Coldwater Creek Grows With Location IntelligenceColdwater Creek Grows With Location Intelligence

Women's clothier relies on geospatial analysis to track customers and avoid overlaps as it adds more than 60 retail locations per year.

Doug Henschen, Executive Editor, Enterprise Apps

March 17, 2008

3 Min Read
information logo in a gray background | information

When you don't have many retail locations, it's easy to find places to open new stores, but the task gets harder as the retail network grows. That's when location intelligence (LI) technology goes from being extremely helpful to being absolutely essential, and it's exactly the point when retailer Coldwater Creek took a deeper dive into spatial analysis.

It was six years ago when the successful catalog direct marketer surpassed 40 stores. With the emphasis on retailing growing, Coldwater Creek bought MapInfo software (now Pitney Bowes MapInfo) and began the task of spatial-enabling the company's 14-terabyte customer data warehouse.

"The data integration starts at the transaction level, so if Suzy Smith bought a red blouse, you need to aggregate that into the relevant geographic data structures, such as census block groups," explains Carl Brenner, manager of statistics of geography.

It's all relatively straightforward, says Brenner, but you have to build aggregates, custom data feeds and structures, and you need to integrate that information with the spatial tools. You also have to handle all the details such as slowly changing dimensions. In the case of MapInfo, SpatialWare is the data management tool, and it also lets you do spatial analysis through SQL, asking questions such as "which customers are within a particular store's trade area?"

Defining trade areas is part of the art of LI that only comes through experience, says Brenner. "Every company's business strategy is a little bit different. For Starbucks, it may be okay for stores to be across the street from each other because the cost of crossing the street is greater than the cost of the cup of coffee. For Coldwater Creek, shopping is more of a social activity. You are going with your friends or your family, and you may be willing to go long distances. Our trade areas are a lot more complicated and a lot more fluid."

Coldwater Creek surpassed 300 stores in 2007, and the company is adding 65 new stores each year. It takes a day of analysis in MapInfo "just to kick the tires" on a prospective new location, says Brenner, examining existing customers, local demographics, road networks, barriers such as bodies of water and the potential impact on existing store sales. It may be three or four days before the analysis is complete, but between Coldwater Creek's experience and improvements in software and hardware in recent years, that's down considerably from the two weeks it used to take just for the data management steps.

Of course, the analyses are bit more involved now that there are that many more stores. "For example, we opened a store in Manhattan last year, so we had to ask ourselves, 'do we market to people in New Jersey, and if so, how far do we go?'" says Brenner. "You don’t know for sure what impact a new store will have until it opens, but there are plenty of hints out there, such as commuting patterns and road networks. If the customer is close to the George Washington Bridge, chances are they are going into New York City."

Coldwater Creek starts planning new locations as much as two years in advance. The goal, says Brenner, is neither over-saturating the network and hurting existing stores sales nor under-exploiting the potential to reach new customers. It helps that Coldwater Creek can draw on a rich data warehouse to better understand customer behavior."

"Most companies will add demographics into a tool like MapInfo and call it good," he explains. "We have the luxury of knowing who lives where and what they're buying. It's an added wrinkle — and a complicated one — but it makes the analysis that much more accurate."

Read more about:

20082008

About the Author

Doug Henschen

Executive Editor, Enterprise Apps

Doug Henschen is Executive Editor of information, where he covers the intersection of enterprise applications with information management, business intelligence, big data and analytics. He previously served as editor in chief of Intelligent Enterprise, editor in chief of Transform Magazine, and Executive Editor at DM News. He has covered IT and data-driven marketing for more than 15 years.

Never Miss a Beat: Get a snapshot of the issues affecting the IT industry straight to your inbox.

You May Also Like


More Insights