“The worst possible thing in the world is to have someone who doesn’t believe you.” – Elizabeth Holmes

When it is all over, and the whole truth is known, Theranos will undoubtedly join the great frauds of business history, competing with the likes of Enron and Worldcom. More lines have been written about Elizabeth Holmes than, as one article argued, on Elvis. So, what’s new? I believe that everybody missed how she and her partner convinced Investors to put money in a company that promised wonders and delivered only ruin.

In this post, I analyze the accounts of fraud detailed in the recent indictment and debate what Investors should, and could, have done about it. My conclusion is that the most salient issue underlying the Theranos case is not technical, such as governance or inexperience in due diligence—although they play a role—but has psychological roots: confirmation bias.

What Was The Real Problem With Theranos?

There is a surprisingly high level of confusion on the causes for the Theranos debacle.

Anti-Silicon Valley critics argue that fraud shows the limits of the “fake it till you make it” ethos that seems pervasive in Startupland. The Edison machine, touted for its almost magical capabilities—it was supposed to deliver as much as 200 test results with a single drop of blood—never worked. Like the illustrious inventor whose name they borrowed for their device, Theranos executives probably found “10,000 ways that won’t work.” But none other.

Silicon Valley Venture Capitalists, on the other hand, point to the fact that no reputable VC firm had invested in Theranos. For these advocates, Theranos is the poster child of what goes wrong when new entrants in the asset class play VC. They highlight investors including Rupert Murdoch and Larry Ellison, who are phenomenal entrepreneurs but not professional Investors, and wealthy families such as the Waltons (behind Walmart).

A closer look at Crunchbase’s database shows that such finger-pointing is not totally honest. One of the early Investors in Theranos, and a valiant defender of Holmes until she was indicted (more on that later), is Tim Draper. A controversial figure, Draper’s track record indicates savviness. He’s credited for Hotmail’s success and has led early investments in Tesla and SpaceX.

I can’t be bothered to go into this again, but no, *telling lies* is not the ‘Silicon Valley ethos’. Tolerance for failure, yes. Bootstrapping, yes. Unrealistic ambition, yes. Lying, no. This is like claiming that Jayson Blair was an indictment of ‘the journalism ethos’ https://t.co/930SKvK9II

— Benedict Evans (@benedictevans) January 4, 2022

If not a culture of lies and a lack of experience, what is the cause of the Theranos fiasco?

Some voices have aimed at a distinct failure in governance, starting with its Board of Directors. It has been said that Theranos’s Board was better suited to attack a country than ask tough questions on the workings of a medical device. The list of high-rolling ex-government officials includes Henry Kissinger and George Shultz (former Secretaries of State), Jim Mattis (retired Marine Corps. four-star general), and William Perry (former Secretary of Defense).

A well-functioning Board may have been able to uncover the truth and force some change, as the Benchmark handling of Uber showed. However, in some cases, even astute Investors known for their independent minds hesitate to speak up—a possibility most recently highlighted in the WeWork disaster.

Having worked in private equity for over fifteen years and sat on as many Boards, I believe that the proper dynamics were not at play at Theranos. Holmes and her partner Sunny Balwani were good at hiding stuff, and Board members were good at looking the other way. They were even, in some cases, incentivized to do so; and fell into the confirmation bias trap.

Could Investors Have Known—And Done—Something Earlier?

I read the indictment recently released in the Theranos case. It is a fascinating document that clearly lays out the charges against Holmes and Balwani. The indictment also recaps the history of Theranos. The first significant date for our purpose is September 2013, when Holmes came out of stealth mode to announce a partnership with pharmacy chain Walgreens—the (in)famous Wellness Centers.

To understand what went wrong, we must first analyze the accusations against Theranos’s leaders.

There are two main counts as two groups of people were defrauded in the Theranos case: Investors and patients. The violation is that of Title 18, United States Code, Section 1349, which stipulates that:

“Whoever, having devised or intending to devise any scheme or artifice to defraud, or for obtaining money or property by means of false or fraudulent pretenses, representations, or promises, transmits or causes to be transmitted by means of wire, radio, or television communication in interstate or foreign commerce, any writings, signs, signals, pictures, or sounds for the purpose of executing such scheme or artifice, shall be fined under this title or imprisoned not more than 20 years, or both.” (Source: Cornell law school)

Specifically, the Theranos defendants are accused of:

• Against Investors: soliciting investments through making false and fraudulent representations
• Against patients: soliciting, encouraging, or otherwise inducing doctors to refer and patients to pay for and use its laboratory and blood testing services under the false and fraudulent pretense that Theranos technology produced reliable and accurate blood test results

The final verdict form dated January 3rd, 2022, indicates that the defendants were found guilty of the first count (against Investors), but not the second (against patients). Sunny Balwani’s trial is expected to occur later this year, so this verdict chiefly concerns Elizabeth Holmes.

Let’s dive into the scheme to defraud Investors.

Although nine false representations are mentioned in the indictment, they belong to two categories: technology and financials.

We now detail each of them and assess whether Investors could have avoided being duped.

False Representation #1: Technology

Theranos’s top executives used several strategies to fool Investors about the supposed capabilities of their device (mentioned as TSPU, Edison, or miniLab).

Firstly, Holmes and Balwani lied about what the machine could do. Holmes kept touting the device’s prowess: it could perform more tests with a smaller blood sample, do it more accurately than conventional labs, and faster.

The reality, however, was much starker. The indictment claims that “Holmes and Balwani knew that Theranos’s proprietary analyzer had accuracy and reliability problems, performed a limited number of tests, was slower than some competing devices, and could not compete with larger, conventional machines in high-throughput, or the simultaneous testing of blood from many patients, applications.”

Demonstrations organized for Investors were running a “null protocol” that gave the impression the machine was operating. As became clear later, the device collected blood, but the samples were then analyzed on third-party analyzers.

Could Investors have done anything about these representations?

Typical VC due diligence includes a technical or technological check to verify what is under the hood, especially in the case of hardware startups. Although Walgreens did conduct such verifications, the process was botched. The expert retained by Walgreens explained, in abc’s exceptional podcast about the Theranos debacle, The Dropout, that he never got access to the machine.

The appearances were there. Walgreen opened 40 Wellness Centers with Theranos, giving Investors the false impression that they had vetted the machine. In reality, no processing was done locally. Theranos shipped the blood samples to a central lab—which, as mentioned earlier, analyzed the samples of third-party analyzers.

In a 2017 SEC deposition, Balwani explained that the change in operations allowed Theranos to circumvent FDA clearance. As we know from the indictment, this was also a lie.

What is surprising of—and damning to—the Board and Investors is that they don’t appear to have asked for an external audit of the technology after it became clear that test results were faulty.

False Representation #2: Financial Projections

Startups famously don’t meet their financial projections. As I demonstrate in this webinar, Investors ask Founders to establish them for other purposes.

How could Holmes and Balwani be indicted for making false representations in that area?

Let’s first look at the numbers Theranos communicated to Investors. As late as Q4 2014, Holmes allegedly projected that the company would generate $100 million in revenues for 2014 and $1 billion in 2015.

The problem is, Theranos only had revenues in the hundreds of thousands of dollars at that point in 2014.

Could Investors have done anything about these representations?

It is not inconceivable that Theranos could have generated far more revenue in a few months than its entire existence combined. Startups, after all, can turn into revenue-generating machines.

VCs assess that possibility by doing several checks, including:

• Verifying existing contracts with clients
• Challenging Founders on their probabilized pipeline
• Calling prospects to assess the potential for new business
• Contacting past clients to understand why they stopped working with the company

Theranos’s Investors don’t appear to have done any of these routine diligence checks. The indictment shows that they took the company’s statements at face value.

Take, for instance, the Walgreens partnership. The indictment states that it was clear, by late 2014, that “the Walgreens rollout had stalled because of several issues”. Yet, Theranos raised over $105 million on October 31st, 2014, from two Investors. It is the date of the wire, which figures in the indictment, not the deal’s announcement.

Worth noting here, Holmes and Balwani are sued for potential fraud against Investors who wired c. $155 million between December 2013 and October 2014. According to Crunchbase, the company raised $140 million before those dates (including $50 million from Walgreens) and a staggering $450 million between March 2015 and December 2017, including $100 million in debt extended to avoid bankruptcy. (A $580 million secondary market operation took place in May 2017, but that is not cash to the company. Most likely, some Investors who saw problems arising offloaded their positions to a new player.) In September 2018, Theranos’s acting CEO announced to shareholders that the company would be dissolved.

Another lie that didn’t come up on Investors’ radar is the supposed business with the U.S. military. Holmes repeatedly put that contract forward to boast Theranos’s “combat-grade” technology. She later confirmed that “Theranos had limited revenue from military contracts and its technology was not deployed in the battlefield.”

To be fair, auditing a startup is not easy and by no means a science. A probabilized pipeline, in particular, can prove treacherous because the probability rates of success hide the fact that there is a binary truth underneath: either you get the contract, or you don’t.

Also, Founders are optimistic by nature—if they weren’t, they wouldn’t be entrepreneurs. They may get it wrong in good faith.

Finally, prospects may also strongly vouch for the reality of the business under consideration at a given moment, to later refuse to enter into business with the startup. I remember calling over a dozen prospects of a fintech startup back in 2010 for a Seed deal, all of which confirmed to us that they thought the solution was very promising indeed. Eighteen months later, none of them had signed a contract with the startup.

But this is not what we are talking about in the Theranos case. A routine due diligence check would have exposed lies—not optimistic, fake-it-till-you-make-it statements but outright lies—put forward by Holmes and Balwani on the reality of their business relationships and potential future revenues.

I never permit myself to get false impressions from anything any one tells me. I form my own judgments.

Hercule poirot – The ABC murders

I could go on and on about the diligence and governance oversights pervasive through the Theranos story. As these have been covered elsewhere, instead, I’d like to focus on a cognitive bias that helps explain how so many experienced business people, journalists, regulators, and even a few scientists got fooled by Holmes. Let’s start with the most obvious one: confirmation bias.

How Confirmation Bias Explains Diligence & Governance Failures In Venture Capital

Among the dozen cognitive biases—errors we make in interpreting the world—that beset humans, confirmation bias is near the top of the most damaging list. Confirmation bias is “the seeking or interpreting of evidence in ways that are partial to existing beliefs.”

For instance, it is one of the primary reasons explaining the current polarization of the political scene. The tendency to only take into account elements that confirm a prior position, to the exclusion of anything else, expresses itself in many portions of our daily lives: entering into disputes or altercations at work, professing truths with friends, or holding up to our side of the argument at home.

The problem is that confirmation bias is often unconscious.

It may lead the individual exercising it to unknowingly:

• Ignore data justifying an alternative position (restriction of attention)
• Give greater weight to data supporting an existing point of view (preferential treatment)
• Focus research on element favoring a preestablished belief (selective testing behavior)

If the narrative potential Investors in Theranos believe in is that Elizabeth Holmes is a genius—the next Steve Jobs, whose black turtleneck she famously copied—and that her invention will revolutionize the world, they are less likely to ask tough questions on the device’s performance or the reality of military contracts.

Once one has taken a position on an issue, one’s primary purpose becomes that of defending that position.

Raymond S. Nickerson (1998). Source: click here

The subtle and often invisible confirmation bias will push to transform unlikely shreds of potential evidence as truths. “This is hard stuff to do, it takes time”, “Entrepreneurs talk in the present of what they will do in the future, she’s no different”, or “10 out of the top 15 pharmaceutical companies have validated the science” (which, by the way, was false) are arguments often heard to justify pouring hundreds of millions of dollars in a company that consistently failed to deliver on its promises.

How Can Venture Capitalists Avoid Confirmation Bias

There are various ways in which VCs try to escape from confirmation bias, including:

• Being aware of it (always the first step!)
• Making investment decisions after debating the opportunity at an investment committee made of other VC firm partners who have not met the Founders
• Adopting a skeptical approach and trying to find contrarian arguments
• Not relying on work supposedly done by others

In the Theranos case, a simple way to better evaluate the science would have been to seek out former scientists who had left the company and ask them why. The scientific method, which includes peer review, was built to confound confirmation bias. Scientists are trained to take a skeptical approach to their work, and that of others.

Most scientists who quit the company after it started stalling on its progress seem to have done so because of strong ethical concerns about how top executives were handling the truth: staging fake demonstrations, selecting data that suited them, and so on.

Aram Founder

Aram is a veteran investment professional with 20 years of experience. He’s realized over 45 transactions across Project Finance, LBO Financings, Growth Equity, Venture Capital, and M&A in half a dozen countries on three continents.

See Full Bio

Venture Capital Entrepreneurship Fundraising