Magicstats

Twitter Continues Tweet Recommendation Experiments With Fav Happy @Magicstats

Next Story

Box Acquires dLoop To Enhance Security With Fine-Grained Data Analytics Technology

Twitter continues to pull the string on personalization and recommendation with a new account called @magicstats. The account, whose description reads ‘I favorite the best tweets I see in real-time’ appears to be doing just that.

Though the account is protected and its activity closed down, we’ve been tracking it for several weeks now and have gained a bit of insight into how it works. Several of my personal tweets and some TechCrunch tweets have been favorited by the account, which appears to be working on several metrics including velocity of activity (like favs and retweets from other accounts). We investigated the account and it’s followed by many members of Twitter’s search and relevance team, just like Twitter’s other experiments @magicrecs and @eventparrot.

Screen Shot 2013-11-26 at 5.07.21 PM

The account, from what we understand, was originally dedicated to monitoring the Magicrecs experiment, but was repurposed to its new role once that was rolled into the product proper. Previously, the account’s profile said it was recording data related to Magicrecs. Twitter declined to comment about the status of the account.

The account’s tweets were not originally protected, and we were able to browse its favs for a while. The account often awarded a fav to a tweet very quickly after it was posted, but we never saw it fav a tweet ‘first’. It always followed other favs and RTs very quickly. Judging by this, it’s likely that the account is looking for tweets that are getting rapid attention in its network or on the Twitter network as a whole.

Twitter has noticed the rise of products like the third-party fav-and-RT-tracking Favstar, which a source with knowledge of its user base told us was ‘surprisingly popular’. The service offers a visual dashboard of re-tweets and favorites that are tied to user profiles and presented very well. It’s fun, visual and addictive.

Screen Shot 2013-11-26 at 5.09.32 PM

Twitter also has a statistics package, but accessing it is a bit obscure as you have to go to the ads.twitter.com page and sign in with your account there. The page is interesting and information packed, but it’s not nearly as fun as the Favstar profiles are. I check Favstar several times a day at least, just to see how recent tweets faired, and it can be an interesting long-term indicator of what the best tweets on your account are.

Screen Shot 2013-11-26 at 5.35.01 PM

And that, in the end, could be the bit that interests Twitter the most, which tweets are the most popular, how fast to they become popular and how far they reach. Some of Twitter’s most popular vanity metrics are centered around the ‘most re-tweeted’ tweets, like the famous Obama ‘four more years’ tweet that was called out in its S-1 filing.

This focus on finding the tweets most likely to ‘go viral’ or simply those that are garnering the most attention appears to be continuing on the back of the success of Magicrecs. As I stated when we broke the news of Twitter’s ‘breaking news’ experiment @eventparrot, Twitter’s best bet to keep existing users engaged and happy — and to demonstrate the utility of the service to new users — is to focus on personalization.

“When we think of the purpose of Twitter, what we’re able to do, making it so any person in the world can communicate with any other person, connecting all the people on the planet, that is an incredible mission to be on,” said Twitter SVP of Engineering Christopher Fry in a recent Wired interview. “We’re probably still early in that mission, but that is the goal: that any one person can communicate with every other person in the world.”

Making everyone’s Twitter feel helpful in a unique way could be a powerful motivator for people to keep using the service. It could also contribute to Twitter becoming what I like to think of as an ‘Internet pillar’, like Google’s search product or cloud services like Amazon’s AWS. Yes, it will definitely keep iterating on consumer products in order to bring in ad money, but the potential as a personalized tool for communication is what’s most exciting.

Image Credit: Atalanta#5