OK so we really don’t version our projects like that. The point is that this is a really early stage project.
From the very beginning, prior to even deciding whether or not to have a paid version, we recognized that there was a lot of opportunity in targeting ads to sports fans. SportsLizard does this in a very basic way. On our Price Guide we target eBay ads based upon what card/collectible the user is valuing. Search for a Michael Jordan card, and we’ll show you a banner ad of Michael Jordan cards. It’s worked really well for us over the years, far better than AdSense. As a whole, the advertising revenue for SportsLizard almost doubled in 2010 and has continued to grow into 2011.
The premise for the LockerPulse ad platform is that using the eBay and Amazon API’s we can hyper target ads to users based upon what they’re viewing, their preferences, their history, their location, and other factors. I’m of the belief that there is something on eBay or Amazon at any given moment – tickets, memorabilia, apparel, books, movies, etc – that every single sports fan is interested in looking at. It’s a matter of us being smart enough to figure out what to show, how to show it, and when to show it. We’ve (predictably) seen a lot more sign ups come through since making the site completely free, which will help us gather even more data. The more active the user, the better ad we’ll be able to show them (in theory).
We have been running AdSense tests for a few months now. The click-through-rates were very encouraging and act as a great baseline to try to beat. Which is why to start we essentially mimicked AdSense in as many ways as possible, including the fonts, the colors, the size, and the location on the page. We tried to not make too many educated guesses. The goal at the moment is to simply collect data and try to beat AdSense. Right now there is an ad at the end of every single story on both the regular site and the mobile site. Assuming this works, the sky is the limit.
I spent a ton of programming time integrating effectively with Amazon and eBay, and then building caching systems on our end so we’re not creating an ad on the fly every time you access a story. I’ve done my best to set everything up to scale well as we grow and add functionality. Both API’s are OK to work with, but are littered with documentation for deprecated functionality (or functionality that is soon to be deprecated). The challenge was in figuring out prior to starting what documentation is actually valid. There also wasn’t much sample code out there, I sort of figured it all out on my own. They also don’t integrate well with their affiliate programs, which have separate logins and documentation. For all of the potential with these API’s, I wonder how many are turned away by the unnecessary complexity?
Anyway, just getting this launched is very exciting for us. We could have spent months more adding features and fine tuning things, but to be honest we really don’t know will work and what won’t. As with most web projects, the data and customer feedback will go a long ways in dictating our next move and ensuring we don’t waste time on unnecessary stuff.
Oh, and if you happen to come across an ad that isn’t relevant I’d greatly appreciate you leaving a comment or dropping me an email. For instance, we’ve discovered situations where we’ll show ads for “Buffalo Bill” the Western movie(s) instead of Buffalo Bills stuff. I’m making a list of these things. I’ll go back and clean them all up on the next iteration.