The clearest lessons from failed angel investments come from public cases and base rates. CB Insights found most recent venture-backed shutdowns ran out of capital, but poor product-market fit was the deeper cause in many; cases like Theranos, Juicero and Fast add three specific checks: verify claims, test the product, and compare money raised with revenue.
Every angel portfolio contains failures, and the useful question is what the failures have in common. This page answers it with documented public cases and published data, not anyone's private portfolio. Most of the companies below were backed by venture funds rather than angels alone, but the mistakes they illustrate are the same ones angels make with smaller checks.
How do startups fail, according to the data?
CB Insights' March 2026 report looked at 431 venture-backed startups that shut down since 2023. "Ran out of capital" topped its list at 70 percent, but the report says that is almost always the final cause of death, not the root problem. The deeper causes it counts are poor product-market fit (43 percent), bad timing (29 percent) and unsustainable unit economics (19 percent), and many companies cited more than one. The median time from last fundraise to shutdown was 22 months.
For an angel, the base rate is the starting point. Rob Wiltbank, who has studied angel returns for years, wrote in a 2012 TechCrunch column that an angel is more likely than not to lose money on any single investment, and that 90 percent of cash returns came from the top 10 percent of exits. Losing on a deal is the normal outcome. The skill is in not losing for avoidable reasons.
What can Theranos teach about checking claims?
On January 4, 2022, the US Attorney's Office for the Northern District of California announced that a jury had found Theranos founder Elizabeth Holmes guilty of one count of conspiracy and three counts of wire fraud in a scheme to defraud investors. According to the release, the evidence showed she made materially false statements to investors about the company's blood-testing analyzer and its finances, and gave investors reports carrying the logos of pharmaceutical companies in a way that falsely suggested they endorsed Theranos.
Fraud is rare, and a fraud case is not a typical failure. The checks it points to are still cheap:
Ask for independent validation of any technical claim, and read it yourself or have an expert read it.
Treat logos and names as claims, not evidence. Ask the named company whether the relationship exists.
Compare stated revenue with what customers say. Speak to a customer who is not on the founder's reference list.
What can Juicero teach about testing the product?
Juicero sold a connected juice press that cost hundreds of dollars and had raised about $120 million. In April 2017, Bloomberg reported that two of its investors were surprised to learn that its juice packs could be squeezed by hand without the machine. Months later, a New York Times story syndicated by Gulf News reported that the company was shutting down, after cutting its price and laying out plans to find a buyer with an existing national fresh-food supply chain.
The lesson for an angel is to use the product and then ask what the cheapest alternative is. Buy it, run the core workflow and see whether the promised benefit survives contact with something simple. A product that needs an elaborate machine, a proprietary supply chain or special hardware has costs you should price before you invest.
What can Fast teach about revenue and valuation?
Fast, a one-click checkout startup founded in 2019, shut down in April 2022. Global Venturing reported that its 2021 revenue came to approximately $600,000. By then, according to Payments Dive, it had raised $124 million, including a $20 million round led by Stripe in March 2020 and a $102 million round in January 2021. Global Venturing, citing The Information, said Fast had tried to raise new funding at a $1 billion valuation in 2022 without success, and Payments Dive reported The Information's account that it was exploring a sale after attempts to raise more capital failed. Chief executive Domm Holland announced the closure, saying the company had made "the difficult decision to close our doors."
Early investors in a company like this may never see the later numbers unless they ask. The lesson is to ask for revenue and burn at every update, and to treat a rising valuation as the market's opinion, not proof the business works. The guide to evaluating burn rate and runway shows what to track.
What should an angel do differently?
These steps follow from the cases and from the research on angel returns:
Spend real time on diligence. In the 2007 Kauffman-funded study of angel-group members, summarized by the Angel Capital Education Foundation, deals with high diligence returned 5.9x against 1.1x for low diligence, with a median of 20 hours. That is an association, not proof, but it points the right way. Use the early-stage diligence guide.
Test the core promise yourself before you wire money.
Track burn against revenue and ask what happens at the next 12 months of cash.
Hold enough positions. Wiltbank found median returns above 1x only once an investor had at least six investments and suggested a dozen or more. The portfolio-building guide covers pacing.
Write a pre-mortem and a post-mortem for every deal. Note why you invested, what would change your mind and, after a shutdown, what you missed.
Know your passes. The red flags for passing on a deal are easier to apply before the check than after.
Separating bad luck from bad process is covered in the luck versus skill analysis. More guides are in the angel investing section.
Frequently Asked Questions
Q:What percentage of angel investments fail?
It depends on the dataset and on how failure is defined. Wiltbank wrote in 2012 that an angel is more likely than not to lose money on any single investment, and that a few exits produce most of the cash.
Q:Why do most startups fail?
CB Insights' 2026 analysis of 431 shutdowns found that running out of capital was the final cause in 70 percent of cases, with poor product-market fit, bad timing and weak unit economics behind it. The report reads the cash shortfall as a symptom of those deeper problems.
Q:Did angels lose money on Theranos, Juicero and Fast?
The sources describe investors in general or named venture firms, and they don't break out angels. These are cautionary cases about diligence, not records of angel losses.
Q:Should I keep a log of my failed investments?
Yes. A short post-mortem for each failure, covering what you believed, what turned out false and what you'd check next time, is the cheapest way to improve your process. Review the log before you write your next check.