BACKGROUND: Vermont Country Store is a family-owned catalog company in Manchester Center, Vermont, that started as a retail store in 1952. Today, the company uses data mining to target-mail more than 40 million catalogs a year.
PLATFORMS: Enterprise Miner software runs on a Windows NT client and server.
PROBLEM SOLVED: Enterprise Miner software is used as part of Vermont Country Store's targeted marketing strategy to turn large amounts of customer data into knowledge to improve the targeting of a variety of catalogs.
PRODUCT FUNCTIONALITY: SAS Enterprise Miner gives us the ability to investigate literally hundreds of variables. Currently, we are using house-file internal variables and purchased data. In the future, we will likely use research data from our customers as well. Enterprise Miner is giving us the ability to incorporate a lot more into the scores that go into target mailings. It lets us address many different business objectives. In the future, we look forward to planning further out and identifying optimal campaigns for individuals in terms of what titles we are sending out and when.
STRENGTHS: Enterprise Miner delivers a point-and-click GUI with all the power of traditional programming. Competing packages have a GUI, but not the power, control and flexibility. Enterprise Miner's strength is in allowing lots of control over what's going on, and it lets you develop and interact more as you learn more about your data. One feature I love is the project organization and documentation. This lets me put it all code nodes and comparing scores to different measures on the customer base in one spot. That way, others can come back and build on what I've been doing. It's great that if someone leaves a programming area, it's easy for someone else to pick up the history.
WEAKNESSES: Temp files take up a lot of space on the hard drive, although a lot of them probably don't need to stay there. Some of the charts can be difficult to navigate to and organizing the libraries is slower than I'd like. Printing out a decision tree can be challenging, too.
SELECTION CRITERIA: SAS Institute's reputation and its training and technical support were major factors. SAS Institute is probably the leader in data mining. I've been using SAS software for six and a half years. For me, it's great to be able to use SAS software programming within Enterprise Miner. I love that feature.
DELIVERABLES: I like that the output is similar to what I'm accustomed to. I also like being able to do my own checks on models. Enterprise Miner generates complete scoring code for all stages of model development including variable reduction and transformations. Scoring code can be applied immediately to new data, which saves time. An assessment facility converts data mining results into business terms that can be shared via the Web. The data mining method that SAS Institute calls SEMMA (sample, explore, modify, model and assess) is fully implemented in Enterprise Miner.
VENDOR SUPPORT: We have had very good support, although we would have liked to spend more of the consulting time working with the product. I've always gotten a response within an hour from the technical support desk, so I've been especially pleased with the technical and sales support.
DOCUMENTATION: There's a lot of good, basic online help. I'd like to see more examples, such as setting up profit vectors. I'd like a tutorial, something on how to build a model with Enterprise Miner step by step. There's so much functionality in SAS Enterprise Miner, I want to make sure I'm not missing out on something.
Register or login for access to this item and much more
All Information Management content is archived after seven days.
Community members receive:
- All recent and archived articles
- Conference offers and updates
- A full menu of enewsletter options
- Web seminars, white papers, ebooks
Already have an account? Log In
Don't have an account? Register for Free Unlimited Access