TL;DR
Most major social platforms offer controls that let users nudge recommendation systems to show more or less of particular topics. The Verge overview catalogs options on Facebook, Instagram, Threads, X, TikTok and notes that some platform details (for example, on YouTube) were not fully available in the source.
What happened
An overview published on Dec. 29, 2025, outlines the ways users can influence algorithmic recommendations across several large social apps. The piece explains that recommendation systems generally use demographic signals, account activity and patterns from similar users to surface content, and that most services include “more/less” style controls. On Facebook, users can unfollow pages, clear activity history from an Activity log and mark individual posts as “Interested” or “Not interested”; settings also let people reduce political or sensitive content suggestions. Instagram offers an AI-generated summary of interests for Reels where users can add, remove or downrank topics; the main feed also supports Interested/Not Interested flags. Threads supports “Not interested,” muting words, restricting profiles and is testing a text-based “Dear algo” request tool in beta. X and TikTok provide feed-level menus and toggles to uncheck interests, mute keywords, flag videos as not interesting and, in TikTok’s case, a “Refresh your For You feed” option to retrain recommendations. The article mentions YouTube’s tuning tools but the source text cuts off before details are provided.
Why it matters
- Users can reclaim some control over what algorithms surface, shaping discovery and daily browsing.
- Platforms provide tools to reduce exposure to sensitive or political material, which affects what appears in feeds.
- Adjusting recommendations can change what content reaches you without removing accounts or leaving networks.
- Algorithmic tuning affects the signals platforms use to personalize content, which can influence time spent and what creators reach you.
Key facts
- Facebook: Settings & privacy > Content preferences includes an Unfollow people and groups option; Activity log lets you delete interaction histories; individual posts offer “Interested” or “Not interested.”
- Instagram: Reels show an AI-generated summary of topics; users can delete interests, add new ones or mark topics to see less of; Interested/Not Interested is available on Reels (mobile) and the main feed (mobile and desktop).
- Threads: Users can mark posts as “Not interested,” mute words, block or restrict profiles; a beta “Dear algo” text request feature is being tested and was seen by the article’s author in the UK.
- X: In the For You feed you can mark individual posts “Not interested” from the feed timeline; Settings > Privacy and safety > Content you see lists interests you can untick and lets you mute accounts and keywords.
- TikTok: Long-press or use a video’s three-dot menu to mark content not interesting; mobile Content preferences include Manage topics with sliders and a Restricted Mode; Refresh your For You feed can temporarily show different popular videos to retrain the algorithm.
- Platforms commonly base recommendations on demographics, past activity, and interactions from similar users, per the article.
- Meta has announced additional upcoming Feed controls for Facebook, including options to explain why items may be irrelevant, according to the article.
- The source indicates that some tuning features are mobile-only, and that desktop capabilities can be more limited on certain platforms.
- YouTube: the article refers to YouTube tuning tools but the source text is truncated and specific details are not available in the provided material.
What to watch next
- Meta’s planned feed controls and any wider rollout of Threads’ “Dear algo” requests — the article says new Facebook Feed feedback options are coming and Threads’ feature is in beta.
- Whether Threads’ Dear algo testing expands beyond the author’s observed UK beta — availability was described as limited in the source.
- How quickly changes take effect on X after users untick interests — the article notes X warns changes “may take a little while to go into effect.”
- YouTube tuning details: not confirmed in the source.
Quick glossary
- Algorithmic recommendations: Automated systems that select and order content for users based on signals like activity, preferences and similar users’ behavior.
- For You / Reels feed: A personalized content stream on platforms (e.g., TikTok’s For You, Instagram Reels) that surfaces items the platform predicts a user will engage with.
- Interested / Not interested: User-facing feedback controls that tell a platform to show more or less content like a specific post or topic.
- Activity history: Records of a user’s interactions — such as likes, comments, searches and watches — that platforms may use to inform recommendations.
- Muted keywords: A filter that prevents posts containing specified words or phrases from appearing in a user’s feeds.
Reader FAQ
Can I make a platform show me more of a topic I like?
Yes — several services let you mark posts or topics as “Interested,” add interests manually, or follow topics to increase related recommendations.
Will clearing my activity stop those recommendations?
The article says deleting interaction histories on platforms like Facebook may help reduce similar suggestions, but results can vary.
Is Threads’ Dear algo available to everyone?
Not confirmed in the source — the article notes the feature was seen by the author in the UK and is in beta, which may not be available to every user.
How can I retrain TikTok’s algorithm?
TikTok offers a “Refresh your For You feed” option that temporarily shows different popular videos so the app can learn new preferences.
TECH CREATORS HOW TO How to tweak your online platform algorithms Most platforms give you some control over what appears in your recommendations and ‘for you’ feeds. by Jess Weatherbed…
Sources
- How to tweak your online platform algorithms
- Social media algorithm: 2025 guide for all major networks
- Everything You Need to Know About Social Media Algorithms
- Daring Tactics For Combatting Algorithms In Social Media
Related posts
- Experimental Carnegie Mellon camera lens can focus every object at once
- Tesla appears to be developing Apple Wallet car‑key support for iPhone
- How One Hand Operation+ and RegiStar fixed my Galaxy’s ergonomics