25 theses about Elon Musk's first three weeks at Twitter

Musk's leadership has been even more erratic than I expected.

It’s time to do a paid-subscriber-only Q&A post! Please leave questions for me in the comments to this post, and I’ll answer some of them in a future post. Thank you for supporting the newsletter!

Twitter has been in constant turmoil over the last three weeks, as Elon Musk has announced a flurry of layoffs and new products—then quickly suspended some new products that didn’t work out as intended. Rather than write a conventional article with a clear thesis, I thought it might be better to offer some loosely-connected thoughts about how things have gone so far.

1. Let’s get this out of the way: so far, Elon Musk seems to be doing a worse job running Twitter than I expected. I expected things to be chaotic, but I didn’t expect them to get this chaotic this fast.

2. I don’t like the way Musk has treated Twitter’s employees. Layoffs are always painful, but Musk seems to be going out of his way to do it in a haphazard and disrespectful way. This must be an immensely stressful time for Twitter employees—especially those who are supporting families or have immigration visas tied to their jobs.

3. It would be nice to say that Musk’s efforts are doomed to failure because this kind of callous management style never works. And if I had to guess, I’d say it probably isn’t going to work. But I’m not as confident as a lot of his critics.

4. Musk is clearly trying to make Twitter more like those at SpaceX and Tesla, where employees routinely work punishingly long hours. You might think that was unsustainable, but each of these companies has been doing incredible work for close to 20 years now.

5. We’ve known for almost 50 years that programming productivity doesn’t scale linearly with the number of programmers. A major reason is that programmers have to coordinate with one another, and the cost of this grows more-than-linearly with the number of programmers.

6. So I don’t think it’s that crazy to think that a 2,000-employee Twitter with a SpaceX-like work culture could be more productive than a 7,500-employee Twitter with a conventional big company culture.

7. The big question is whether Musk can convince enough people to sign up for this. Reporting so far suggests that a ton of engineers are likely to quit beyond the 3,000-some people Musk fired earlier this month.

8. SpaceX and Tesla are dream jobs for a lot of engineers. Lots of kids grow up dreaming to build rockets, and Tesla is plausibly making a big difference in combatting climate change. It’s not clear that Musk’s Twitter will have that same kind of draw.

9. At the same time, working for one of the world’s leading social media companies will always be more prestigious than the average programming job. I also think liberals might be underestimating the number of engineers who like Elon Musk, broadly share his politics, and might like the idea of working at a tech company that’s not on the woke bandwagon.

10. There may also be some young, ambitious engineers who see this as a career opportunity. Lots of people in their 20s put in long, punishing hours at law firms or investment banks that most people have never heard of. With a lot of senior people quitting, some junior people might be excited at the opportunity to move up through the ranks quickly.

11. On the other hand, there’s a big difference between building a workaholic company from the start—as Musk did with SpaceX and Tesla—and trying to force a big change on an existing workforce. Firing 50 to 75 percent of workers is bound to create morale problems that could deep schisms inside Twitter’s workforce and make it hard for Twitter to recruit new engineers for years to come.

12. In other areas, I think Musk’s tolerance for risk might be an under-appreciated advantage. For example, earlier this week Musk announced he was turning off a bunch of microservices because he claimed most of them weren’t needed for Twitter to work.

13. This was widely mocked online, especially after Twitter’s two-factor authentication stopped working, making it impossible for many people to log into the site. But the problems seem to have been fixed within a few hours, with no lasting damage to Twitter.

14. Unlike a conventional CEO, Musk doesn’t have to worry that his board will fire him if he screws up. So he can try a lot of stuff and then quickly reverse decisions that don’t work out. This allows him to learn rapidly.

15. This might also be the right way to think about last week’s pay-for-verification fiasco. A bunch of people paid $8 for verified accounts purporting to be famous brands and tweeted out funny parody tweets, forcing Twitter to stop selling blue checkmarks.

16. While this has been widely portrayed as a huge blunder, it’s not obvious that this did any permanent damage to Twitter. Obviously it’s not going to help Twitter sell ads over the next month or two. But if the site is working well a year from now, I have to imagine most of these brands will come back.

17. If anything, those parody tweets produced a round of funny stories that kept Twitter in the news.

18. This basic approach—iterating rapidly without worrying too much about failures—has been a hallmark of Musk’s leadership at SpaceX. For a while in late 2020 and early 2021, SpaceX was launching a prototype of its new Starship rocket almost every month. Most of those launches failed, but SpaceX engineers learned a lot from the failures. Maybe the same approach will work for Twitter.

19. A lot of people have mocked Musk’s idea of charging people $8 per month to get a blue “verified” checkmark. And I agree that it’s silly to expect people to pay for a blue checkmark solely as a status symbol.

20. But Musk has made it clear that that’s not all he’s doing. Musk has said that verified users will have priority in replies, mentions, and search, which will provide another incentive for people to pay.

21. In the long run, the way things might work is that anyone can sign up for a free account if they just want to read other people’s tweets and occasionally reply to tweets by their friends. But anyone who wants to build up popular Twitter profile will need to pay $8 per month to become verified, because those without a blue checkmark will be almost invisible to people who don’t already follow them.

22. Obviously, Musk isn’t going to get literally all of Twitter’s users to pay. But if he can get, say, 10 percent of users to pay, that would represent $2 billion in additional annual revenue.

23. Josh Barro has an interesting (paywalled) post suggesting that Musk could monetize Twitter by turning itself into an OnlyFans competitor. In Josh’s telling, the blue checkmark would function not only as identity verification but also age verification.

24. It can sometimes be useful to have polite fictions about what people use a product for. If Twitter formally announced a porn subscription service, many men might be reluctant to sign up for fear of angering their wives. It would also cause friction with Apple’s app store gatekeepers.

25. But if you charge people for blue checkmarks and then later announce that you’re protecting children by limiting porn to verified adults, that has the same practical effect while not ruffling any feathers.

It’s time to do a paid-subscriber-only Q&A post! Please leave questions for me in the comments to this post, and I’ll answer some of them in a future post. Thank you for supporting the newsletter!