| Title: | Prove It: An enterprise-wide control group can help measure a database’s value |
| URL: | http://www.rimmkaufman.com/rkgblog/2006/10/15/prove-it-an-enterprise-wide-control-group-can-help-measure-a-databases-value/ |
| Printed: | January 5, 2009 |
| Source: | The Rimm-Kaufman Group Blog, info@rimmkaufman.com |
- October 15, 2006
- 1 comment
Great article by Richard Tooker in October’s Direct titled Prove It: An enterprise-wide control group can help measure a database’s value.
Almost all marketers employ control cells to measure test lift. Tooker advocates establishing an additional holdout control panel of 50k customers who receive no marketing communication — ever. This panel allows a company to determine the aggregate lift of their customer communication efforts. The panel answers the question, “What we getting from all of our housefile marketing investment?”
Worth reading and considering.
If you like this post, consider subscribing to our RSS feed. You can also have new posts sent to you via email.
Possibly Similar Posts
- Customer Type Matchback Analysis & Lifetime Value
- Banners v. Search
- Video: Customer Centric Paid Search Marketing
- Yegge: If It’s Something You Want, Then You Already Know What The Requirements Are
- “List, Offer, Package” in the Age of Online Marketing And Site Usability
Back in 1993-1994, we ran a year-long test at Lands’ End. There were five business units that contacted customers via seven mailing strategies. We employed a 2^7 factorial design. One of the 128 combinations received no mail for the whole year.
At the end of the year, we had 128 different mailing strategies to evaluate. We could evaluate all possible combinations of strategies, down to profit. It was the most fun I had in my career.
Long-term holdout groups provide results that are different than what you see when executing short-term holdout groups. I wholeheartedly encourage long-term testing.