Now that I have satellite TV and 3 kids I’ve cancelled Netflix, but I’ve always respected their attention to usability and the way movies are rated and recommended. The NY Times has a great article about the success of their rating system and their recommendation engine, but also Wal-Mart’s dismal failure in providing recommendations. When Wal-Mart.com user’s viewed movies such as a box set about Martin Luther King, Jr. they were recommended “Planet of the Apes” among other embarrasing titles. They had to remove all recommendations and control the PR Spin.
“The most reliable prediction for how much a customer will like a movie is what they thought of other movies,” Mr. Hunt said. The company credits the system’s ability to make automated yet accurate recommendations as a major factor in its growth from 600,000 subscribers in 2002 to nearly 4 million today. NY Times article (free registration required)
Not only are the recommendations useful to the subscribers, but it also helps move back-inventory of older movies, relieving some of the pressure (cost) to keep up with massive quantities of new releases.
At NetFlix, the online DVD rental company, for example, roughly two-thirds of the films rented were recommended to subscribers by the site – movies the customers might never have thought to consider otherwise, the company says. As a result, between 70 and 80 percent of NetFlix rentals come from the company’s back catalog of 38,000 films rather than recent releases.
One last note… Recommendation engines are not useful if they are not kept updated. The recommendation engine at CommunityCoffee.com was built in 2004, but has not been updated since. The engine is run off of a static file because the products rarely change and it was based on about 20 years of purchase history. I know because I architected the site and the engine:)
I haven’t done any work for Community Coffee since early 2005, but since I’ve been gone, they’ve changed the SKU numbers for many of the products, of which the engine is based. This item, French Vanilla coffee, should recommend another flavored coffee (featured products in the right margin), such as Hazelnut, but it is defaulting to the most popular SKU numbers on the site.