Whether Fav.or.it was abandoned due to it being a failure isn’t clear from CEO Nick Halstead’s post, although as Louis Gray points out in his coverage, the main driver behind abandoning development of Fav.or.it was the success of the company’s newer product Tweetmeme.
We’ve not written much about Tweetmeme previously; Steven Hodson mentioned Tweetmeme when writing earlier today about the risks involved with basing a business on the Twitter API, and I do agree with his analysis. There is though another aspect about Tweetmeme that needs to be explored: that’s the Techmeme 2.0 aspect.
There is a lot to like about Tweetmeme: the sites provision of retweet buttons for example has been a huge success, although how you monetize that is another matter. The overall idea of creating a meme tracker based on Twitter links is fairly sound, and of all the similar services in this space, there’s little doubt that it is currently the market leader.
The problem I have with Tweetmeme is the results. It’s not entirely the developers fault, but the results show an over-riding bias to sites that have Twitter accounts that are offered as “recommended subscriptions” by Twitter itself when new users sign up. It’s Techmeme 2.0, although with Mashable riding alongside TechCrunch in this instance (Mashable barely gets a look in on Techmeme.) Tweetmeme doesn’t primarily reflect the best content shared on Twitter, but the content shared by those with the most followers because those accounts will always have more shares.
Of the 15 top items on the front page of Tweetmeme as a write this, two items are ads, six items are from Mashable, and three items are from TechCrunch. Four items are from accounts that don’t get favored status from Twitter, and would appear by all accounts to have legitimately earned their place on the front page through organic sharing.
If the purpose of a meme tracker is discovery of new and interesting content, Tweetmeme fails: I can easily subscribe to and read Mashabe or TechCrunch, and in the case of Mashable I do.
What Tweetmeme needs to do is develop an algorithm that levels the playing field to some degree. A Twitter account with 10,000 followers who has 100 retweets should be ranked more highly in the results than a Twitter account with 1 million followers who has 100 retweets. It isn’t as simple as this in implementation (you couldn’t ratio down to 100 followers for example due to spam issues), but it’s not impossible either.
It is fairly early days for Tweetmeme, and as much as company head Nick Halstead would be no fan of mine, he would appear to be highly talented and able to make the change. All that is needed is the will to make Tweetmeme a better product.