The Theranos case is unique in the NAVETRA™ series because it is the only one where the governance failure produced direct physical harm to patients. False cancer diagnoses. A woman who received a miscarriage notice from a faulty blood test. False positive HIV results. Real people received medically actionable information from a device that its own engineers knew could not reliably produce accurate results — while a board of distinguished public servants looked on, unable to ask the questions that would have revealed what was happening, because none of them knew which questions to ask.
What Actually Happened
Elizabeth Holmes founded Theranos in 2003 at age 19, after dropping out of Stanford's chemical engineering programme. The vision was genuinely compelling: a miniaturised laboratory device — the Edison — that could run hundreds of blood tests from a single finger-prick, making diagnostics faster, cheaper, and accessible to patients who couldn't afford or feared conventional blood draws. Holmes promised the technology would eventually be available within five miles of every American household.
The vision never became reality. The Edison device was unreliable from the outset, producing inconsistent results that its own scientists flagged internally. Rather than disclose these limitations to investors and partners, Holmes and COO Sunny Balwani made a series of decisions that collectively constitute what John Carreyrou — the Wall Street Journal reporter who broke the story — described as one of the most epic governance failures in American capitalism. The company ran most of its tests on commercially available Siemens machines rather than the Edison, diluting the finger-prick samples to run on equipment designed for larger blood draws. Only 15 of the 240 tests offered in Walgreens Wellness Centers were run on the Edison. When Vice President Biden visited the lab in 2015, Holmes and Balwani had created a fake lab for the tour, concealing the true operating conditions.
2003–2013: Theranos operates in near-total secrecy. Holmes cultivates a culture of extreme information compartmentalisation — employees are told not to discuss their work with colleagues in other departments, NDAs are aggressively enforced, and the trade-secret framework is used to prevent partners, investors, and regulators from seeing the underlying technology. The board during this period is composed almost entirely of political figures and business leaders with no healthcare or laboratory science background.
2013: Theranos partners with Walgreens to offer blood tests in 40 stores. Holmes knows at the time of signing that Theranos cannot meet Walgreens' technical requirements. Former NASA engineers are assembled to modify third-party analysers as a substitute. Walgreens believes — because it is told — that the Wellness Centers are using Theranos's proprietary technology. The board approves the partnership without the scientific capacity to evaluate what the company is actually deploying.
2012–2015: Safeway invests $350 million retrofitting 800 locations with clinics for Theranos blood tests. The deal is terminated in 2015 after missed deadlines and unreliable trial results — without Safeway ever having offered the service publicly. The board has no mechanism to understand why the deployment is failing.
2013–2015: Internal whistleblowers including microbiologist Erika Cheung and Tyler Shultz — grandson of board member George Shultz — raise concerns internally. Cheung approaches Balwani; she is dismissed. Shultz raises concerns in writing; Balwani responds by belittling his scientific understanding and invoking his grandfather's position. Shultz resigns. As he leaves the building, Holmes calls his grandfather on the board. Theranos deploys private investigators to surveil former employees and threatens litigation against anyone speaking to reporters or regulators. The board takes no action in response to these employee exits.
October 2015: John Carreyrou publishes his Wall Street Journal investigation: Theranos is not using its own machines, and the Edison produces unreliable results. Holmes denies the allegations and threatens litigation. Walgreens suspends services. CMS begins a regulatory inspection.
January 2016: CMS sends Theranos a letter reporting that its California lab caused "immediate jeopardy to patient health and safety" — specifically regarding a warfarin dosing test that, if inaccurate, could cause fatal blood clots or haemorrhaging. Tens of thousands of test results are subsequently voided. Walgreens ends its partnership and sues for $140 million. Safeway dissolves its relationship.
2018: The SEC charges Holmes and Balwani with massive fraud. Holmes settles, agreeing to pay $500,000, give up voting control of her shares, and accept a 10-year bar from serving as an officer or director of a public company. A federal grand jury indicts both on wire fraud and conspiracy charges.
2022: Holmes is convicted on four counts of wire fraud and conspiracy. She is sentenced to eleven years and three months in federal prison. Balwani is convicted on all twelve counts and sentenced to nearly thirteen years. Theranos had been shut down in 2018.
The governance dimension that NAVETRA™ focuses on is not the fraud itself — that is a criminal matter, adjudicated in court. It is the structural conditions that allowed the fraud to persist for more than a decade while a prestigious board watched, investors deployed hundreds of millions of dollars, a major pharmacy chain put patient-facing blood test services in forty retail locations, and the only people who raised the alarm were junior scientists who were then harassed into silence.
"One of the most epic failures in corporate governance in the annals of American capitalism. No one was truly policing the business's processes or offerings. The board had no system in place to monitor Theranos's compliance with laboratory regulations or identify any of these problems."
The Five NAVETRA™ Domains That Were Failing
Theranos presents the NAVETRA™ framework in its most concentrated form. Every domain failure reinforced every other. The absence of scientific expertise on the board made technical deception possible. The culture of fear made internal escalation impossible. The secrecy framework prevented external validation. The absence of compliance infrastructure meant regulatory requirements went unmonitored. And the knowledge that did exist — inside the heads of scientists who knew the Edison didn't work — had no governance path to the board, the investors, or the partners whose decisions depended on it.
Is the board receiving an accurate picture of what the core product actually does — based on independent scientific assessment — rather than the founder's representation of what it will eventually do?
At Theranos: every board member was highly accomplished. Not one had meaningful healthcare, diagnostic science, or laboratory regulation experience. The board had no independent mechanism — no scientific advisory function with board-level reporting, no technical due diligence process, no external expert retained to assess the Edison's actual performance — that would have given it a picture of the technology separate from Holmes's own description of it. When one investor testified at trial that he had "little visibility into the company" and had invested based on his relationship with Holmes, he was describing Leadership Alignment failure in its most direct form: the board and investors were operating from a picture of the company that only Holmes could provide, because no governance structure existed to provide an alternative one.
Is the organisation's culture aligned toward honest upward communication of technical reality — or has a culture formed in which the fear of the CEO and COO, combined with extreme information compartmentalisation, makes it structurally impossible for scientific truth to reach the people who need it?
At Theranos: the culture was one of the most thoroughly documented instances of toxic Organisational Alignment failure in Silicon Valley history. Employees were forbidden from discussing their work across departments — a policy justified as trade-secret protection but functioning as an information barrier that prevented any single employee from assembling a complete picture of what the company was doing. Balwani regularly fired employees who raised technical concerns. A microbiologist who advocated for industry-standard laboratory safety protections was dismissed. Erika Cheung, who attempted to escalate inaccurate results through proper channels, was dismissed. Tyler Shultz, who raised concerns in writing, was belittled in an email that invoked his grandfather's board position as the only reason Balwani was responding at all. The organisation was aligned to maintaining Holmes's vision — not to producing accurate blood tests.
Does the organisation have compliance infrastructure capable of identifying and escalating regulatory violations, laboratory standard breaches, and product performance failures — to the board, to partners, and to regulators — before patients are harmed?
At Theranos: the board had no system in place to monitor Theranos's compliance with laboratory regulations. The company operated its certified laboratory for months without a director — a direct licensure violation. When it eventually appointed a lab director, the appointee was a dermatologist with no qualification to run a clinical laboratory and was largely absent. Unlicensed personnel conducted quality control procedures. The FDA had approved only a handful of Theranos's tests, yet the company marketed hundreds. The CMS inspection in 2015 did not find a compliance function that had identified and was addressing these issues. It found a laboratory in a state the agency described as causing "immediate jeopardy to patient health and safety." Internal Risk Management at Theranos was not failing. It had never been built.
Are the board, the scientific/technical function, the commercial function, and the regulatory function working from a shared, accurate picture of what the product does — or is each function receiving a different version of reality, with the board receiving the most optimistic one?
At Theranos: the engineers knew the Edison was unreliable. The lab scientists knew the commercial tests were mostly being run on Siemens machines, with diluted samples that degraded accuracy. Walgreens believed its forty Wellness Centers were deploying Theranos's proprietary technology. The board believed it was governing a revolutionary diagnostics company. Investors believed they were funding a product that worked. The commercial partners had signed agreements based on technical representations that the scientists inside the building knew were false. Cross-Functional Alignment failure at Theranos was not just an absence of shared information — it was an active, structured separation of what different functions were told, maintained by Holmes's information compartmentalisation policy and enforced by litigation threats against anyone who attempted to close the gap.
Does the technical knowledge that exists inside the organisation — about what the product actually does, where it fails, and what regulatory requirements it is not meeting — travel to the board, to partners, and to investors in a form that enables informed governance decisions?
At Theranos: the scientists who knew the Edison didn't work were the most informed people in the organisation about the central question on which every investor, partner, and patient decision depended. That knowledge had no governance path to the board. Employees who attempted to communicate it internally were dismissed. Those who communicated it externally — to regulators or reporters — were surveilled by private investigators and threatened with litigation. The trade-secret framework, which Holmes used as the primary justification for preventing external scientific validation, was specifically engineered to close the only remaining paths through which technical knowledge could have reached the board. Knowledge Transfer Gaps at Theranos were not an organisational accident. They were the mechanism that made everything else possible.
The Board That Couldn't Ask the Right Questions
The Theranos board composition is the governance failure expressed in its most visible form. Former Secretary of State George Shultz. Former Secretary of State Henry Kissinger. Former Secretary of Defence William Perry. Former Secretary of the Treasury George Shultz. Former US Senator Bill Frist. Former Wells Fargo CEO Richard Kovacevich. Former US Navy Admiral Gary Roughead.
These were not unintelligent people. They were among the most accomplished public servants in American history. What none of them were was a diagnostic scientist, a laboratory medicine specialist, a clinical pathologist, or a regulatory expert in CMS laboratory certification. The board had the prestige to open doors and the political relationships to attract partners. It did not have the technical capacity to ask whether the Edison could actually perform a warfarin dosing test accurately enough not to kill someone.
The Idaho State Bar's governance analysis of the case makes the critical structural point: no board member faced prosecution or indictment. They were not participants in the fraud. They were, like almost everyone else, deceived. But the question NAVETRA™ asks is not whether the board was culpable. It is whether the governance architecture would have caught the deception before patients received false cancer diagnoses — if the board had been composed differently, or had an independent technical advisory function with board-level reporting, or had required external scientific validation of the Edison's performance before approving the Walgreens partnership.
The answer is almost certainly yes. The knowledge that the Edison didn't work existed inside Theranos from early in its commercial deployment. It had no path to the board because the board had no mechanism to receive it independently of Holmes.
$9 billion valuation. $700 million raised. Two former US Secretaries of State on the board. Zero healthcare experts. Tens of thousands of voided test results. Patients who received false cancer diagnoses, false positive HIV results, and incorrect miscarriage information. Eleven years and three months in federal prison for the founder. Theranos is not primarily a story about a fraudulent CEO, though the criminal record is clear. It is the most extreme available demonstration of what happens when a board's composition, culture, and compliance architecture make it structurally incapable of independently verifying the central claim on which every governance decision depends — that the product works.
In this case, the cost of that structural failure was not measured in shareholder value. It was measured in patient harm.
The Question for Every Board Governing a Technical Product
Theranos is not only a startup governance story. It is the most consequential demonstration of a board composition failure that happens routinely — at smaller scale, with less catastrophic results — in industrial, healthcare, energy, and manufacturing companies across every sector.
A board that governs a technical product without independent technical capacity to evaluate that product is not governing the product. It is governing the founder's description of the product. In most cases, founders are honest. In some cases, they are not. In all cases, the governance architecture should be capable of finding out the truth independently — before the consequences reach patients, customers, or regulators.
NAVETRA™ measures the five structural conditions that determine whether a board is receiving a true picture of the organisation it governs. At Theranos, all five were absent or actively suppressed. The two junior scientists who tried to raise the alarm — Erika Cheung and Tyler Shultz — should not have been the governance mechanism of last resort for a $9 billion company serving patients in forty retail pharmacies.
A board of statesmen cannot govern a laboratory. The prestige that gets you into the room is not the expertise that tells you what is happening inside it.