More data mining….

Last week, I discovered Netflix can email you a history of all the movies you’ve rented, and when they were checked out and checked in. I thought, “Neat, I can use this to tell how many movies I’ve rented and what my average rental time for a movie is.” So I started writing a parser. Tonight, I’ve been working on adapting that parser to the web so other people can run stats on their Netflix histories.

You can see an example of what it does here. The sample is just a small subset of my history. I’ve run it on the full thing, and I know my average turnaround time is about 27 days and each movie over my entire history with Netflix has cost me just under $7. I know the shortest turnaround time from Champaign to St. Louis, even if I watch them as soon as I get them and put them in the mail the next day is 7 days. Pathetic, eh?

What I hope to accomplish by this parser is a better understanding of how I use Netflix, and hopefully increase my usage of it to make it more economical (the more movies I rent per month, the more value I get out of it.) I guess another fun thing to do would be to compare my rental ‘out’ times to what it would cost me to do the exact same thing with Blockbuster and see if Netflix, Blockbuster online, or Blockbuster traditional rentals (with or without late fees,) would be a better deal for me.

If you are a Netflix subscriber and want to help me test it, or think it should have new stats/features, please let me know. The only “bug” I’m aware of at this time is it assumes a 3-out plan. I should fix that to either figure out how many movies you have out by the history, or prompt the user. I’ll probably clean it up a bit and lure some of the Netflix bloggers to it and see what they think.

Geeky, but cool? Fun, and useful? I can live with that.