Twitter is experimenting with personalized lists of Who to Follow accounts. Interestingly, the company is accomplishing this using data generated by millions of embedded Tweet buttons around the web.
Above: Twitter thinks I need a coach
The Who To Follow functionality has been on Twitter for several years. It’s part of the effort to ensure that your experience of the service is engaging, but the list is currently extremely simplistic: almost all users see the same suggested accounts.
Now Twitter is starting to customize suggestions. Not simply, as you might initially imagine, by using an Amazon.com style “people who followed @JustinBieber also follow @philosophy.” That’s part of the data, but not the whole story.
Here’s how Twitter described it in a blog post today: “We receive visit information when sites have integrated Twitter buttons or widgets, similar to what many other web companies — including LinkedIn, Facebook and YouTube — do when they’re integrated into websites. By recognizing which accounts are frequently followed by people who visit popular sites, we can recommend those accounts to others who have visited those sites within the last ten days.”
In other words, Twitter is harnessing the wisdom of crowds to drive personalized, targeted who-to-follow suggestions that are highly likely to be interesting to you … because people who visit technology-centric sites are more likely to follow technology-centric Twitter accounts. Exactly how intelligent the algorithm is, Twitter is not revealing.
As with all online targeting, tracking, and analysis endeavors, privacy is clearly a concern here. Twitter does support the Do Not Track initiative, however, and will honor DNT browser settings. In addition, users can opt out on their Twitter settings page.
For existing users, the Who to Follow feature will look much the same as it currently does. But new users may see a new style of suggested accounts to follow when they sign up for Twitter.
Photo credits: Jez on Flickr, Twitter
VentureBeat’s VB Insight team is studying marketing and personalization...
Chime in here, and we’ll share the results