More Bootcamp Layoffs and Program Cancellations

Today, I want to cover another dismal month for coding bootcamps and, as a former founder who left the space, give my observations on how they got here and where things are heading.

Turing School Layoffs

First, we have Turing School, which let 14 staff members go without notice. This drops their staff from 52 people at their peak to now 22 people. Once again, the staff were released without notice, and students working with those teachers were left wondering what would come next.

App Academy Layoffs

Next, we have App Academy, one of the more well-known brands in the bootcamp space cutting 31% of their staff. In the blog post, the CEO, Kush Patel, also claims they will have fewer cohorts and leverage "internal and external software, AI, and external contractors". This, naturally, is about driving down costs.

Codeup Layoffs

A few weeks after moving into its new physical location, Codeup had a "roughly 25% reduction in force layoff".

Codesmith Put on Blast by Former Student

​Codesmith threads on Reddit are very strange. As others point out in the comments, many posts that criticize or critique the program get systemically downvoted, and there are quite a few posts that look like paid accounts.

The common thread among criticism posts is that they use an open-source learning project to pretend to be work experience to fool recruiters. If true, that's a ticking time bomb that will devalue every current and former student's experience if it goes off.

The Financial Problems with Coding Bootcamps

Warning! Math ahead...

I spent several years (2013-2017) in the bootcamp space and then several more years delivering training and curriculum to B2B clients, including enterprises, universities, and the occasional coding bootcamp.

It wasn't until after we sold the bootcamp to a larger, for-profit education company that things started to turn for the worse. Immediately, a layer of management, marketing, sales, etc., was layered on top of the program, which dramatically increased the costs of running the program. Most education companies follow the same playbook, and as the space gets more crowded, they have to dump more money into marketing.

The acquisition cost (ac) calculates what it costs an education offering to get a paying student, and many bootcamps have an AC of $2,000-4,000. Let's set this at $3,000.

Then, you have the instructional costs. A bootcamp generally hires experienced developers, alumni as TAs, or both. Regardless, you'll see around a 1:7 to 1:20 instructor-to-student ratio in most places. The inexperienced TAs generally make about $50k/year, and experienced developers range from $90k to $150k/year. To simplify things, let's set instruction costs at $25k/month for 30 students.

Next, you have the overhead costs. You have all these admissions people, marketing people, executive team members, physical space, hosting, software tools, computer hardware, etc. In a typical bootcamp that runs around 100 students per year, this is at least $600k/year or $50k/month. (a lot more if you're in San Francisco or NYC)

These numbers are broad, general, and oversimplified, but they are useful in proving the point. Let's compare two cohorts, one with 30 and one with 10 students, and set tuition at $15,000, the average for bootcamps nationwide. The broad numbers look like this:

As you can see, if bootcamps aren't filling their cohorts, they lose money, and many costs aren't easily reduced. Even when cohorts are full, you need to stack up cohorts with monthly rolling starts to turn a profit that will satisfy investors. And therein lies the problem:

• There's almost no budget available for curriculum and process improvements​
  This is why some reviews complain about being sent to public sites, outdated content, or having lectures consisting of TAs reading slides.
• Instruction is the Easiest Place to Make Cuts​
  Hence the layoffs. Instruction is the largest variable cost in the business outside of executive overhead, and we all know executives aren't keen on cutting their pay or benefits. This is why we've seen bootcamps making large staff cuts. They are making a bet that students don't need as much support.
• They've Been Betting on "Easy" and Cheap Skills
  ​There's a reason why most bootcamps teach the front-end JavaScript / MERN stack. It's one of the easiest to get started in. You don't have to be a strong programmer because it's mostly stringing frameworks together, and when there was high demand, employers were willing to take people with less depth.

So, there are only two ways out of this situation for the bootcamps:

1. Pivot into More Depth and Quality

They can pivot and fundamentally change their model to pursue higher quality education and produce students with more depth of expertise. If done right, this would set their students apart and ahead of bootcamp grads and CS majors.

However, this is a non-starter for their investors because it would require investing in a new curriculum, which takes time and money, the cohort lengths would have to be longer, and their enrollment processes would have to be more selective. This would increase costs because, as they are currently structured, they are not efficient organizations.

2. Try to Increase Margins and Weather the Storm

This is the path most of the camps are trying to take, and it's a loser. They are cutting instruction, not investing in new products or curricula, and maintaining the current model, which involves pushing learners through as quickly as possible with as few resources as possible to maximize their margins.

This is what we've seen for most of 2023, and as this trend continues, I predict we will see more camps shutting down, more student complaints about the quality of the learning experience, and more tarnish on the concept of bootcamps. The future of many of these programs looks like what happened to ITT Tech.

It's not all Doom and Gloom

Things are rough for bootcamps and will probably continue until Q2/Q3 of 2024. I am seeing some indications of recovery in some sectors.

I encourage learners to slow down, take their time, and pick something other than the MERN stack to go deep on. Enterprise development jobs generally are the least impacted and quickest to recover.

Still, those established non-FAANG companies have stacks that mostly consist of tried-and-true enterprise languages like Java, .NET, and SQL. Most of these enterprises run about 2-3 years behind the latest ideas and frameworks, but the core of professional application development evolves slowly and methodically. So stay true to the fundamentals, and you'll become a stronger candidate than someone who rushes through a 12-week MERN stack and becomes a "framework developer" - not a software engineer.