WASHINGTON—You might think that the best place to put a Redbox DVD rental kiosk would be at a retail store in a highly populated area with a lot of foot traffic. And that’s a fair assumption. The trickier question is: where do you put the next kiosk?
The answer to that question is not necessarily intuitive, as Redbox representatives explained during their presentation at the sixth Annual Tableau Customer Conference, on Sept. 9 in Washington, D.C. Matt James, senior director of strategy and analytics at Redbox, and Andrew Thompson, an analyst of strategy and analytics, described how the company uses Tableau Software’s data visualization tool to figure out the best places to put new kiosks.
Originally developed by the venture arm of fast-food chain McDonald’s in the early 2000s, Redbox today has 44,000 kiosks across the United States. Early this summer, it passed the three-billion mark in number of DVD rentals. The company now claims more than 50 percent of the U.S. DVD rental market, according to James.
After such explosive growth, Redbox is now concentrating on finding the optimal locations for additional kiosks, says Thompson. The right location is critical to growing revenue because each new kiosk could potentially cannibalize too many sales from existing kiosks. The company tracks every transaction (DVDs typically rent for $1.29 to $2) at each kiosk, including number of disks rented down to each individual movie. It also tracks the movement of each disk between kiosks.
“Today, about 50 percent of our disks get returned at a different kiosk than they were rented from,” says James. In some cases, people rent the DVD in Washington, D.C., and return it in Chicago, for example – presumably after watching the disk during a flight. But most of the paths the DVDs travel are more local.
The information on kiosk location, sales figures and the movement of inventory gives Redbox a good sense of how these factors impact the return on investment for each kiosk. Now Redbox is analyzing data to assess the relationship between kiosks. The company is examining the “trade area” of each kiosk in several different ways. The first way is by distance, says James. For example, you can assume that customers within a two-mile radius of a given kiosk are likely to rent from that location. But another way to look at it is by drive time: examine the transportation routes and figure that kiosk may have customers within a five-minute drive of the location.
But by tracking the path of each DVD, the company has identified other types of relationships between kiosk locations, adding another level of data that’s showing some surprising results. In one case, the data visualization can highlight an unexpected relationship between kiosks miles apart. In one such instance, it showed a strong correlation between a kiosk just north of Oceanside, Calif., and a location just east of downtown Los Angeles in Whittier, Calif., over an hour away.
“People were driving by hundreds of other Redboxes where they could have returned that movie,” says Thompson. “We asked, ‘what’s going on here?’” By studying the areas, the analysts realized that military personnel were renting movies at Camp Pendleton, a Marine base in Oceanside, and returning them at their training location in Whittier.
“This is the type of thing we’d never see or understand unless we looked at it in Tableau,” says Thompson. And it’s the type of data the company hopes to leverage to attain the best ROI possible from new Redbox locations.
Tam Harbert is a freelance writer based in Washington, D.C. She can be contacted through her website.