Platformer obtained documents obtained from Twitter engineers and interviews with people familiar with the events involved. Musk sent a message to Twitter engineers at 2:36 on Monday morning.
Musk, a cousin of the Twitter CEO, posted in Slack that “we are debugging an engagement issue across the platform” and tagged “@here” so everyone would be notified. Any people who are capable of creating dashboards and writing software are welcome to help solve this issue. This is of high urgency. If you are willing to assist, please thumbs up this post.
The nature of the emergency became clear as engineers logged onto their laptops: Elon Musk’s tweet about the Super Bowl received far fewer engagements than President Biden’s.
Nearly 29 million impressions were generated by Biden’s tweet, in which he said he would root for the Philadelphia Eagles with his wife. Musk, who also tweeted his support for the Eagles, generated approximately 9.1 million impressions before deleting the tweet in apparent frustration.
On Sunday night, Twitter’s CEO flew his private jet back to the Bay Area following those losses – the Eagles to the Kansas City Chiefs and Musk to the president.
Within a day, the consequences of that meeting would reverberate around the world, as Twitter users opened the app and found that Musk’s posts overwhelmed their ranked timeline. Platformer can confirm that it wasn’t an accident: after Musk threatened to fire his remaining engineers, the company built a system to ensure Musk — and Musk alone — received previously unheard-of promotion of his tweets.
The amount of engagement Musk receives on his posts has been a major concern to Musk in recent weeks. Platformer broke the news last week that one of the company’s two remaining principal engineers had been fired after he told the engineer that views on Musk’s tweets were declining because Musk’s popularity was falling. During the weekend, his deputies informed the rest of the engineering team that they would also lose their jobs if the engagement issue was not “fixed.”
Twitter deployed code to automatically “greenlight” all of Musk’s tweets meaning his posts will bypass Twitter’s filters designed to show people the best content possible. The algorithm now artificially boosted Musk’s tweets by a factor of 1,000 – a constant score that ensured his tweets rank higher than anyone else’s in the feed. Internally, this is called a “power user multiplier,” although it only applies to Elon Musk, we’re told.
On Tuesday afternoon, Musk posted a version of the popular “forced to drink milk” meme in which a woman labeled “Elon’s tweets” bottle-feeds another woman labeled “Twitter” and pulls her hair back.
His tweets Monday were sent during a call with Twitter engineers to test whether their solutions were working as he expected.
The artificial boosts applied to his account remain in place, although the factor is now lower than 1,000, we’re told. Musk’s handful of tweets Tuesday reported around 43 million impressions, which are on the high end of his recent average. Absurd as Musk’s antics are, they do highlight a tension familiar to almost anyone who has ever used a social network: why are some posts more popular than others? Why am I seeing this thing, and not that one? Engineers for services like TikTok and Instagram can offer partial, high-level answers to these questions. But ranking algorithms make predictions that can be opaque even to those who design them.
For better and for worse, the answer hasn’t been good enough for Musk. As Twitter’s most prominent user, with nearly 129 million followers, his posts often get 10 million or more impressions, as counted by Twitter. (There are good reasons to doubt the accuracy of these counts, but better data is not readily available.) But Musk’s view counts still fluctuate widely. The bottle-feeding tweet got a reported 118.4 million impressions; his next one, a joke observation previously posted to Reddit and satirically attributed to Abraham Lincoln, got 49.9 million. Some of his tweets from earlier this year (such as one about smoking) have had far fewer views than some of his more recent tweets (such as one about colonizing Mars).
This discrepancy is primarily due to people’s perceptions that some tweets are better than others. However, you don’t have to do that: you could change the ranking algorithms so that they always show your posts. This is what Twitter engineers are now building in fear of losing their jobs.