Bad Blood: The book that unmasked Theranos and Elizabeth Holmes, and a broken start-up culture

Elizabeth Holmes founded Theranos - a blood-testing technology company - in 2003, when she was 19.

The convicted founder of Theranos, Elizabeth Holmes, and her former partner Ramesh “Sunny” Balwani had managed to con investors and partner companies for years.

The reporter who finally exposed the scam, John Carreyrou, who used to work with The Wall Street Journal, had written an award-winning book Bad Blood published in 2018. Here are some interesting excerpts and tidbits from the book, which reveal the pitfalls of a FOMO-driven culture, the millions it could cost investors and, most importantly, the lives it could put at risk.

Faking it had become the norm. VCs have a term for startupreneurs overstating forecasts–the hockeystick forecast. It shows revenue stagnating for a few years and then magically shooting up in a straight line. Holmes was great at projecting optimism, even using “extra-ordinary” in her mails to staff, italicising extra and hyphenating for emphasis. She reserved death stares for anyone who questioned the company’s claims of progress. Theranos’ co-founder Shaunak Roy told company’s CFO Henry Mosley that investor presentations faked real-time analysis, and showed pre-recorded results, to hide the product’s breakdowns.

Dissenters weren’t tolerated. In 2006, its valuation was $165 million dollars, based on pharmaceutical contracts that weren’t ever shown to the CFO Henry Mosley citing “legal review”. When Mosley raised concerns about the reader’s unreliability and fooling investors, Holmes fired him. Later, when a board member and former head of software engineering at Apple, Avie Tevanian expressed doubts, he was also encouraged to resign and threatened with a lawsuit. IT head at Theranos, Matt Bissel was sometimes asked to prepare dossiers on people who quit, so that they can later be used for leverage. Bissel was always uncomfortable with Holmes’ demand for absolute loyalty from her employees, and her hostility against those she suspected of not having that.

Even the brightest minds can be misled. In 2003, when Elizabeth did her summer internship at Genome Institute of Singapore, testing specimens to check for SARS, she thought there must be a better way than the low tech syringes and nasal swabs. Back home in Houston, she sat at her computer for five days straight, sleeping very little and came up with an idea for an arm patch that could both diagnose and treat medical conditions. She filed for a patent. At Stanford, she presented it to Robertson who was awed by her determination and drive and he encouraged her to keep going. Shaunak, then a college peer, was skeptical about the concept but he got caught up in Robertson’s enthusiasm.
Little ‘outsider’ oversight in the beginning. Holmes got a lot of help from her immediate social circle, which was also powerful. It stopped people from asking tough questions. One of the early investors was Tim Draper, father of her childhood friend, and Victor Palmieri, a long-time friend of her fathers. Draper’s firm DFJ had a formidable reputation after making a killing from early investments in tech companies including in Hotmail. In contrast, in 2004, MedVenture Associates, a venture capital firm that specialised in medical technology investments asked her difficult questions about her diagnostic patch, andshe left in a huff.

The reader was a big engineering problem. It was finally solved by modifying a three-thousand dollar, glue-dispensing robot from a New Jersey-based Fisnar. The team miniaturised the mechanical arm attached to a gantry, and fixed a pipette at the end. Then, they had it mimic a human chemist’s actions.

Holmes thwarted an attempt to unseat her from CEO post. In 2007, one of the sales heads Todd Surdey and the general counsel Michael Esquivel took their suspicions about the Theranos machines being unreliable and about investors being misled to the board. An emergency meeting of the board was called and Holmes was to be removed from the CEO post. But when she was called in, she was contrite and charming and convinced everyone that she would mend her ways and allow a more transparent management. After she won that round, she fired Surdey and Esquivel.Sunny “disappeared them”. Ramesh “Sunny” Balwani was appointed the vice chairman of Theranos but employees were unsure of what role he really played. He seemed to hold a lot of influence over Holmes and one of Holmes’ closest friends thought of him as “bad news”. Many thought of him as “boastful and patronising” and he gained a reputation as a hatchet man, having an ability “to disappear” an unwanted employee in a reference to the secrecy with which people were dismissed. Holmes had met him at Stanford’s Mandarin programme, and no one is clear when they become romantically involved. Sunny had made his money when employer and a B2B platform CommerceBid was sold to Commerce One for $232 million and, as CommerceBid’s second highest-ranking officer, he got $40 million.

Walgreens’ FOMO blunder: Between 2009 and 2010, when the US was recovering from the deep recession, investors were rushing to invest in startups. Facebook had an insane valuation of $23 billion, which jumped to $50 billion in six months, Twitter had been valued at $1 billion and an obscure cab-hailing app UberCab had a beta launch of its service in San Francisco. Also, interest rates were at their rock bottom. Managers of East Coast hedge funds, who were until the interested only in publicly traded stocks, and old, established companies were sending messengers to the West in search of the next big thing and innovation. That’s how Walgreen’s Dr J or Jay Rosan ended up considering and defending a partnership with Theranos. Even when their business consultant Kevin Hunter raised alarm over the start-up’s reluctance to be transparent about their product and its efficacy, Walgreens execs turned a deaf ear because they feared Holmes would take her business to their competitor CVS. Hunter was sidelined and even asked not to attend meetings, when he began asking difficult questions to Holmes and Sunny.

Holmes’ brush with the army: When Holmes tried to get the army to use the reader in warzones, Lieutenant Colonel David Shoemaker asked questions about regulatory approval. With a PhD in microbiology, years spent in medical research and the first army offer to complete a year’s fellowship at FDA, he knew the ins and outs of the regulatory maze. What Holmes was proposing to do, to deploy her product, seemed like a clever way to escape oversight. He therefore wrote to the FDA, asking them to check if her method was legitimate. It led to an enquiry, which infuriated Holmes and Sunny, and they stirred up trouble for him by complaining to a much-higher James Mattis, head of the US Central Command. Shoemaker suggested a compromise to Mattis, which involved testing the efficacy of Theranos test against the results from army’s regular testing labs. Mattis agreed but Theranos never took up the offer.

A team fooled Sunny for laughs. To survive “a dysfunctional corporate culture”, a few employees devised some hilarious tactics. John Carreyrou writes, “it dawned on them that they were dealing with an erratic man-child of limited intellect and an even more limited attention span”. Sunny was lost in discussions about engineering. To mask it, he threw technical terms around without really understanding it. Arnav Khannah, a young mechanical engineer, and his team cottoned on to this. In one of the meetings, a team member had used the word “end effector”, for the claws at the end of a robotic arm. Sunny seems to have heard it as “endofactor” and kept referring to these fictional endofactors. Arnav’s team presented “Endofactors Update” in their next meeting and Sunny “didn’t bat an eye and the meeting proceeded without incident”. They had a good laugh. The team also began using the word crazing “liberally and out of context” to see if Sunny would parrot it. He did. Crazing is what causes fine cracks on a surface. Even early in the day, Khannah had figured out a way to get Sunny “off his back”--write lengthy replies to Sunny’s emails, preferably over 500 words; and send invites to biweekly meetings and he would soon lose interest or forget to show up.

The Frankenstein goof-up. One of the problems with Theranos’ miniLab was that it could not process more than one sample at a time, a fault which would have resulted in long queues of customers and ridicule, since the company had promised fast results. To get over this, they decided to stack six miniLabs and all of them would share the same sample. This “Frankenstein machine” had one problem–each of the miniLab stacks produced heat that messed with the tests. An new employee, a 22-year-old graduate who had just been out of college, was shocked that the company had missed something so “basic”.

Rupert Murdoch’s $125 million investment. He was Theranos’ biggest investor and the startup was his single biggest investment outside of his media empire. Carreyrou, whose former employer WSJ is owned by the media mogul, writes that Murdoch usually did “no due diligence to speak of” when putting in his money and simply followed his gut. But he proved a good boss--when the story was breaking and Theranos founders tried to get the story pulled out through Murdoch, he refused to play ball. After she got his funding, Holmes tried to convince Murdoch that Carreyrou’s story was based on false information and that it would do great damage to the startup. Murdoch refused to interfere, saying “he trusted the paper’s editors to handle the matter fairly”. As the reporter was getting close to the story’s publication, Holmes tried once again to convince Murdoch to kill the story but was politely refused.

She fought back with “bald lies”. Holmes showed up for WSJ’s D.Live conference, a few days after the damning story came out, though many assumed that she would back out. Carreyrou and his colleagues had expected her combative behaviour but what they were shocked by were her “bald-face lies in a public forum”. She insisted over and over again that the nanotainer (the small container which contained the blood samples) were withdrawn by Theranos voluntarily. The truth was, FDA had demanded the withdrawal, declaring the nanotainer an “unapproved medical device”. She said that the Edison reader cited in WSJ article was an old model that was no longer used for their test. The truth was it was still being used, as shown by a regulator’s investigation. But the biggest lie to Carreyrou’s mind was her categorical denial of diluting the finger stick samples to run their tests. His source and the start-up’s former head of laboratory had given clear evidence how Sunny and Holmes were aware this was done.

Holmes made another attempt at selling miniLab as a different product. Even after she was banned from running clinical laboratories, she made a presentation to a group of clinical laboratory scientists trying to sell miniLab. At the talk, she shared data but based on venous (drawn from the vein) blood as opposed to the finger-stick test, which had been Theranos’ big idea. There was little data based on finger-stick tests but they covered only 11 tests, a big draw down from the hundreds of tests they had promised earlier. It was a weak pitch but Carreyrou notes her “amazing” skills as a salesperson. “Like her idol Steve Jobs, she emitted a reality distortion field that forced people to momentarily suspend disbelief,” he writes. The effect didn’t last long however–two academicians from the crowd raised questions on miniLab’s highly scaled down capabilities and the technology that didn’t seem very different from Theranos' laboratory method.