I’ll never forget the first time I got a chance to actually look at a client’s house list for the first time. I was shocked. The list was filthy. The list had a 8 ½ % duplicate rate. And one lady was on the list 14 times. You don’t think that’s possible? Well I didn’t want to intrude on a client’s data so I found this address on a real estate website. It’s a 3 bedroom, 2 bath rancher for$128,000 – sounds like a great deal (here's a picture in case you're interested).
Anyway, the address I found is
1512 S No Le Hace Ave, Tucson, AZ 85713
Quickly, how many ways can you come up with to confuse a duplicate elimination program?
15125 North Le Hace Ave, Tucson, AZ 85713
1512 S N Le Hace Ave, Tucson, AZ 85713
1512 S Nolehace Ave, Tucson, AZ 85713
1512 South North Le Hace Ave, Tucson, AZ 85713
Without trying very hard, I could probably come up with a dozen other easy to make errors that standard duplicate elimination software would have a hard time finding. In fact, a lot of these errors – like changing “No” to “North” or “N” are probably caused by the software we all use to standardize and help eliminate errors on our lists.
But if you think the donor who was on the list 14 times was angry, you should have heard me. I’d been fundraising council for that client for 6 years. I’d worked so hard tweaking packages, testing new ideas, and doing my best to increase our response rate by just a quarter of a percent or so. Now I’m finding out that for 8 ½ % of the names I was sending out were duplicates? I could have a much bigger impact by getting rid of duplicates than I could ever have by increasing response. That’s when I decided to learn a lot more about data.