AI-powered
podcast player
Listen to all your favourite podcasts with AI-powered features
Facebook is facing a significant challenge from TikTok, as the latter is capturing the attention of young people. The issue for Facebook is that its social graph, which is built around friends and family, is less engaging than the algorithmic video recommendations offered by TikTok.
Tom Allison, head of Facebook at Meta, discusses his efforts to make Facebook and Instagram feeds more like TikTok. This involves showing more content from accounts users don't follow, a departure from the previous focus on showing content from people users know. The aim is to prioritize content that captivates users and fosters meaningful social interactions.
Facebook's feed has undergone significant changes over the years. The introduction of the meaningful social interactions (MSI) approach shifted the focus towards content that fosters meaningful connections. However, with the rise of short-form video and changing user preferences, Facebook is now incorporating more recommendations and content from creators. Efforts are also being made to improve content quality and combat abuse within the recommendations.
The podcast episode discusses the efforts made by a team of AI engineers at a company to enhance the quality of recommendations. The team examines both the content and user preferences to make recommendations more personalized. The goal is to make the recommendations as good or even better than the connected content in the feed. They aim to increase social activity on the platform, such as friend posting and commenting, as a measure of success. However, they recognize that there are multiple goals to consider, including user engagement, revenue growth, experience quality, and trust and safety.
The podcast delves into the different eras of the company's News Feed. The transition from web to mobile was a significant milestone and required a massive engineering effort. The company also had to adapt its design approach to create a seamless mobile product. Later, they focused on integrating various formats and content types into the feed, including video and groups. The shift to stories was initially met with skepticism but eventually gained traction. The current phase involves format innovation, such as short-form video, and wiring new ways for users to express themselves. The podcast also mentions the increasing role of AI in recommendation systems and highlights how Facebook aims to strike a balance between AI-driven recommendations and social interactions.
We’ve got a special episode of Decoder today – an interview between Verge deputy editor Alex Heath and Meta’s Tom Alison, the head of Facebook. Alex is the co-host of the newest season of Vox Media’s podcast Land of the Giants. This season is about Facebook and Meta. The season finale comes out tomorrow.
Alex has been reporting for Land of the Giants for many months, and along the way he interviewed Tom. Facebook has a lot of challenges, but it seems like the biggest problem is TikTok: Facebook's problem is that it spent years – you spent years – building out a social graph that, it turns out, is less interesting than just being shown content that the company thinks you might like. Alison has been at Facebook for more than a decade and previously ran engineering for the News Feed, so he knows more than almost anyone about the history of feeds and where they are going.
Links:
Facebook is changing its algorithm to take on TikTok, leaked memo reveals
Facebook is revamping its home feed to feel more like TikTok
Transcript:
https://www.theverge.com/e/23092319
Credits:
Decoder is a production of The Verge, and part of the Vox Media Podcast Network.
Today’s episode was produced by Creighton DeSimone and Jackie McDermott and it was edited by Callie Wright.
The Decoder music is by Breakmaster Cylinder. Our Sr Audio Director is Andrew Marino. Our Editorial Director is Brooke Minters. And our Executive Producer is Eleanor Donovan.
Learn more about your ad choices. Visit podcastchoices.com/adchoices
Listen to all your favourite podcasts with AI-powered features
Listen to the best highlights from the podcasts you love and dive into the full episode
Hear something you like? Tap your headphones to save it with AI-generated key takeaways
Send highlights to Twitter, WhatsApp or export them to Notion, Readwise & more
Listen to all your favourite podcasts with AI-powered features
Listen to the best highlights from the podcasts you love and dive into the full episode