Facebook Your algorithm: a publisher’s content check
Meta announced topic controls for Feed and Reels. Review what is confirmed and use a practical editorial checklist without promising more reach.
On October 13, 2026, Meta announced the rollout of Your algorithm, letting people add and manage Feed and Reels topics through an updated three-dot menu. Interested or Not interested feedback may trigger an optional request for clarification in their own words. Meta says recommendations can respond within the same session. That is the company’s claim, not our test result. The announcement provides no country or language rollout list and announces neither Ads Manager targeting changes nor guaranteed publisher reach. Checked October 18; the announcement date remains October 13. Meta’s official announcement.
Editorial analysis: check the promise of a post
The following is our publishing workflow, not Meta ranking guidance. Its useful output is a clearer post and a record of the changes made. A reader adjusting personal recommendations does not give publishers control of other people’s feeds. Do not ask audiences to coordinate interest settings in an attempt to promote your page.
Choose one post whose purpose is difficult to identify. A fictional example about planning software might open with “make your team more efficient.” A clearer approach would explain how to assign responsibility for one task. This is an editorial exercise, not a customer case or a claim that a particular product supports that workflow. Verify any named product before demonstrating it.
A checklist before the next post
- Write the reader’s question in one sentence. Learning about a topic, completing a task and choosing a product are different intentions. Avoid a headline that promises to solve all of them at once.
- Compare the headline, opening screen and complete answer. Readers should receive the promised explanation without having to follow a commercial link. Remove unsupported earnings or time-saving claims.
- Check the example and its limitations. Use an authorised demonstration without customer records. Label an illustrative diagram clearly instead of presenting it as a real product screenshot.
- Inspect the next step. The destination should match the recommendation, and the affiliate relationship should be clear beside it. Explain when a simple spreadsheet is sufficient.
- Assign responsibility for answering questions and correcting errors. Do not collect personal interests, screenshots of readers’ recommendations or contact details merely to review your writing.
Observe outcomes without inventing causation
Before editing, record the date, address, headline, format, promise and link status. Note paid promotion, a different publishing time or a change in audience separately. After the edit, record substantive questions and voluntary responses. Where comparable click measurements are available, use the same observation window and event definition. Do not combine unrelated totals just because they look like a funnel.
A difference between two posts is not evidence of an algorithm effect. Topic, format and distribution can also change; this small editorial review does not isolate a cause. Silence is not proof of failure either: a reader may have found the answer without commenting. With sparse evidence, leave the conclusion open and first assess whether the content fulfils its promise.
If the new setting is absent from your own account, record that as a local observation. Do not generalise it into a country-wide availability claim or make the setting a prerequisite for the content plan. Reviewing clarity does not require the button to appear. Do not borrow identities or bypass access restrictions to create a demonstration.
A relevant next step for affiliate content
In our AffiliateCore platform, compare the prepared material with the actual terms of an available offer: geography, permitted traffic source and payable event. Existing publishers can sign in to AffiliateCore with a short description of the source and the post’s promise. Offer acceptance and an active campaign are separate checks. AffiliateCore does not control Facebook recommendations or guarantee a suitable offer.
For video content, see our guide from a short video to a measurable visit. The output of this review is a clear answer, a transparent recommendation, a working destination and an honest list of unknowns. Additional reach, registrations or revenue remain hypotheses, not results established by this announcement.