Following on from my previous post, I’ve been spending a lot of time looking at retail sites over the last few weeks and it struck me that they have the ultimate infinite choice problem.
What’s really interesting is how poor most of them are when it comes to filtering the products they sell in a way that is meaningful to the user. Even the market leaders are way behind the price comparison engines like shopping.com and pricegrabber. There is a lot we can learn from these guys who live and die by getting the user to an affiliate as fast as possible.
In retail that means making an on-line sale, in publishing it could mean downloading a whitepaper, watching a video or reading a review. So what does it take to create great filters? The answer is simple, anything the user will recognise is a great filter. Here are my top 5
- Brands - some users will buy brands way above all other criteria
- Price - likewise some users will buy based on price rather than quality
- Editorial recommendation - the trusted third party
- User recommendations/ratings - 62% of people trust their peers more than anybody else
- Automated recommendations - “users who bought this, also bought that”
Obvious common sense stuff, so why aren’t more retailers/publishers doing this? Two reasons, firstly, you can’t do comparisons if your data is dirty and most big databases of content or products are very messy. Secondly, tying user opinions/ratings to that content is not always straightforward. Integration between E-Commerce systems and Content Management Systems is usually pretty light resulting in disconnected data.
What the price comparison engines have invested in is great data tools so that their databases are as clean as possible. This lets them create all manner of filters , safe in the knowledge that the results will be right. It’s time for the rest of the on-line world to take that on board as the volumes of content increase exponentially otherwise overwhelming confusion will reign.
