Written by Lucy Hall and reviewed, fact-checked and signed off by a SocialDay editor before publication. Read our editorial standards and corrections policy. Spotted something wrong? Tell the newsroom.

The platform now predicts how long users will spend on content and prioritises posts people read to the end, making lightweight engagement tactics obsolete.

On October 20, 2026, Elon Musk announced two updates to X's timeline recommendation system: a new AI ranking model built to better predict how long users will spend on content, and a retrieval model that favours posts people actually read to the end.

The change marks a deliberate move away from traditional engagement signals such as likes, reposts and replies, toward what X believes is a better proxy for content quality: whether someone actually finishes reading it. The underlying logic shifts from rewarding light engagement to rewarding deep reading, with the algorithm now directly predicting and allocating traffic toward posts with high dwell time and giving priority to content that users are likely to read in full.

For brands and social media teams, the immediate consequence is tactical. Posting strategies optimised around quick-hit engagement, likes and shares, or formats designed to game legacy signals, are now working against the system rather than with it.

The retrieval model pulls in posts with high completion, meaning posts that users tend to read all the way through rather than abandon halfway, while the new ranking model's job is to predict dwell time, the industry term for how long someone lingers on a piece of content.

Source link