Facebook’s “Your Algorithm” lets users edit feed topics directly

Meta is bringing Instagram’s Your Algorithm to Facebook, letting users mark topics interested or not interested in feed and Reels. Creators and social teams may need to adjust content strategy as recommendations update in real time.

What happened

On October 13, 2026, The Verge reported that Facebook is introducing a new “Your Algorithm” feature, letting users directly add and manage topics they want to see in their feed and Facebook Reels. Meta says recommendations “respond right away” after people hit “interested” or “not interested” for different topics.

The feature was previously launched on Instagram earlier this year, and Meta also introduced a similar “Dear Algo” feature on Threads. On Facebook, users can open Your Algorithm or choose “interested” or “not interested” from the three dots menu on individual posts. Meta says the buttons may also appear under posts, and the platform “may ask follow up questions for more detail on what you liked or disliked. You can easily tell us, in your own words, what you want to see more or less of, and your recommendations will update in real time.”

Meta describes the update as part of a set of “AI-powered” tools used to customize what users see on Facebook. The company also says it is using LLMs more in its ranking systems. In Meta’s description, LLMs “can incorporate the interests, context, and intent behind natural language,” letting users tell the algorithm what they want more or less of.

Key facts

• Facebook’s new “Your Algorithm” feature lets users directly add and manage topics for their feed and Facebook Reels.
• Meta says recommendations “respond right away” after users hit “interested” or “not interested” for topics.
• Your Algorithm previously launched on Instagram earlier this year; Meta also launched a similar “Dear Algo” feature on Threads.
• Users can access the controls from the three dots menu on individual posts, and Meta may ask follow-up questions for more detail on likes or dislikes.
• Meta says it is using LLMs more in ranking systems and that recommendations adapt within the same session.

Our analysis

The shift matters because it moves Facebook’s ranking controls from mostly behavioral signals to explicit, user-managed topic inputs. Before, creators often had to infer why content performed well; now users can directly tell Facebook what they want in their own words, and Meta says the updates happen in real time. That likely reduces the lag between audience preference changes and distribution adjustments.

For social teams, the introduction of LLM-based ranking suggests Meta can parse more natural, nuanced requests than a simple “show less.” If a user says they want more practical tutorials and fewer memes, the system may be able to apply that across feed and Reels within the same session. This creates a faster feedback loop, but it also means topic misalignment may be penalized more quickly when users mark content “not interested” or answer follow-up questions.

“Tell it what you want more or less of, and your recommendations respond in real time, adapting within the same session to reflect what matters to you right now.”

That is a meaningful difference from previous ranking descriptions. The feature is not only a settings panel; it is becoming an interactive layer that can adjust on the fly. Creators and brand pages should treat clear topic labeling and audience prompts as distribution levers, not just content hygiene.

What it means for operators

• Make topic fit obvious. Because users can add and manage topics they want to see, posts and Reels should clearly signal their topic category so interested users can surface them.
• Use the feature as a research tool. Monitor your own feed and Reels, use the three dots menu to mark “interested” or “not interested,” and observe how recommendations change in real time to understand topic competition.
• Encourage specific feedback. If Meta asks follow-up questions, prompt your community to describe in their own words what they want more or less of from your page; this can feed the LLM signals for your content category.
• Sequence content within sessions. Since Meta says recommendations adapt within the same session, avoid posting off-topic content immediately after a strong positive topic signal; batch related posts around the same theme.