One of the projects I was working on for a few years involved cleaning up Address data for a customer database. I found out the admin did a lot of manually data correction. That works well for a customer base of less then 1,000, maybe even less than 500. In our situation, we had 40k customers at any given moment. That is a lot of tedious work.
When the admin was gone for vacations, I would take over the responsibility of maintaining the member’s database. Due to the restriction of having only x amount of hours in a day, I took some time to test out fuzzy lookups to manage the data.
The basic concept is that I can automate the computer to take a look at one list (in this case, the member’s database) and compare it to another list to see what are the strongest matches. Not being familiar with Hawaiian street names at the time, it was difficult to do it manually. So I decided I would try to compare a confirm address list and the customer list I had to maintain.
I never convinced the admin that this would help them with their job. I’m thinking it’ll be useful in the future to minimize “Lost in Translation” moments.
Anyway, to help you understand better, here’s a quick video on Fuzzy Lookups that I thought was useful: