Analyze This

Most retailers have the capacity to capture an enormous amount of data about their customers’ shopping habits. The problem: what do you do with all that data?

Programs that analyze shopping patterns and predict the types of promotions likely to increase purchase size or induce customers to buy specific items certainly exist, but are typically expensive, cumbersome to operate and slow to get data into the hands of retailers.

Quantivo Retail gives retailers a “query tool” to install on their back-end systems. It collects all of the data from the retailers’ POS equipment and integrates that data with customer demographic information kept in the retailers’ files. That information is then shipped to San Diego-based Quantivo which, within five days, returns analysis and predictions to help the retailer promote and sell additional products.

“With the current analytical solutions in the market, retailers have to make a huge investment in the hardware and software and it takes a Ph.D. to write the programs,” says Quantivo CEO Brian Kelly. “We can get practical and useful information into the hands of the marketing managers” more quickly and less expensively.

Quantivo provides retailers with “canned” questions to which most retailers want answers, and marketing managers can write additional questions based on their specific needs. Some typically asked questions: What lift has a specific promotion had on the intended product sales? What other products saw increased sales during that promotion? What products are typically sold together?

“Retailers want to know if they put a certain product on sale, what other products will also see a lift along with it,” Kelly says.

Purchase data can be updated on a weekly, monthly or quarterly basis. Questions can ask about comparison sales at various stores to determine where certain promotions are most effective; they can also compare the results of promotions at various times of the day, as well as compare sales results by gender, age or region.

The main features of Quantivo Retail include:
• Market basket analysis of overall product combination purchases
• Customer purchase analysis that identifies product purchases of repeat customers
• Segmentation analysis
• Affinity 360 finder, which performs an exhaustive search for all relevant affinities, often uncovering new patterns

Added flexibility
A large chain of home repair stores conducted a nine-month test of Quantivo Retail that involved the analysis of one billion pieces of information: It was able to increase analytic capacity relative to previously-used analytical tools, as well as gain more flexibility.

“They were able to allow junior analysts to ask questions, and they were able to dig down in understanding even more subtle changes in consumer behavior,” says Albert Gouyet, vice president of marketing for Quantivo. “Prior systems did not allow them to slice and dice the information so much over long periods of time.”

An electronics store with two billion pieces of data to be analyzed was able to find out what customers typically purchased just before they made big-ticket purchases, Gouyet says. It was also able to determine what loss leaders resulted in the biggest number of big-ticket purchases. On a smaller scale, a local chain of home and garden stores – which could not previously justify the cost and personnel resources – was able to analyze purchase data for the first time.

The exact cost for Quantivo Retail varies based on the data volume that has to be analyzed, but the program starts at about $5,000 a month for a smaller retailer, Gouyet says.