Facebook Your Algorithm: Take Back Your Feed
Social MediaFacebook Your Algorithm lets you tell your feed, in your own words, what you want more and less of. What changed, how it works and what it means for you.
You open Facebook to see what your cousin did at the weekend. Ten minutes later you've watched three cooking videos from strangers, scrolled past an AI-generated picture of a "miracle" waterfall, and still haven't found your cousin. You close the app feeling slightly worse than when you opened it.
If that sounds familiar, Facebook has just made a change worth knowing about. On October 13, 2026, Meta brought a feature called Facebook Your Algorithm to the main app. For the first time, you can tell Facebook, in your own words, what you want to see more of and what you're tired of, and your feed and Reels adjust right away.
It isn't a magic fix, and it doesn't switch the algorithm off. But it's the biggest change in years to how much say you have over your own feed. Here's what it does, how it works, and what it means for the way you use Facebook.
What is Facebook Your Algorithm?
Facebook Your Algorithm is a set of controls that let you shape the recommendations in your Feed and Reels by describing your preferences in plain language. Instead of hunting through settings or hoping the app notices what you skip, you can simply say something like "more local news and gardening, fewer celebrity videos" and Facebook will try to follow it.
According to TechCrunch, the app may ask follow-up questions to understand what you mean, and the changes show up in your recommendations in real time. Meta says the feature is designed to make content recommendations more relevant, and coverage of the launch says it uses large language models, the same kind of AI behind today's chatbots, to understand what you type.
There's also a quicker route for everyday use. The three-dot menu on individual posts now offers "Your Algorithm", plus "Interested" and "Not interested", so you can correct the feed the moment something feels off.
Where the idea came from
Facebook isn't the first Meta app to try this. Instagram introduced Your Algorithm for Reels in December 2025 and later expanded it, and Threads has a related feature called "Dear Algo". Facebook is the last of Meta's big apps to get it, and arguably the one that needed it most.
TechCrunch described the launch as a shift away from traditional recommendation systems, which quietly decide for you based on what you watch, like and linger on. The new approach sits closer to streaming services, where you can tell the app what you're in the mood for.
Why your Facebook algorithm feels messed up
"My Facebook algorithm is messed up" is one of the most common complaints people type into Google, and it's easy to see why. A few things have changed over the years.
More posts from strangers. Facebook used to be mostly friends and family. Today, much of what you see comes from pages and creators you don't follow, chosen because similar people engaged with them.
Recommendations follow behaviour, not intentions. If you pause on a car crash video out of shock, the system reads that as interest. It doesn't know you felt worse afterwards. Understanding how AI actually works helps here: these systems learn patterns from what you do, not from what you meant.
AI slop. Feeds are now full of AI-generated images and recycled clips built to grab attention. Some coverage of the launch framed it directly: tired of AI slop? It's no coincidence that the founder of Orkut is now campaigning to bring Orkut back with a network built around real people.
Put together, many people feel their feed no longer reflects who they are. Your Algorithm is Meta's answer to that feeling.
How Facebook Your Algorithm changes your feed
The practical difference is that you can now be direct. A few examples of the kind of thing you can ask for:
- More posts from local groups and fewer viral videos.
- More cooking and travel, less politics.
- Fewer Reels about a topic you've had enough of.
- More from friends and family, if that's what you miss.
Because the feedback is applied in real time, you can see quickly whether it worked and refine it. And because the controls also sit on each post, you don't have to remember to visit a settings page. If a post annoys you, you can tell Facebook why, right there.
This is also the closest thing to an answer for people searching how to stop the Facebook Reels algorithm from showing them the same thing again and again. You can't turn Reels recommendations off, but you can now tell them, clearly, to move on.
What Your Algorithm doesn't change
It's worth being honest about the limits, because the name promises a lot.
- The algorithm is still in charge. You're giving it better instructions, but Facebook's systems still rank and choose what you see. Coverage of the launch stressed that the underlying ranking system hasn't been replaced.
- It isn't a chronological feed. Your Algorithm tunes recommendations; it doesn't turn your feed into a simple time-ordered list.
- Ads aren't going away. Meta is an advertising company, and nothing in this update suggests fewer ads.
- Results depend on the AI understanding you. Language models are good at reading intent, but not perfect. Vague requests will get vague results.
Is it good for your privacy?
Here's the trade-off nobody puts in the headline. When you tell Facebook, in detail, what you like and dislike, you're also giving it a clearer picture of who you are. That's useful for your feed, and it's also useful for a company that earns its money from targeted ads. Some privacy advocates raised exactly this concern when the feature launched.
That doesn't mean you shouldn't use it. It means you should use it the way you'd use any AI tool that learns about you: share what helps, skip what's personal. You don't need to tell your feed about your health, money worries or family problems to get fewer celebrity videos.
The same principle applies across Meta's AI products. Meta Muse, the company's AI agent, can act inside your accounts once you connect them, which is powerful and worth handling with care. It's also why the wider industry is debating whether AI has a kill switch that people can use when it goes wrong. If keeping your information on your own device matters to you, tools like Google's offline AI Edge Foresight show that some companies are moving in the other direction.
Can you reset your Facebook algorithm?
People often ask whether they can wipe the slate clean and start again. This update is about telling the algorithm what you want rather than erasing what it already knows. In practice, a few weeks of clear signals, using Your Algorithm, "Interested" and "Not interested", and simply not lingering on content you don't want, will reshape your feed more than any single button.
A useful habit is to be deliberate rather than reactive. Instead of correcting your feed only when something annoys you, take two minutes once to describe what you actually want from Facebook. It's the same lesson that applies to AI tools in general, and a big part of why most people use AI wrong: they let the tool decide instead of telling it what they need.
Why this matters beyond Facebook
For years, recommendation systems have worked quietly in the background, one of the most widespread uses of AI beyond chatbots. They decide what billions of people see every day, and most users have had almost no way to argue with them.
Letting people speak to the algorithm in plain language is a real shift in that relationship. It treats users less like data points and more like customers who can say what they want. If it works, expect other platforms to follow. If it doesn't, it will at least show that people want more control, which is a message the industry has been slow to hear.
It also fits a pattern we've seen with AI assistants. The most useful tools are the ones that work from what you actually tell them, which is the core idea behind the NotebookLM vs ChatGPT debate: one answers from your own sources, the other from everything it has read.
A small tip for people who share a lot of photos
If you post family photos, recipes or travel pictures often, smaller files upload faster on mobile data and look just as good. A bulk PNG to WebP converter can shrink a whole folder of pictures at once, right in your browser, without uploading them anywhere first. It's one of our free online tools that keep your files on your own device.
The bottom line
Facebook Your Algorithm won't turn back the clock to the days when your feed was just your friends. But it gives you something you haven't really had before: a voice. You can now tell Facebook what you want, in your own words, and see the change straight away.
Use it, be specific, and keep your personal life out of the conversation. Your feed should feel like yours again, not like something that happens to you.
FAQs
What is Facebook Your Algorithm?
It's a Facebook feature, launched on October 13, 2026, that lets you describe in your own words what you want more or less of in your Feed and Reels. Facebook adjusts your recommendations in real time and may ask follow-up questions.
How do I control my Facebook algorithm?
Use Your Algorithm to describe the topics you want more or less of, and use the "Interested" and "Not interested" options in the three-dot menu on individual posts. Consistent signals over a few weeks reshape your feed the most.
Can I stop the Facebook Reels algorithm?
You can't turn Reels recommendations off, but Your Algorithm and "Not interested" let you tell Facebook to show less of a topic, and the changes apply straight away.
Does Your Algorithm remove ads or give me a chronological feed?
No. It tunes which posts and Reels are recommended, but ads remain and the feed is still ranked by Facebook's systems rather than shown in time order.
Is Facebook Your Algorithm safe for my privacy?
It's safe to use, but anything you tell it helps Facebook understand your interests, which also matters for ads. Share preferences about topics, not personal details about your health, finances or family.
Full Stack Developer & Tech Writer
Alex is a software engineer with over 5 years of experience specializing in web development, cloud computing, and developer tooling. He writes in-depth technical guides, tutorials, and articles on modern web architecture.


