I recently ran a Twitter poll asking my followers to estimate how many engineers had been laid off from US-based startups and tech companies in 2022 and 2023 so far.
As you can see, 75% of respondents thought that 50K or more engineers had been laid off, and over a third of respondents thought it was at least 100K. Only about a quarter of respondents guessed 10K or fewer.
As it turns out, most people were off by roughly an order of magnitude – the real number is likely somewhere between 7.5K and 10K… which means that the majority of people are operating on incorrect information when they make important career decisions. For instance, how might your answers change to questions like these if you knew that layoffs aren’t as pervasive as you thought?
How do I act at work? Should I even look for a job; it probably makes sense to hunker down and keep my head down, right?? If I’m in the middle of a job search, should I take the first offer? If I just got an offer, should I even try to negotiate… won’t they rescind it if I ask for more money?
We’ll attack some of these questions, head on, in a future post, but in this post, we’ll just share our estimate and how we got there, and hopefully we’ll dispel some of the panic around eng layoffs and help engineers make decisions based on data rather than fear.
To get to our much lower estimate, we first looked at how many people total were laid off in 2022 and 2023 so far, independent of department. Then we did some analysis to figure out how many of those people were engineers.
To figure out how many people got laid off, we looked at layoffs.fyi. At some point in the last 2.5 years, you’ve probably visited layoffs.fyi. It was launched by Roger Lee in February 2020, just when concerns that this COVID-19 thing might affect the economy went mainstream. The site does exactly what it sounds like – it tracks layoffs at tech companies. Every time a company conducts a public round of layoffs, it gets added to a growing list. Each entry includes a layoff count, and a small subset of entries include a link to a list (usually in a Google doc) of actual people who were let go, as well as some info about them (name, LinkedIn, geography, title, and so on).
The meat of layoffs.fyi is a giant Airtable embedded into the site, which means you can filter the data, like so (I had to go city by city rather than just saying “United States” because many rows had the country mislabeled):
Once you filter on just the U.S., you get 163,296 layoffs total in 2022 and 50,263 layoffs total in 2023 so far, making for a grand total of 213,559.
We saw this estimate roughly corroborated in a recent post by Crunchbase News. They also used layoffs.fyi as a source but ended up with a more conservative estimate of 153,000.
So, TL;DR somewhere between 150K and 200K people got laid off total.
We recently did an analysis, where we looked at a mix of layoffs.fyi and LinkedIn data to figure out how much layoffs affected engineers vs. other departments. As it turned out, about 5% of total layoffs were engineers, as of the publication date of that post (October 2022)1. I expect that this figure has held up over the past few months.
If we put all this info together, we’re looking at something like 7.5K-10K engineers who were affected by layoffs between the start of 2022 and now, which is a far cry from the 50K-100K+ that most people had guessed.2
While our hearts go out to all those affected, especially in cases when the layoff was a function of poor planning by the company rather than anything to do with the engineer’s performance, it’s important to put these numbers in perspective and temper the rising urge to panic with some cold, hard data.
If you’re curious, even though only 5% of total layoffs were engineers, about 10% of engineers were laid off. ↩
A few high-profile companies with layoff lists are likely outliers: Meta and Coinbase. Both of those companies had engineers comprise more than 5% of layoffs, but in both cases, the lists were primarily composed of junior engineers. That doesn’t make it any less painful for the people involved, of course, but from what we saw, eng layoffs were not generally so skewed toward junior engineers. For a really useful analysis of Meta layoffs specifically, please see this Github repo. ↩
Given a string s, find the length of the longest substring without repeating characters.
Given a string s, return the longest palindromic substring in s.
You are given a path to a file as a string. The path can contain the symbols: “..” for the parent directory and “.” for the current directory. Convert the path into its simplified form.
Interview prep and job hunting are chaos and pain. We can help. Really.