There is a story out there that has been told again and again by the big Internet companies. It goes something like this, “We track your searches, scan your emails, and store every click you make on our site, to deliver you the best user experience possible.”
I’ve rarely seen any improvements in my user experience brought about by storing anything except my login information and the data that I specifically asked to have stored.
Most instances of personal data being used to deliver a “better user experience,” revolve around advertising, and the incredibly large premium that advertisers will pay for a well targeted ad versus and untargeted ad. Storing user data has very little to do with delivering you a better online experience and everything to do with delivering you to advertisers at a higher price. It’s not about user experience, it is about user segmentation.
This is why I was so happy to see Jason Calcanisis suggest that AOL should simply trash its log files immediately.
Frankly, I want us to NOT KEEP LOGS of our search data. Yep, you heard that right… we shouldn’t even keep this data. I know that’s crazy, but I learned this week that Wikipedia turned off their log files. They did this for tech reasons, but they now are keeping them off and not looking to solve the problem because of the huge upside of users knowing their searches on wikipedia DON’T EVEN EXIST!
I think we should use this as a way to brand AOL Search: We don’t record your searches!

After reading Jason’s post my second thought was that the entire community of net billionaires was going to crucify him for saying this, which is why it isn’t a surprise that he announced a blogging vacation yesterday.
I think I’m gonna take the rest of the week off from blogging as a “cool down” period. I don’t want to say something I regret, and I don’t want to become the spokesperson for the entire company–that’s not my job and it’s not my desire. I just want to build cool stuff with cool people I respect.
Did some AOL higher-ups get to him and suggest a short break before he cut the heart out of the Internet’s tried and tested model for sensational profitability (delivering you to the advertisers)?
I think that the idea of ditching all user data except that which we ask companies to store is a brilliant idea. Advertisers are not really going to understand what a person wants by looking at this data gathered on the sly. It just doesn’t tell enough of the story. Even the NY Times article on Searcher No. 4417749 doesn’t tell you who Thelma Arnold really is. If you want to know who someone is and what they want just ask them.
I know this is crazy talk to technophiles, who think human beings can be reduced to numbers, but asking Thelma what ads she wants to see is, well, the fastest way to determine what ads Thelma wants to see. Don’t try to be omniscient. Advertisers always get it wrong when they do this. Just ask.
We the readers understand that advertising based media companies will always require that we see their stupid ads, but if they really care about us as they often profess, then they will let us chose the types of ads that we see. Give us a choice of ad channels and we’ll tell you the best ads to serve. Give us a universal advertising profile that we can customize to defend ourselves against all the stupid ads that we don’t want to see, and you will always know exactly what we want. It’s a relatively simple idea to put the ranking and streaming of ads in users hands, but more difficult to put into practice. I’m interested to learn about any organizations that are actively engaged in this type of project.