Sign up for free to read the full article.

Take the first step (it's free).

Already a registered user? Sign in now.

Loading...

ABOUT THE AUTHOR

Rishi Bhandari is vice-president of analytical solutions and senior scientist at Marketing Management Analytics (www.mma.com). Reach him at Rishi.Bhandari@mma.com.