Link Tracking – (lies, damn lies &) Statistics?

The last couple of months have shown a shift in Internet traffic – from being search engine (read Google) directed to social media driven. I believe, the main reason for this is what Fred Wilson calls the Power of Passed Links. That is, I click more often on the links that my friends tweet in Twitter or share on Facebook than on those I find by searching.  Fred has seen this dramatic shift on his own website. Earlier TechCrunch presented a similar observation and was quick to embrace a personalized short domain name (tcrn.ch).

In a recent article titled “How Twitter Might Send Far More Traffic Than You Think“, Danny Sullivan, editor-in-chief of Search Engine Land, compares the statistics provided by Bit.ly and Google Analytics and is awed by the difference. My own sense is that before all of us get rolling on this Social Media Optimization (SMO) effort, it would help to take a hard look at how link statistics are gathered and tracked by tools such as Bit.ly.

Simply put link tracking tools store the number of times a short link has been looked up. When such tools first emerged they captured imagination of a lot of people as potentially a simple replacement for Google Analytics.  However, the statistics provided by link tracker function in tools such as Bit.ly can be very misleading.

Since only the link look-up is counted by link trackers, all search engine and Twitter crawlers that de-reference a short link get counted erroneously as look-ups. Most of the newer Twitter clients de-reference a short url, and show the target domain name, complete url or the title in the tweet message. Which effectively means that the moment you publish a Bit.ly link on Twitter, it’s link count could potentially shoot up to as many followers as you have, even if none of them actually clicked through the link.

In my view, this is one of the biggest challenge that Bit.ly and others of its ilk will face to present an actual view’ed link as against merely looked-up link.

In the near future, I anticipate that more people will use their own short domain names and their own url shortening service to protect their brands.  Pure play URL shortening services are bound to get commoditized. For pioneers such as Bit.ly to retain their market share, it is critical that they evolve to provide vanity shortening services and emerge to be the analytics tool of choice for SMO by providing ever more accurate viewed statistics as against looked-up statistics.