Sorry I missed your question Jeff; I just saw it now.
fI have never used the longest common substring to solve a business problem. I was posted on a forum few years ago and I thought it was a fun and challenging exercise. Today I use it to when showing people how to use tally tables - it seams to make a good impression on people new to the concept.
One thing you can do with a Longest common substring function is do some preliminary plagiarism detection. I have a couple examples but can't find them at the moment. Let's say, for example, you have a website like SSC where authors can submit articles. You could (after some initial cleanup such as making breaks/line feeds/carriage returns uniform earlier in the routine) compare a submitted article to all existing articles in your database to check for a longest common substring longer than <user defined number>. If the query finds that an article has a shares a, say, 90-character substring with another article, the editor is notified to review and see if the submitting author is steeling someone's work or is correctly referencing another article where they give credit.
"I cant stress enough the importance of switching from a sequential files mindset to set-based thinking. After you make the switch, you can spend your time tuning and optimizing your queries instead of maintaining lengthy, poor-performing code."
-- Itzik Ben-Gan 2001