Lessons of Theranos

I’m halfway through listening to the audiobook of John Carreyrou’s Bad Blood: Secrets and Lies in a Silicon Valley Startup. It’s an interesting time to be learning about this company, as Theranos’s CEO Elizabeth Holmes was just convicted of lying to investors.

I’ve helped launch or participated in my share of startups and been briefed by many more as an analyst. Carreyrou’s book describes a startup environment different from any I’ve ever encountered — and far more toxic.

Basically, to understand why Theranos was so different from the typical startups, you need to understand the difference between failed aspirations and lies.

Failed aspirations sound like this: “We are going to ship in August. The product will cost $375. We will change the world.”

Startup leaders say stuff like this all the time. Basically, look at anything Elon Musk says about Tesla — for example promises about full self-driving and ship dates for the Cybertruck — and you get an idea of what these sorts of overpromises sound like. They are most often statements about the future.

Lies are statements that are factually false. “The blood test is working now [when it isn’t]. It can test for 100 different conditions [when it can only test for one-tenth that many]. We have deals with a five of the top ten pharma companies [when there are no actual finalized deals].” Notice that these are statements about the present.

Holmes and Theranos’s other senior managers shared actual lies, not failed aspirations.

Signs of trouble

The best startups that I worked with were as transparent as it was possible for them to be. They shared financial results when they could. They told people what was happening at company meetings. And while they certainly energized people with soaring aspirations, for the most part, they didn’t out-and-out lie about things that were happening.

Carreyrou’s story of Theranos is rife with warning signs that I’ve never seen in any successful startup anywhere.

• Lying to investors (the crime for which Holmes was just convicted).
• Lying about facts to employees.
• Huge amounts of turnover — a complete lack of stability in employees and management. You can’t get anywhere without some stability in employees.
• Firing people for privately shared concerns about problems. Certainly, if you share confidential bad news outside the company, you could get fired. But if you shared it with your boss, they would typically take your concerns seriously and either address them or try to explain why they weren’t a problem — not fire you. Sharing bad news with your boss is not disloyalty; it’s essential to elevating and addressing actual problems.
• Not allowing people in different groups to share information. At Theranos, the only ones who had the whole picture seemed to be the CEO Holmes and the president of Theranos, “Sunny” Balwani. They kept groups in the dark about what each other were working on.

No organization that operates this way can succeed. This is a baseline — if you don’t do these things, you’re certainly not guaranteed to succeed, but if you do do them, you will surely fail.

I’ve seen all sorts of abusive behavior in startups. Managers who continued to insist on impossible deadlines, just to get people to work themselves nearly to death. Sneaky behavior to get access to other companies’ confidential information. Bonuses that somehow dried up when the time came to pay them. Those are all reprehensible things to do. But they’re not unusual. Theranos crossed over into some zone that was so far beyond the normal that it could never succeed.

When loyalty is more important than facts and facing problems squarely, an organization will never accomplish anything and must eventually collapse.

If you are working for a company and you see any of these behavior in your management, quit. Quit right now. It’s a good time to look for another job. The chances of your financial incentives (like stock options) paying out are far less than the chances that your company will crash and burn.