When Henry Rosales joined Lambda School, he thought it would be a way out of working a low-wage call-center job in Las Vegas. The school's inspirational Instagram ads promised to take people like him with no technical experience and train them for lucrative professions like UX design or web development in nine to 18 months.
After a little over a year of working part time and taking online classes at the school, Rosales felt duped. The school had failed to provide the type of instruction that would have allowed him to make a career as a UX designer, he said. "By the end, I just stopped, because it was a waste of my time," he told me.
Rosales is not alone. Leaked documents from company all-hands meetings in the summer of 2020 and January and February of this year, led by the school's now former chief operating officer, Molly Graham, who resigned earlier this month
, and others led by its chief business officer, Matt Wyndowe, showed that Lambda School placed only 30% of its 2020 graduates in qualifying jobs during the first half of 2020. This figure is in stark contrast to the 74% placement rate it advertised for its 2019 graduates, the latest figure the school has made publicly available.
In a tweet, Graham wrote that her mission was to "get the company through a pivotal phase" and position it to "operate well without me." These documents, given to Insider by a person familiar with the meetings, alongside over a dozen interviews with former Lambda School students and instructors, suggest that Graham is leaving with that mission far from accomplished.
Lambda School's internal placement rates that it shares with investors are different from what it advertises publicly
Cofounded in 2017 by the tech entrepreneurs Austen Allred and Ben Nelson, with help from the startup accelerator Y Combinator, Lambda School offered a nontraditional path for those seeking careers in computer science. In lieu of a four-year degree, students could take a crash course in programming while paying no tuition up front; an income-share agreement allowed students to pay the school a portion of their salary after being hired in a tech job with an annual salary of at least $50,000. Blog posts advertised it as
With the global edtech industry worth more than $106 billion
as of this year, schools have popped up across North America promising to teach students using a similar business model. Lambda School itself has raised a total of $122 million from venture capital.
Lambda School enrolls thousands of students a year and has indicated it plans on growing many times over to give investors profitable returns on the investments they've made.
"I don't see any reason Lambda School shouldn't scale up to tens of millions of people per year," Allred recently said in an interview on "The Quest Pod with Justin Kan."
But the school also has a history of advertising questionable figures. Last year, I reported for New York magazine
that Lambda School's placement rate was closer to 50% in 2019 while it publicly advertised a rate of over 80%
When asked for comment on the school's 30% placement figure presented at the all-hands meeting, Allred redirected Insider to its outcomes report for the second half of 2019, which showed a placement rate of 74%. But to calculate that rate, it excluded about half of the 516 graduates in this report for reasons including graduating late, being unresponsive, or indicating they were no longer seeking a job related to their studies.
While the school's internal documents and those it shares with investors consider only graduates who make $50,000 annually — which roughly mirrors the national average for entry-level coding salaries — as "qualified placements," its advertised placement figures include graduates who make as little as $10,000 a year, less than what the vast majority of students made before they enrolled in the school.
Allred defended the school's methodology as common in the boot-camp industry and said it was audited by an independent group. But the school's private auditors only confirm the results of the report, rather than question the process used to obtain them.
When pressed, Allred blamed the low placement figures on the pandemic. A spokesperson for the Bureau of Labor Statistics told Insider that while jobs in tech suffered some during the first months of the pandemic, the industry was comparatively resilient and among the fastest to recover. It also has more jobs now than it did when the pandemic began, the spokesperson said.
Allred didn't respond to Insider's request for placement figures for the latter half of 2020, though slides suggested initial placement rates from this period were slightly lower than those from the beginning of the year.
If you know that your curriculum isn't good enough, why would you still hold somebody accountable for $30,000? - Henry Rosales
According to Sheree Speakman, the former CEO and now advisor to the Council on Integrity in Results Reporting, a nonprofit formed to track boot-camp results, this type of cherry-picking has become common among big coding schools. The nonprofit has strict guidelines for reporting placement rates, but Speakman told Insider the largest boot camps rarely follow them.
Boot camps that report to the council must ask students whether they are seeking jobs in their field of study before the start of their program, and count all students who respond affirmatively in their job-placement figures. Speakman questioned why so many Lambda School students who'd chosen to enroll in a tech boot camp would report after completing it that they actually weren't seeking employment in tech. While the council does allow the removal of some students in a placement rate — for example, those who graduate late can be removed so long as the school also explicitly discloses the percentage of students who graduate on time and counts late graduates in the next graduating class — Lambda School's methodology diverges from these practices.
"They're shrinking the denominator," Speakman told me.
Lambda School stopped reporting to the nonprofit in 2018 after its first batch of graduates.
Lambda School needs to become profitable. Its efforts may hurt its students.
The all-hands documents also indicate that Lambda School's business model may not need to be as "incentive-aligned" as the school has people believe.
Documents from a meeting last year suggested the school was losing an average of $7,250 per student. Another of Graham's slides from January illustrated the balance the school must strike between enrollment volume and placement rate to achieve profitability. The slide said the school's goal for the first half of 2021 was to reach 50% to 70% placement with 500 students enrolled per month, but it also acknowledged a more uncomfortable possibility: The school could still reach profitability by enrolling 2,000 students a month while placing less than half of its graduates in qualifying jobs.
Lambda School slides from July 2020 showing that the school's total cost per student is $13,000 and that its average revenue per student is $5,750.
Lambda School All Hands Meeting - 07/09/2020
When asked about the slides, Allred denied that the school could be profitable with a low student-success rate and told Insider that some of the figures presented in them were inaccurate, because the school's variable costs would increase according to the number of students it enrolled. Allred declined to share what those variable costs were or explain why slides with ostensibly inaccurate forecasts of the company's paths to profitability were presented at an all-hands meeting.
Lambda School slides showing that its goal is to enroll 500 students a month and place 50% to 70% in qualifying jobs. The graph also shows that Lambda School could become profitable if fewer than half of its students find jobs if it enrolls 2,000 students. "To break even, we need to increase # of students and QP%," the slide says. "Break even happens when we maintain these numbers for 14 months."
Lambda School All Hands Meeting - 01/28/21
A person familiar with the all-hands meetings said the presenters told staff that the school could be sustainable on a per-student basis with a lower student-success rate but it desired better outcomes for its students. Allred denied that they made this statement.
Some students feel there is already inadequate support for those who are struggling in class. "Those students that are in the bottom 10%, why would they invest resources into helping those students succeed? Just let them fail out after six months, let them waste six months of their lives, six months of paying rent," a former web-development student who spoke on the condition of anonymity told me.
Despite its lack of profitability, the school raised over $70 million in Series C funding in August of last year. In the following months it implemented dramatic cost-cutting measures, including shortening its curriculum by three months and eliminating a majority of its paid teaching assistants, or "team leads," who were responsible for helping instructors teach, grade, and review students' code.
"The truth of the matter is that these cost-cutting measures hurt students," said a former Lambda School data-science instructor who spoke to me on the condition of anonymity for fear of retribution from the company. "Because we were always teaching, there was no time to really get the curriculum to where it ought to be."
Allred promised to replace the teaching assistants with more instructors, but in April he laid off a third of the school's staff.
"Class sizes are like 150 students to one instructor — I've only heard that number going up," the web-development student said. The data-science instructor similarly suggested that student-instructor ratios in some programs were around 100 to 1.
Allred told Insider that Lambda School's student-instructor ratio was closer to 13 to 1. This figure included "instructors, teaching assistants, student success coordinators, and other student-facing employees," he said. He declined to provide a number for instructors excluding additional staff.
"Instructor-to-student ratio is the wrong metric for measuring the level of support any student receives," he added.
Lambda School stopped being a school for many students
Tomás Phillips, who enrolled in Lambda's data-science program, complained that the truncated program made it difficult to learn the skills the school had promised to teach.
"It stopped being a school," Phillips said.
"They're very into doublespeak," Phillips added. "They say you'll get hired sooner because you don't have to do those three months."
Last year, the school announced it would terminate the UX-design program and give students the option to either change courses or cancel their income-share agreement. The school's official statement said UX design was no longer a priority, but some students believed the quality of the program was to blame.
Rosales, one of the students in the canceled course, told Insider that he wasn't aware the school had given students the opportunity to change programs or terminate their income-share agreement. When he found out, it was too late — he had missed the deadline to do either. He completed the program, but he told Insider he didn't feel he learned the skills to receive a job in the field.
"If you know that your curriculum isn't good enough, why would you still hold somebody accountable for $30,000?" Rosales said. "It doesn't make any sense."
Another student, Pablo Vahanian, said he joined Lambda School for the hands-on help its teaching assistants provided. When it got rid of the teaching assistants, Vahanian was so incensed that he stopped attending class and negatively reviewed the school with a tech YouTuber. Vahanian said that when he tried to terminate his income-share agreement, the school demanded he take down the interview and pay it $15,000. He ultimately declined.
"There was nothing wrong with reviewing a product," he said. "I purchased a product, and it was faulty."
One student, who requested they remain anonymous, said that learning UX design at Lambda School changed their life and that they're now pursuing a lucrative career in the profession. "I make really good money now," they said. "What I did has changed my life, and I don't want that to go unsaid. I'm a fortunate UX graduate of Lambda School. I'm doing very well."
At the same time, they told Insider, the school didn't provide the educational experience or the career services it had promised. "I have a fortunate enough life situation that I could continue my own education and … fill the gap in education that Lambda failed to give me so that I could get a job," they said.
Allred refused to comment on students' specific accusations against the school, but he said that allegations that the UX-design program was shut down because of its quality were false.
The instructor said that for many students, money was tight. (The school's own 2020 diversity report
said more than a quarter of students cared for children or other dependents while studying.) "These are people often in difficult life positions," he said.
"The focus internally is on telling specific, tear-jerking stories of people whose lives were so profoundly changed by Lambda for the better. Those stories were true — but the question is how often did these true stories occur?"
The school is experimenting with its students — and they aren't happy
This chorus of dissatisfaction has led to a series of legal complaints against the school, including by the National Student Legal Defense Network, a student-protection nonprofit. This May it filed a series of arbitration cases accusing Lambda School of defrauding students by advertising false placement records.
"If I would've known their real job-placement rate — not to mention how hard it was to actually learn at the school — I never would have signed up," said Jonathan Stickrod, one of the students involved in arbitration with the school.
Stickrod, who now works at a café in Medford, Oregon, left community college to attend Lambda School. He said he dropped out after a year because of the school's poor curriculum and cost-cutting measures.
While the students' agreements bar them from bringing a class-action lawsuit, the NSLDN is hoping Strickrod's case and two others will pave the way for a larger case.
John Danner, an early Lambda School investor and board member, told me he was very aware of the complaints against the school.
Danner, who invested $1 million shortly after Lambda School's Series A, compared the complaints to his own experience launching the charter school Rocketship Education, which received public backlash for teaching elementary-school students on laptops without instructors for part of their day.
"They said we were experimenting on the backs of children," Danner said.
"But when SpaceX launched their first five rockets and they blew up, was that OK?" he continued. "We're in a more high-stakes world of human development. Still, you can't say that you don't like the way things are but don't want people to try new things."
For nearly a decade, the boot-camp industry has grown on investments from venture capitalists willing to bet millions on the promise of disrupting traditional education with little oversight and regulation. For VCs it's a given that some of the companies they gamble on will lose money.
But while such trials might be fine for unmanned rockets, experimenting on students has come at a cost.
Rosales is still without a design job and dreads getting one, knowing he'll have to pay Lambda School if he does. The school recently sent him a letter demanding his banking information so that it could track direct deposits from a job. If he doesn't comply, it threatened to charge him his full tuition of $30,000 regardless of whether he gets a job.
Despite this, he's holding out. "I don't think I should be held liable for an education that I didn't receive," he said.