Accessibility Adjustments

Use these optional tools to adjust reading and display preferences. These tools cannot resolve every accessibility barrier. Please contact the website owner if you need assistance.

  • Text adjustments
  • Content scaling 100%
  • Font size 100%
  • Line height 100%
  • Letter spacing 100%
  • Colour adjustments
  • Orientation adjustments

Facebook adds words to its recommendation controls

Listen to this article

Meta announced Your algorithm for Facebook on October 13. The rollout lets people manage topics for Feed and Reels and explain their preferences in ordinary language.

Where to find the controls

Open a post’s three dot menu to look for Your algorithm. After Interested or Not interested feedback, Facebook may also ask people to explain what they liked or disliked.

Meta says large language models interpret those requests alongside recommendation signals, with changes taking effect during the same session. That responsiveness is a company claim.

The announcement does not specify rollout countries, eligible accounts or a completion date.

Illustrative photograph by Moses Cursor Ssebunya. CC0 via the WordPress Photo Directory.

A reason carries more information than a reaction

Facebook already had direct feedback controls. Its October 2022 announcement described Show more and Show less as ways to temporarily raise or lower ranking scores for a post and similar content. A March 2025 update to that page records their later names, Interested and Not interested. The original announcement also placed Favorites, snooze and reconnect among the options in Feed Preferences.

Those controls address different problems. A person might want more from a particular account, a break from its posts or fewer recommendations resembling one item. An explanation can potentially narrow the reason behind that choice. Someone watching a camera review, for example, might care about its photography advice while disliking its emphasis on expensive equipment. A single negative reaction leaves that distinction unstated.

The useful test is whether the resulting selection preserves the part the person wanted. Removing every camera video would be easy to notice, but it would miss the point of that example. More detailed input only becomes meaningful control when the system applies the intended distinction. This is a question for testing, rather than a result established by the announcement.

Choosing topics and choosing an ordering serve different needs

Meta’s 2023 explanation of ranking described systems that make predictions about relevant content, including recommendations from accounts a person does not follow. It also documented Facebook’s Feeds tab as a way to view a chronological feed, and Favorites as a way to prioritize selected accounts. That history helps explain why adjusting subjects and choosing how posts are ordered are separate decisions.

A topic preference is useful when someone wants discovery within a subject. A chronological view answers a different question, what the accounts they follow have posted recently. Readers should choose the control that fits that goal. Giving a recommendation system a clearer instruction still leaves a recommendation system deciding which eligible items to show.

Similar Meta controls have different rules

Threads provides a concrete comparison. In June 2026, Meta announced Your Algo as a private setting for requesting more or less of specified topics. People could choose a duration of one, three or seven days. Meta said those requests would be visible only to the person making them and managed alongside Dear Algo in a shared hub. Those details belong to the Threads feature.

The comparison gives Facebook users practical questions to ask when the controls reach their account. How long does a preference stay active? Can they inspect and revise it later? Does the explanation change the subject, the format or both? Similar names across apps are a poor substitute for checking the choices actually presented on screen.

For an initial check, make one specific, ordinary request and inspect the recommendations that follow. Keep the wording narrow enough to judge. Then repeat the check on another visit. A pleasing first result is useful feedback, but consistent behavior across sessions is a stronger reason to trust a control.

Jordan Reid
Jordan Reid

Jordan Reid is focused on AI tools, agents, developer products, and the way technology changes everyday work. Jordan approaches a launch from the user’s side of the screen. What can it actually help someone finish? The voice is practical, conversational, and skeptical of products that turn a simple job into five new settings. Coverage follows coding assistants, creative software, browser agents, and the workflows around them, with attention to pricing, permissions, setup, and the human work that remains.

Leave a Reply

Your email address will not be published. Required fields are marked *

Gravatar profile