Brick-And-Mortars Are Taking A Data-Driven Approach To The E-Commerce Challenge
The digital transformation brought a sea change to the marketing world. Data around campaigns from email to pay-per-click to website configurations made everything measurable, and by extension, testable. Most leading marketers understand the value in running A/B tests. However, for maximum business outcomes, that same mindset should also be applied to more traditional real-world marketing, such as marketing trials like product launches and promotions, and in-store experiences like shelf layouts and signage.
Recently, Gartner touted the value of fostering a “test-and-learn” culture rather than relying on opinion or guesswork when addressing the customer experience (CX) around products and services. Its 2017 CMO Survey found that marketing leaders consider CX, customer retention, and growth key to supporting marketing strategies.
Brick-and-mortar stores have contended with competition from the likes of Amazon and the steady growth of e-commerce, where testing is easily done. Yet brick-and-mortars can also take a data-driven approach to the e-commerce challenge. In-store experimentation based on advanced data science allows them to test everything from the store CX to its operations with relative ease and in a scalable way.
To foster a testing culture, Gartner offers several suggestions: begin with internal alignment from the top down; build a dedicated team that is skilled in the testing process; establish a process that begins and ends with customer data and is aligned with business goals; and share the testing plan and results across the organization.
Trial Run, a data science-based test management product that automates experimentation lifecycles, took a close look at how brands apply testing and optimization to real-world marketing challenges like product merchandising.
Case in Point
For our deep dive, we analyzed an in-store testing cycle by a leading U.S. manufacturer seeking to promote an organic line of sunscreen and moisturizer. In doing so, the manufacturer decided to move one of its organic line products from a lower shelf to a top one in stores.
Before making the change, the manufacturer partnered with a retailer in a limited test to understand how it might affect its organic category sales. It wanted to learn if moving the product would cannibalize sales of its own non-organic category as well as draw from its competitor’s organic line.
The retailer ran the test in approximately 150 stores. Advanced data science algorithm used underlying data to create virtual control stores similar to the test stores.
The results found a 20 percent increase in moisturizer sales and a 12 percent increase in sunscreen sales. Seventy-one percent of stores saw a positive response, while 29 percent had a negative outcome. Drilling more deeply into the data, stores in higher-income areas had a more positive response than those in lower-income areas (93% vs. 57%) , and sales in freestanding stores greatly outperformed those stores in strip malls (87% to 48% positive response rate).
The test found a significant impact for the organic sun care product, although it was inconclusive for cannibalization of the brand’s non-organic line. Since incremental impact for organic category was significant but cannibalization was inconclusive due to lower confidence, a selective rollout was suggested. After a selective expansion of the new merchandising strategy, a larger-scale rollout based on the test resulted in a $300,000 sales gain over the course of a few months.
This is a very specific example of in-store testing in action, but it illustrates how a methodical, science-based approach supported by experimentation (rather than opinion or gut instinct) gave the manufacturer and its retail partner empirical evidence that the new merchandising strategy would succeed.
Real-world, science-based testing isn’t limited to product merchandising. It can be applied across a wide range of brick-and-mortar challenges, new product launches, store remodels, loyalty programs and more. A test-and-learn culture like the one described here can take a company’s research capability to the next level, helping to avoid failed ideas, fuel faster new product rollouts, maximize marketing ROI, and ultimately driving better business results.
Jerry Roche is CEO of Trial Run.