Skip to content

Facebook's Your Algorithm lets you describe your feed

Meta launched Your Algorithm on Facebook on October 13: users describe in their own words what they want more or less of in Feed and Reels.

By Tech AI Wire Team

3 min read

XLinkedIn
Two phone screens from Meta's demo: a Facebook post menu offering Interested, Not interested and Your algorithm, and a Your algorithm page listing topics.
Photo: Meta

Meta launched a feature called Your Algorithm on Facebook on October 13, 2026. It lets people describe, in their own words, what they want to see more or less of in their Feed and Reels. A large language model reads that description and adjusts recommendations in real time. It is the clearest sign yet that the platform's ranking system will take direct instructions, not only clicks.

How Your Algorithm works

According to Meta's announcement, the feature sits in the three-dot menu on posts. Meta updated that menu to make the controls easier to find. From there, users can add and manage topics they want to see.

The bigger change is free text. Users can type what they want more or less of, the way they would describe it to a friend. After someone taps "Interested" or "Not interested," Facebook may ask a follow-up question about what they liked or disliked. Meta calls this a "more direct, conversational way to shape your experience."

In Meta's demo video, the Your algorithm page opens with a short summary of the user's recent interests. Below it, topics are sorted into "Interested in" and "Not interested in" lists, with an option to add more. The page notes that the interests are based on recent activity and summarized by AI.

What runs underneath

Meta says large language models interpret the interests, context and intent in what users write. Those readings are combined with the signals Facebook already uses for ranking, such as likes, comments and watch behavior.

Recommendations update in real time and adapt within the same session, Meta says. TechCrunch reports the same: changes show up in recommendations right away.

Where it came from

Facebook is not the first Meta app to get this. TechCrunch traces the feature's path across the company:

AppFeatureWhen, per TechCrunch
InstagramYour Algorithm for ReelsDecember 2025
InstagramYour Algorithm extended to Feed and ExploreJune 2026
Threads"Dear Algo," for influencing recommendationsAlready available
FacebookYour Algorithm for Feed and ReelsOctober 13, 2026

What Meta has not said

Meta says the feature is rolling out but gives no countries, languages or dates. Neither source mentions Groups. Meta also does not say how long the text people type is kept, or whether it is used for anything beyond recommendations.

What this means for developers

For people who build products on top of Facebook reach, or run teams that depend on it, a few things change:

  • Expect stated preferences to matter more. Ranking now blends what people say they want with how they behave. Publishers and creators who rely on engagement alone may see reach shift as users name the topics they are tired of.
  • Write clear posts on clear topics. If a language model matches posts to described interests, a post whose subject is obvious is easier to match. Vague teasers give it less to work with.
  • Watch the privacy questions. Users are now typing personal preferences into a box that feeds an AI system. Teams handling compliance should look for Meta's answers on retention and use before building strategy around the feature.
  • Track your Reels separately. The feature covers both Feed and Reels. Reach changes may show up in one before the other, so measure them apart.
  • Expect the same pattern elsewhere. Instagram, Threads and now Facebook all let users steer recommendations. Teams building their own recommendation systems can treat plain-language controls as a feature users will start to expect.

Sources

  1. Your Algorithm, Your Words: New Ways to Shape What You See on Facebook - Meta
  2. Facebook is giving users more control over their algorithm - TechCrunch

Related articles

The weekly digest

One email every Friday with the week's top stories from all six desks: AI, dev tools, coding, the tech industry, startups and what's next. Free, no spam.

Unsubscribe anytime with one click.