{"id":17020,"date":"2026-01-15T21:20:00","date_gmt":"2026-01-15T21:20:00","guid":{"rendered":"https:\/\/yohrt.com\/news\/meta-outlines-how-its-improved-reels-recommendations\/"},"modified":"2026-01-15T21:20:00","modified_gmt":"2026-01-15T21:20:00","slug":"meta-outlines-how-its-improved-reels-recommendations","status":"publish","type":"post","link":"https:\/\/yohrt.com\/news\/meta-outlines-how-its-improved-reels-recommendations\/","title":{"rendered":"#\n                Meta Outlines How It\u2019s Improved Reels Recommendations"},"content":{"rendered":"<div id=\"ez-toc-container\" class=\"ez-toc-v2_0_84 counter-hierarchy ez-toc-counter ez-toc-custom ez-toc-container-direction\">\n<div class=\"ez-toc-title-container\">\n<p class=\"ez-toc-title\" style=\"cursor:inherit\">Table of Contents<\/p>\n<span class=\"ez-toc-title-toggle\"><a href=\"#\" class=\"ez-toc-pull-right ez-toc-btn ez-toc-btn-xs ez-toc-btn-default ez-toc-toggle\" aria-label=\"Toggle Table of Content\"><span class=\"ez-toc-js-icon-con\"><span class=\"\"><span class=\"eztoc-hide\" style=\"display:none;\">Toggle<\/span><span class=\"ez-toc-icon-toggle-span\"><svg style=\"fill: #dd3333;color:#dd3333\" xmlns=\"http:\/\/www.w3.org\/2000\/svg\" class=\"list-377408\" width=\"20px\" height=\"20px\" viewBox=\"0 0 24 24\" fill=\"none\"><path d=\"M6 6H4v2h2V6zm14 0H8v2h12V6zM4 11h2v2H4v-2zm16 0H8v2h12v-2zM4 16h2v2H4v-2zm16 0H8v2h12v-2z\" fill=\"currentColor\"><\/path><\/svg><svg style=\"fill: #dd3333;color:#dd3333\" class=\"arrow-unsorted-368013\" xmlns=\"http:\/\/www.w3.org\/2000\/svg\" width=\"10px\" height=\"10px\" viewBox=\"0 0 24 24\" version=\"1.2\" baseProfile=\"tiny\"><path d=\"M18.2 9.3l-6.2-6.3-6.2 6.3c-.2.2-.3.4-.3.7s.1.5.3.7c.2.2.4.3.7.3h11c.3 0 .5-.1.7-.3.2-.2.3-.5.3-.7s-.1-.5-.3-.7zM5.8 14.7l6.2 6.3 6.2-6.3c.2-.2.3-.5.3-.7s-.1-.5-.3-.7c-.2-.2-.4-.3-.7-.3h-11c-.3 0-.5.1-.7.3-.2.2-.3.5-.3.7s.1.5.3.7z\"\/><\/svg><\/span><\/span><\/span><\/a><\/span><\/div>\n<nav><ul class='ez-toc-list ez-toc-list-level-1 eztoc-toggle-hide-by-default' ><li class='ez-toc-page-1 ez-toc-heading-level-1'><a class=\"ez-toc-link ez-toc-heading-1\" href=\"https:\/\/yohrt.com\/news\/meta-outlines-how-its-improved-reels-recommendations\/#Meta_Outlines_How_Its_Improved_Reels_Recommendations\" >Meta Outlines How It\u2019s Improved Reels Recommendations<\/a><\/li><\/ul><\/nav><\/div>\n<h1><span class=\"ez-toc-section\" id=\"Meta_Outlines_How_Its_Improved_Reels_Recommendations\"><\/span>\n                Meta Outlines How It\u2019s Improved Reels Recommendations<br \/>\n            <span class=\"ez-toc-section-end\"><\/span><\/h1>\n<div>\n<div class=\"text-to-speech\">\n    <button class=\"text-to-speech__button button\"><br \/>\n        <img decoding=\"async\" class=\"text-to-speech__button__icon\" src=\"https:\/\/www.socialmediatoday.com\/static\/img\/play.svg?500116090725\" alt=\"\"\/><br \/>\n        Listen to the article<br \/>\n        <span class=\"text-to-speech__button__audio-length\">5 min<\/span><br \/>\n    <\/button><\/p>\n<div class=\"text-to-speech__controls\">\n        <audio controls=\"\" class=\"js-text-to-speech\" preload=\"none\"><source src=\"https:\/\/text-to-speech.divecdn.com\/newspost\/809804\/2026-01-16_10.32.17\/metas-outlines-how-its-improved-its-reels-recommendations.wav\" type=\"audio\/mp3\"><\/source><\/audio><\/p>\n<div class=\"text-to-speech__controls__text\">\n            This audio is auto-generated. Please let us know if you have feedback.\n        <\/div>\n<\/p><\/div>\n<\/div>\n<p><span><span><span><span><span><span><span>Meta has published a <a rel=\"nofollow\" target=\"_blank\" href=\"https:\/\/engineering.fb.com\/2026\/01\/14\/ml-applications\/adapting-the-facebook-reels-recsys-ai-model-based-on-user-feedback\/\" style=\"color:#0563c1\">new overview<\/a> of how it\u2019s working to improve your Reels recommendations, by using user response surveys to better gauge which elements are driving interest and engagement. <\/span><\/span><\/span><\/span><\/span><\/span><\/span><\/p>\n<div align=\"center\"><img decoding=\"async\" alt=\"Reels feedback\" data-imagemodel=\"192083\" src=\"https:\/\/imgproxy.divecdn.com\/a-3QGxPQp_E6Kfum9LNvkO11jmanGX6lgkOo7AQ3Vqo\/g:ce\/rs:fit:1600:0\/Z3M6Ly9kaXZlc2l0ZS1zdG9yYWdlL2RpdmVpbWFnZS9yZWVsc19mZWVkYmFjay5wbmc=.webp\"\/><\/div>\n<p><span><span><span><span><span><span><span>No doubt you\u2019ve seen these yourself within the Reels feed, prompts that are shown in-between videos that ask you how you felt about the Reel that you just watched. Meta says that it\u2019s deployed this approach on a large scale, and based on the feedback provided, it\u2019s gleaned more info to help refine and improve its Reels recommendations.<\/span><\/span><\/span><\/span><\/span><\/span><\/span><\/p>\n<p><span><span><span><span><span><span><span>As explained by <a rel=\"nofollow\" target=\"_blank\" href=\"https:\/\/engineering.fb.com\/2026\/01\/14\/ml-applications\/adapting-the-facebook-reels-recsys-ai-model-based-on-user-feedback\/\" style=\"color:#0563c1\">Meta<\/a>:<\/span><\/span><\/span><\/span><\/span><\/span><\/span><\/p>\n<p><span><span><span><em><span><span><span><span>\u201c<\/span><\/span><\/span><\/span><\/em><em><span><span><span><span>By weighting responses to correct for sampling and nonresponse bias, we built a comprehensive dataset that accurately reflects real user preferences \u2013 moving beyond implicit engagement signals to leverage direct, real-time user feedback.\u201d<\/span><\/span><\/span><\/span><\/em><\/span><\/span><\/span><\/p>\n<p><span><span><span><span><span><span><span>So rather than just using likes, shares and watch-time as indicators of interest, Meta\u2019s looking to expand beyond this, and consider more elements that can further improve its recommendations. <\/span><\/span><\/span><\/span><\/span><\/span><\/span><\/p>\n<p><span><span><span><span><span><span><span>And apparently it\u2019s working.<\/span><\/span><\/span><\/span><\/span><\/span><\/span><\/p>\n<p><span><span><span><span><span><span><span>According to Meta, before it deployed these surveys, its recommendation systems were only achieving a 48.3% alignment with true user interests. But now, following the implementation of learnings based on these surveys, that\u2019s increased to more than 70%.<\/span><\/span><\/span><\/span><\/span><\/span><\/span><\/p>\n<p><span><span><span><em><span><span><span><span>\u201c<\/span><\/span><\/span><\/span><\/em><em><span><span><span><span>By integrating survey-based measurement with machine learning, we are creating a more engaging and personalized experience \u2013 delivering content on Facebook Reels that feels truly tailored to each user and encourages repeat visits. While survey-driven modeling has already improved our recommendations, there remain important opportunities for improvement, such as better serving users with sparse engagement histories, reducing bias in survey sampling and delivery, further personalizing recommendations for diverse user cohorts and improving the diversity of recommendations.\u201d<\/span><\/span><\/span><\/span><\/em><\/span><\/span><\/span><\/p>\n<p><span><span><span><span><span><span><span>This approach isn\u2019t new, with Pinterest, for example, detailing how it\u2019s used <a rel=\"nofollow\" target=\"_blank\" href=\"https:\/\/medium.com\/pinterest-engineering\/improving-quality-of-recommended-content-through-pinner-surveys-eebca8a52652\" style=\"color:#0563c1\">similar surveys<\/a> to gather feedback to improve its recommendation systems. <\/span><\/span><\/span><\/span><\/span><\/span><\/span><\/p>\n<p><span><span><span><span><span><span><span>But the rate of improvement is impressive, and it\u2019ll be interesting to see whether this does lead to a significant improvement in relevance for your Reels suggestions.<\/span><\/span><\/span><\/span><\/span><\/span><\/span><\/p>\n<p><span><span><span><span><span><span><span>Though, really, Meta\u2019s still trailing TikTok in this respect. <\/span><\/span><\/span><\/span><\/span><\/span><\/span><\/p>\n<p><span><span><span><span><span><span><span>TikTok\u2019s almighty \u201cFor You\u201d feed algorithm remains the benchmark for compulsive engagement, keeping users scrolling through the app for hours and hours on end. <\/span><\/span><\/span><\/span><\/span><\/span><\/span><\/p>\n<p><span><span><span><span><span><span><span>So what does TikTok\u2019s algorithm have that Meta\u2019s doesn\u2019t?<\/span><\/span><\/span><\/span><\/span><\/span><\/span><\/p>\n<p><span><span><span><span><span><span><span>Primarily, TikTok seems to have developed a better system for entity recognition within clips, which gives the TikTok system more data to go on in matching up your preferences.<\/span><\/span><\/span><\/span><\/span><\/span><\/span><\/p>\n<p><span><span><span><span><span><span><span>Yet, TikTok is also very secretive about how the algorithm works, and won\u2019t reveal much about this particular element, though we do know that TikTok\u2019s system can identify very specific visual elements within clips. <\/span><\/span><\/span><\/span><\/span><\/span><\/span><\/p>\n<p><span><span><span><span><span><span><span>Back in 2019, <a rel=\"nofollow\" target=\"_blank\" href=\"https:\/\/theintercept.com\/2020\/03\/16\/tiktok-app-moderators-users-discrimination\/\" style=\"color:#0563c1\">The Intercept<\/a> came across a set of guiding principles for TikTok moderators, which included a range of very specific instructions for dealing with certain visual cues.<\/span><\/span><\/span><\/span><\/span><\/span><\/span><\/p>\n<p><span><span><span><span><span><span><span>As per <a rel=\"nofollow\" target=\"_blank\" href=\"https:\/\/theintercept.com\/2020\/03\/16\/tiktok-app-moderators-users-discrimination\/\" style=\"color:#0563c1\">The Intercept<\/a>:<\/span><\/span><\/span><\/span><\/span><\/span><\/span><\/p>\n<p><span><span><span><em><span><span><span><span>\u201c[TikTok] <\/span><\/span><\/span><\/span><\/em><em><span><span><span><span>instructed moderators to suppress posts created by users deemed too ugly, poor, or disabled for the platform [as well as] videos showing rural poverty, slums, beer bellies, and crooked smiles. One document goes so far as to instruct moderators to scan uploads for cracked walls and \u2018disreputable decorations\u2019 in users\u2019 own homes.\u201d<\/span><\/span><\/span><\/span><\/em><\/span><\/span><\/span><\/p>\n<p><span><span><span><span><span><span><span>These guidelines were intended to maximize the aspirational nature of the platform, which would then drive more growth. TikTok admitted that such parameters did, at one time, exist, but it also clarified that these specific qualifiers were never enacted in TikTok itself, with the parameters copied from an earlier document intended only for Douyin, the Chinese version. <\/span><\/span><\/span><\/span><\/span><\/span><\/span><\/p>\n<p><span><span><span><span><span><span><span>Though their very existence suggests that TikTok can systematically detect these elements. I mean, you could assume that TikTok\u2019s moderators were looking to manage this manually, and reject videos including these elements based on human detection. But at the platform\u2019s scale (both TikTok and Douyin have hundreds of millions of users) would make this an impossible task, which would render these notes utterly useless. Unless the system could detect such through computer vision.<\/span><\/span><\/span><\/span><\/span><\/span><\/span><\/p>\n<p><span><span><span><span><span><span><span>That\u2019s where TikTok really wins out, in that it can understand a lot more about what you\u2019re looking at, then factor that into your recommendations. So if you spend time looking at a video of a blonde-haired man with blue eyes, you can bet that you\u2019re going to see more content from similar looking creators. <\/span><\/span><\/span><\/span><\/span><\/span><\/span><\/p>\n<p><span><span><span><span><span><span><span>Expand that to any number of physical traits and background elements and you can see how TikTok is better able to align with your specific preferences. <\/span><\/span><\/span><\/span><\/span><\/span><\/span><\/p>\n<p><span><span><span><span><span><span><span>So while TikTok also uses the more common matching, in terms of likes, watch time, etc., it\u2019s also working to keep users glued to their phones by aligning with their more primal leanings. And if the true depth of that process were ever made public, TikTok would likely come under intense scrutiny, because it\u2019s using psychological bias and leanings to compel its users, based, potentially, on problematic and even harmful traits.<\/span><\/span><\/span><\/span><\/span><\/span><\/span><\/p>\n<p><span><span><span><span><span><span><span>That\u2019s where Meta\u2019s losing out, because it can\u2019t implement the same depth of understanding to improve its systems. Theoretically, it could use more psychographic measures, based on user history on Facebook, and with older users who\u2019ve uploaded more of their personal data to the app, that might be effective. But mostly, Meta is relying on more common algorithm signals, and now user surveys, to improve the Reels feed.<\/span><\/span><\/span><\/span><\/span><\/span><\/span><\/p>\n<p><span><span><span><span><span><span><span>Are your recommendations looking better of late? This could be why, while it should also mean that your content is being shown to more engaged audiences. \u00a0<\/span><\/span><\/span><\/span><\/span><\/span><\/span><\/p>\n<\/p><\/div>\n<blockquote>\n<p style=\"text-align: center;\"><strong>If you want to read more like this article, you can visit our <span style=\"color: #ff9900;\"><a style=\"color: #ff9900;\" href=\"https:\/\/yohrt.com\/news\/social-media\/\" target=\"_blank\" >Social Media<\/a><\/span> category.<\/strong><\/p>\n<\/blockquote>\n<p><span style=\"color: black;\"><a style=\"color: #ff9900;\" href=\"https:\/\/www.socialmediatoday.com\/news\/metas-outlines-how-its-improved-its-reels-recommendations\/809804\/\" target=\"_blank\" >Source<\/a><\/span><\/p>\n","protected":false},"excerpt":{"rendered":"<p>Meta Outlines How It\u2019s Improved Reels Recommendations Listen to the article 5 min This audio is auto-generated. Please let us know if you have feedback. Meta has published a new overview of how it\u2019s working to improve your Reels recommendations, by using user response surveys to better gauge which elements are driving interest and engagement&#8230;.<\/p>\n","protected":false},"author":1,"featured_media":17021,"comment_status":"open","ping_status":"open","sticky":false,"template":"","format":"standard","meta":{"fifu_image_url":"https:\/\/imgproxy.divecdn.com\/iuNeUhD_NAotL5UPmY6DhSOcda5O0GLVJiYjOl85zms\/g:ce\/rs:fit:770:435\/Z3M6Ly9kaXZlc2l0ZS1zdG9yYWdlL2RpdmVpbWFnZS9yZWVsc19mZWVkYmFjazIucG5n.webp","fifu_image_alt":"","footnotes":""},"categories":[3],"tags":[1424,1448],"class_list":["post-17020","post","type-post","status-publish","format-standard","has-post-thumbnail","hentry","category-social-media","tag-facebook","tag-instagram"],"_links":{"self":[{"href":"https:\/\/yohrt.com\/news\/wp-json\/wp\/v2\/posts\/17020","targetHints":{"allow":["GET"]}}],"collection":[{"href":"https:\/\/yohrt.com\/news\/wp-json\/wp\/v2\/posts"}],"about":[{"href":"https:\/\/yohrt.com\/news\/wp-json\/wp\/v2\/types\/post"}],"author":[{"embeddable":true,"href":"https:\/\/yohrt.com\/news\/wp-json\/wp\/v2\/users\/1"}],"replies":[{"embeddable":true,"href":"https:\/\/yohrt.com\/news\/wp-json\/wp\/v2\/comments?post=17020"}],"version-history":[{"count":0,"href":"https:\/\/yohrt.com\/news\/wp-json\/wp\/v2\/posts\/17020\/revisions"}],"wp:featuredmedia":[{"embeddable":true,"href":"https:\/\/yohrt.com\/news\/wp-json\/wp\/v2\/media\/17021"}],"wp:attachment":[{"href":"https:\/\/yohrt.com\/news\/wp-json\/wp\/v2\/media?parent=17020"}],"wp:term":[{"taxonomy":"category","embeddable":true,"href":"https:\/\/yohrt.com\/news\/wp-json\/wp\/v2\/categories?post=17020"},{"taxonomy":"post_tag","embeddable":true,"href":"https:\/\/yohrt.com\/news\/wp-json\/wp\/v2\/tags?post=17020"}],"curies":[{"name":"wp","href":"https:\/\/api.w.org\/{rel}","templated":true}]}}