Social Media Algorithms Explained: Facebook’s New Controls, Instagram Ranking

Social media algorithms keep changing, so how should a brand adjust? Start with the part that doesn’t change: on Facebook and Instagram alike, the algorithm is predicting one thing — will this person be interested in this post? Content that clearly fits an interest gets reach. The newest twist is that users can now tell the algorithm what they’re interested in, directly and in their own words.

Social Media Today reports that Facebook has added plain-language controls to its existing “Your Algorithm” option. Users can type, in everyday language, what they want to see more or less of. It works in both Feed and Reels, and changes take effect right away. Facebook frames it as unlocking an algorithm you can actually have a conversation with.

What exactly changed in Facebook’s new algorithm controls?

The original “Your Algorithm” let users add or remove topics from a list of interests the system had inferred about them. The new version adds a chatbot-style prompt where users simply type a description of the content they want.

The second change is follow-up questions. According to the report, after a user taps “Interested” or “Not interested” on a post, Facebook may ask a few more questions so the user can explain, in their own words, what they want more or less of. Meta also says the system adjusts within the same session to reflect what the user cares about in that moment.

Social Media Today also points out that most users rarely change their settings, so the real-world impact remains to be seen. But the direction is clear: algorithms will lean more and more on interests users state out loud, not just on clicks and time spent.

How do social media algorithms rank posts? Facebook and Instagram, placement by placement

“The algorithm” isn’t one system. Each placement solves a different problem, so each ranks on different signals. The table below is a general overview to help you decide what to post where:

The pattern is plain. The more a placement recommends content to strangers, the more it depends on whether the content itself clearly targets one interest. The more it serves existing followers, the more it depends on your relationship with them. Put the same post in the wrong placement and the results can be worlds apart.

Why do brands always feel they can’t figure out the algorithm?

Economists call this information asymmetry: one side of a transaction knows more than the other. The platform sees every user’s behavior; a brand sees a handful of numbers in its own dashboard. Users know what they want to see; brands can only guess. The wider the gap, the easier it is to pour budget and effort into the wrong content.

Facebook’s new controls narrow that gap somewhat. Users’ interests no longer hide only inside clicks and time spent — they get spelled out in words. But that data stays with the platform, and brands don’t see it. What a brand can do is learn to understand its audience through its own channels. That’s the core idea of this article: understanding is the foundation of connection.

A hypothetical example: a Hong Kong handmade soap brand has always built its posts around “natural ingredients.” Only after combing through its comments does it notice that customers mostly ask “Is it OK for sensitive skin?” and “Can I use it on my baby?” Once it shifts the topic from what the brand wants to say (“natural”) to what the audience actually says (“bath time for sensitive skin”), its content finally lines up with what people are interested in.

How will an algorithm you can talk to affect brand reach?

Three things are reasonable to expect. First, the interests users can put into words are usually specific topics, such as “more home organization tips” or “fewer posts that feel like ads.” Vague content is hard to file under any of them.

Second, hard-sell content gets riskier. Once users say outright what they don’t want to see, the system adjusts immediately, so engagement bait and clickbait headlines are more likely to get filtered out.

Third, accounts with a consistent theme gain an edge. If your page posts food today, current events tomorrow and a giveaway the day after, neither the algorithm nor your audience can tell which interest you belong to.

This matters even more for Hong Kong and Taiwan brands getting ready to sell in the US. American users will describe their interests in English, so your topics need to be defined in their words. Rather than translating your Chinese product categories directly, first look at how US customers describe the same thing in reviews and comments.

How do I adjust my social strategy so the algorithm works for me?

1. Define three topics your audience would actually say. Use customers’ words, not marketing jargon: “small-apartment storage,” not “space solutions.”
2. State the topic at the start of every post. In the first three seconds of a video or the first line of a caption, make clear what the post is about.
3. Wrap sales messages inside useful content. Products introduced through tutorials, comparisons and FAQs get skipped less than a shouted promotion.
4. Design Feed and Reels separately. Write Feed posts for regulars; write Reels for people who don’t know you yet.
5. Read comments and direct messages (DMs) once a week. Note the exact words people use to describe their needs, and turn them into next week’s post topics.
6. Track shares, saves and full views. These actions signal “I want to see more” far better than likes do.
7. Test it on your own account. If your personal Facebook already has the feature, tell the algorithm you want content from your industry, then watch whether your posts and your competitors’ show up — and which kinds.

If your team is too small to keep up this pace every day, see how ToucHigh’s Social Autopilot can take on the writing and scheduling.

Which algorithm myths can you let go of?

• “There’s a perfect time to post”: Timing matters, but whether the content targets an interest matters more. Check your own analytics before trusting a generic schedule from the internet.
• “More hashtags are better”: A clear topic matters more than the number of tags.
• “A drop in reach means I’m being throttled”: First check whether your topics have gotten vague or your audience’s interests have shifted, then draw conclusions.
• “When the algorithm changes, I have to start over”: Targeting a clear topic, offering useful content and genuinely replying to your audience hold true through every update.