Twitter moves to cookie-based ads, bans automated following

A new trial will show personalised ads for US users, as well as bringing a raft of minor changes to the way that tweets are displayed on external sites.
Written by Chris Duckett, Contributor

A series of updates released by the social media company overnight have brought slight changes to the way that tweets are shown on third-party sites, but for US-based Twitter users, the big change is the impending arrival of promoted tweets based on a user's browsing history.

Kevin Well, Twitter senior director of product for revenue, wrote in a blog post that the new ads will not increase in quantity, but the intention is to increase the quality of ads displayed to users.

Using the example of a florist shop, Well detailed the process used by Twitter and advertisers for targeting customers.

"The shop may share with us a scrambled, unreadable email address (a hash) or browser-related information (a browser cookie ID). We can then match that information to accounts in order to show them a promoted tweet," Wells wrote.

"This is how most other companies handle this practice, and we don't give advertisers any additional user information."

For users who are concerned about the privacy implications of such a move, Twitter's Do Not Track support spares users from new browsing history-derived ads.

Of interest to developers and third-party users who make extensive use of Twitter's API are the changes to Twitter's display requirements and API terms of service.

Among the changes are the need to opt out users under the age of 13 from any Twitter tracking that occurs with the use of Twitter website buttons, an allowance for languages that display from right to left to have a user's avatar on the right-hand side, and the removal of the need for a tweet's permalink to be linked from the tweet's time stamp.

The display requirement now allows mixing Twitter and non-Twitter data as long as tweets are "explicitly selected by an end user to appear next to non-Twitter content". Examples given in the document of such an instance are manually authored live blogs, manually curated media collections, and commenting systems.

Developers of Twitter-based services are now required to display a privacy policy before the service or application is used for the first time, exporting Twitter datasets as a service is now banned, and automated following or unfollowing of users will now result in suspension.

Editorial standards