# Theranos: a blood testing promise that did not hold up

Source: https://sarang.pro/theranos/
Author: Sarang | Published: 5 July 2026

## The situation

On 3 January 2022 a federal jury in California found Elizabeth Holmes, the founder and chief executive of the blood-testing company Theranos, guilty of defrauding investors. The jury deliberated for seven days after a trial of about three months with 32 witnesses. It convicted her on four of 11 counts, acquitted her on four and could not decide on the other three. Holmes was 37. She had founded the company at 19, and at its height investors had put her paper fortune at $4.5 billion.

The company's claim was simple and attractive. Its Edison device, it said, could run a wide range of blood tests from a finger prick instead of a needle in the arm, quickly and cheaply, with small laboratories in pharmacy and supermarket stores. The evidence at trial showed something different. The technology gave misleading results, patient samples were secretly run on conventional machines, and Holmes had lied about deals with Pfizer and the US military.

This case follows the company from its confident public story in 2014 to the verdict and sentence, and asks what investors, partners, directors and regulators could have seen earlier. It matters to students of strategy because Theranos had many of the surface features of a winning start-up: a large market, a clear customer benefit, famous directors and heavy funding. What it lacked was a product that worked as described.

## Background

### A founder and a promise

Holmes founded Theranos in 2003 and led it through its 15-year history, according to the Associated Press report carried by PBS. She was indicted on felony charges in 2018. The company's pitch was to change laboratory medicine by taking away the large blood draw and the long wait for results, and to put testing within easy reach of ordinary people.

### The 2014 picture

The clearest snapshot of how the company wanted to be seen comes from a June 2014 Fortune profile. It said Theranos had raised more than $400 million in equity, was valued at more than $9 billion, and had about 500 employees. It operated a 140,000 square foot headquarters in Stanford Research Park and a 262,000 square foot manufacturing plant in Newark, California. Holmes said she held more than half of the stock and was a co-inventor on 82 US and 189 foreign patent applications. Her president, Sunny Balwani, had held that role since 2009.

The profile repeated the company's claims: more than 200 tests offered, with plans for more than 1,000, up to 70 tests from a single sample of 25 to 50 microliters held in a small vial the company called a "nanotainer", results within hours, and prices often a half to a quarter of those at independent labs. Fortune also noted that Theranos had published no validation studies in peer-reviewed journals, and that Holmes argued peer review was unnecessary because the methods were conventional. The author later published what he called a protracted correction, in December 2015. The page opened for this case does not include the correction text.

### Partners and board

Theranos had signed up large retail partners. Fortune reported 21 Walgreens collection centres in 2014 with plans for a substantial share of the chain's 8,200 stores. Fierce Biotech reported that the Walgreens relationship began in 2013 and covered 41 stores in Arizona and California, while CNBC dated the working relationship to 2010 and said Walgreens at one point hosted about 40 testing centres, so the sources differ on the start date. Safeway invested $350 million to build clinics in more than 800 supermarkets, in a project codenamed "T-Rex" that was never made public.

The board was an unusual asset. The AP report named Henry Kissinger, George Shultz, James Mattis, William Perry, Sam Nunn and Richard Kovacevich among the directors. Fortune also named Bill Frist and said the lawyer David Boies attended board meetings as de facto legal adviser. The investors included Rupert Murdoch, Larry Ellison and the Walton and DeVos families.

## The industry

Clinical laboratory testing is a large, established service. Hospitals and independent laboratories run tests on blood drawn from a vein, and doctors rely on the results to diagnose and treat. Theranos described itself as a low-cost challenger that would undercut them. Fortune reported that its prices were often a half to a quarter of independent lab prices and a quarter to a tenth of hospital lab prices, and that the company said they never exceeded half the Medicare rate.

Two features of the industry mattered. First, accuracy is the product. A diagnostic test that gives a wrong number is not a cheaper version of the right service; it is a hazard to the patient. Fierce Biotech's account of the Safeway pilot gives an example: at Safeway's Pleasanton headquarters clinic, one executive's high result suggested prostate cancer, but a retest at another lab came back normal.

Second, regulation worked differently for different routes. Fortune explained that Theranos's tests were lab-developed tests, which the article said do not need FDA approval, and that Holmes said the company was voluntarily seeking FDA clearance for all of them. But Fierce Biotech reported that the FDA found the company's nanotainers were uncleared medical devices, after which Theranos stopped finger-prick collection for all but one of its more than 200 tests. The laboratory itself was inspected by regulators, and CNBC reported that in early 2016 they found problems with its practices that they said put patients in danger.

The industry had a further structural feature: the customers and the partners were separate. The people buying the tests were patients, but the people funding the company were investors and the people deciding where to place it were retail executives. Fortune noted that no peer-reviewed validation studies had been published, so outsiders had little independent evidence about how the device performed.

## The decision

For most of its life Theranos faced a recurring choice, which became sharper once it started selling tests to the public in pharmacies: whether to describe the technology as it actually performed, or as it hoped it would perform. The sources show what the company chose. They do not show whether it formally weighed the alternatives below, which are the obvious ones, set out here for analysis.

### Option one: publish the evidence and seek clearance first

The company could have submitted validation data to peer review and to the FDA before selling the service. Fortune noted the absence of peer-reviewed validation studies in 2014, and Holmes's answer was that the methods were conventional. This route would have been slower and might have exposed the device's limits early, but it would have placed the claims on public evidence.

### Option two: narrow the claims to what the device could do

The SEC later alleged that the company's analyzer could perform only a small number of tests, and that most patient tests were run on modified or commercial analyzers made by other companies. A narrower claim, say a short list of tests on a finger prick, would have been honest, though it would have made a smaller story to raise money against.

### Option three: keep promising, and run the tests elsewhere

This is the course the evidence at trial describes. Samples were run on conventional machines while the company presented Edison as the engine of its service. It kept the commercial promise and the partnerships alive. The risk was that every extra patient, partner and investor added a person who might later notice.

### Option four: pause the launch

The company could also have held back commercial rollout until the device matched the pitch. That would have strained relationships with Walgreens and Safeway, and made fundraising harder, but it would have stopped patients receiving results from a service that the company knew was not working as described.

## What they did

### 2013 to 2015: rolling out

Walgreens began offering tests in Arizona and California, and Safeway built clinics. Fortune's 2014 article gave the public account of what Theranos could do. The SEC's later complaint described the claims made to investors, including that the company's products were used by the US Department of Defense in Afghanistan and on medevac helicopters, which the SEC alleged never happened. It also said the company projected more than $100 million of revenue for 2014 while actual 2014 operating revenue was a little over $100,000.

### October 2015 to 2016: exposure

The Wall Street Journal questioned the technology in October 2015. Walgreens said it would end the expansion of its partnership, and froze it pending answers. Safeway then severed ties, and its $350 million deal collapsed. Fierce Biotech reported that the Theranos device handled only a small share of consumer tests sold at the end of 2014. In early 2016, regulators found problems with the laboratory.

### 2017: lawsuits and settlements

Partner Fund Management, a hedge fund that put more than $96 million into Theranos in 2014, accused the company of securities fraud, alleging it was lured in "through a series of lies, material misstatements and omissions." Theranos reached a confidential settlement with the fund in 2017. In June 2017 CNBC reported a tentative agreement under which Theranos would pay Walgreens less than $30 million, against the $140 million that Walgreens sought, a figure that included a $40 million convertible note. Theranos said the suits were without merit.

### March 2018: the SEC charges

On 14 March 2018 the SEC charged the company, Holmes and Balwani with a fraud that raised more than $700 million from investors. Theranos and Holmes settled without admitting or denying the allegations. Holmes agreed to a $500,000 penalty, a ten-year bar on serving as an officer or director of a public company, and the return of 18.9 million shares, and gave up majority voting control. The SEC said that, because of liquidation preferences, she would not profit until more than $750 million was returned to defrauded investors and other preferred shareholders. Balwani contested the claims.

### 2022: trial, verdict and sentence

Holmes testified for seven days. According to the AP report she blamed Balwani for failures in the laboratory and said he had abused her, and that she was close to fixing the technology; she often said she could not recall key events. She was convicted on 3 January 2022. Sunny Balwani was convicted on all 12 charges in July 2022.

The case on one page. Free to save and share with credit.

## What happened

### The verdict

The jury convicted Holmes of investor fraud and cleared her on the counts about patients who paid for tests, which prosecutors had been expected to find harder to prove because she never communicated with patients directly. The AP report put the convictions at three counts of wire fraud and one of conspiracy. CBS News reported two counts of each, so the sources differ on the split. Each count carried up to 20 years.

### The sentences

On 18 November 2022 Judge Edward Davila sentenced Holmes to 11.25 years (135 months) in prison and three years of supervised release, at the low end of a guideline range of 135 to 168 months. Prosecutors had sought 15 years. Her lawyers asked for no more than 18 months. Prosecutors put investor losses at $804 million, and her team at $40 million to $48 million. Balwani received 13 years.

ItemFigure
Raised from investors (SEC, 2018)more than $700 million
Raised from investors (AP, 2022)more than $900 million
Holmes sentence135 months
Balwani sentence13 years
Joint restitution$452,047,268

### Restitution

Holmes and Balwani were held jointly and severally liable for $452,047,268 in restitution. It covered 12 identified individual investors plus Walgreens and Safeway. The largest shares were about $125 million to Rupert Murdoch, about $100 million to the DeVos family's RDV Corporation, $40 million to Walgreens and $14.5 million to Safeway.

### Reporting for prison

Holmes was due to surrender on 27 April 2023. The Ninth Circuit rejected her last attempt to delay, and the court recommended the minimum-security federal prison camp in Bryan, Texas. Balwani reported to a federal prison in San Pedro, California, on 20 April.

## Other views

### The defence

Holmes admitted some mistakes. She said she believed that Theranos was close to fixing its technology and blamed Balwani. She said he had abused her over a long period and controlled her diet, sleep and friendships. The jury heard that account and still convicted her on four counts.

### The prosecution's view of the loss

Prosecutors argued for a loss of $804 million and a sentence of 15 years. The defence said the loss was a small fraction of that. The judge's guideline range sat between the two positions. The gap shows how contested the quantum of harm was, even when the conduct was found proved.

### The partners

Walgreens and Safeway were victims in the restitution order, but their own conduct is also part of the story. Fierce Biotech reported that Safeway's relationship dated to at least 2011 and that its executives referred to the project as a wellness bet. Both retailers had put the Theranos name in front of customers before the technology had been independently tested. The sources opened do not show what checks they carried out.

### The investors

Investors who sued said they had been misled. Partner Fund Management said its 2014 investment of more than $96 million was drawn in by lies, and settled in confidence in 2017. The SEC's 2018 settlement had a notable detail: because of liquidation preferences, Holmes would not profit until more than $750 million had been returned to defrauded investors and other preferred shareholders. That condition shows the regulator treated investor recovery as the priority, ahead of the founder's equity.

### The press and the early profile

Fortune's June 2014 profile gave the company's claims a wide audience, and its author later published a correction in December 2015, which the page opened for this case does not reproduce. The episode is a reminder that profiles built mostly on access to the company and its advisers can pass on claims that the reporter has not been able to test. The Wall Street Journal's reporting in October 2015, as described by CNBC, was the turning point after which Walgreens stopped expanding the partnership.

### Limits of the evidence

The trial convicted Holmes on counts about investors, not about patients. The jury's partial verdicts mean that the legal record does not say everything about the quality of the tests. The sources also differ on the sum raised, with the SEC citing more than $700 million in 2018 and the AP citing more than $900 million in 2022. The figures may count different periods or instruments, and the sources do not say.

## Reading it through a framework

### Stakeholder mapping

Theranos had at least four stakeholder groups with different information. Investors bought on claims and a board. Retail partners bought on claims and reputation. Patients bought on convenience and trust in the pharmacy brand. Regulators saw only what they inspected. Information was most limited for the group that carried the most physical risk. The conviction on investor counts and acquittal on patient counts show how hard it was to connect the company's statements to those who relied on test results.

### Signalling and credibility

When customers cannot judge a product directly, they rely on signals. Theranos supplied strong ones: a board of former statesmen and officials, famous investors, a large valuation and big retail partners. Those signals were not evidence of performance. Fortune's 2014 article itself recorded that no peer-reviewed validation had been published, a gap that the glamour of the board covered. The framework shows why strong signals can delay scrutiny, and why a company that relies on them has an incentive to protect them.

## Since then

The legal story reached its end point in 2023. Holmes's appeal attempts to delay her surrender failed, and the court recommended a minimum-security camp in Texas. Balwani reported to prison in April 2023. The restitution order of $452,047,268 is the largest financial consequence reported in the sources opened. The sources opened for this case do not report later developments in 2024, 2025 or 2026, and none of the company's own accounts of its wind-down could be read, so those are left out.

## What to take from it

- Famous names on a board are not a substitute for evidence. Theranos had Kissinger, Shultz and Mattis among its directors, and its 2014 profile still noted the absence of peer-reviewed validation.

- In health products, accuracy is the product, and failures land on people who never signed a contract with the company. Holmes was acquitted on the patient counts, but the Safeway pilot shows the practical risk.

- Claims made to investors are tested against later facts. The SEC pointed to a projection of more than $100 million of 2014 revenue against actual operating revenue of a little over $100,000.

- Partners carry risk when they put their own brand behind an unproven service, and they may recover only part of it. Walgreens sought $140 million and reportedly settled for less than $30 million.

- Regulatory gaps do not remove responsibility. Lab-developed tests did not need FDA approval, yet regulators still found problems in the laboratory and the nanotainer was ruled an uncleared device.

## Questions to think about

- What would you have asked Theranos for before agreeing to put its tests in your pharmacies or supermarkets?

- Should investors in a private company with a large valuation be entitled to see independent validation of its technology, and who should require it?

- How should a board of non-experts oversee a company whose product it cannot assess?

- Why might a jury accept fraud against investors more readily than fraud against patients?

## Concepts at work

This case draws on information asymmetry, signalling and corporate governance. Explore them in [Insights](/insights/).

*Written from the public sources listed below. Figures are as those sources report them.*