QZ & A

Why scientists make bad entrepreneurs—and how to change that

Untranslatable dreams.
Untranslatable dreams.
Image: Reuters/Kim Hong-Ji
We may earn a commission from links on this page.

Beyond the gifts of nature, we are surrounded by creations that came from science. Yet, scientists are rarely the ones who make money from the advances that shape our world. While Silicon Valley creates billionaires before they turn 30, no Nobel Laureate has ever become a billionaire.

Steve Blank, often called one of the godfathers of Silicon Valley, wants to change that. After being approached by the biggest science-funding body in the US—the National Science Foundation (NSF)—he launched a program to train scientists to become entrepreneurs.

Having trained 500 teams of scientists in the art of entrepreneurship, Blank is one of the best people to explain why scientists still don’t profit from their inventions. Quartz caught up to learn about about what most startups do wrong and how scientists can change the minds of venture capitalists. The following is an edited and condensed version of the conversation.

Quartz: Your work is the inspiration behind the Lean Startup method. What is it? How did it come about? 

Steve Blank: I started working in Silicon Valley in the 70s, and entrepreneurship as we know it was relatively new. Before that, investors treated startups as nothing but a smaller version of big industries. What that meant was that venture capitalists (VCs) expected startups to do everything that a large company would do. If large companies wrote business plans, we needed to write business plans. If large company hired sales, marketing and business development staff, then we needed to do that. And if large companies put up five-year forecasts and revenue estimates, then investors would expect startups to do that and live up to them. It seemed pretty logical because large companies were doing all those things, and they were doing pretty well overall.

steve blank
Making science great again.
Image: Eric Millette

But startups didn’t succeed in that environment. Investors blamed the founders for their failures. They never once questioned the methodology, and just assumed that success was harder in startups. When I retired, it dawned on me that they were making a basic mistake. There was a fundamental flaw in assuming that startups are small versions of big companies.

It turns out that large companies are good at executing business plans. A company is large because they already have many customers and they already know their competitors. To be able to win, they need to do whatever it takes win in a well-defined game. When they launch a second product, it is not that risky because they already have a set of known factors. That’s not true of startups, which are often trying to do something completely new.

So the lean-startup method says that you don’t come here to execute a business plan. You come here because you have a set of untested hypotheses. You test these hypotheses quickly and you can get your business going faster.

This should make perfect sense to a scientist: Experiments are the most rigorous way to figure out if something works or not. So why aren’t there more scientists who are entrepreneurs?

About 30 years ago, the US Congress realized that they’re investing all this money in basic science and that some scientists wanted to commercialize some of these discoveries and inventions they were making. To help them, it effectively setup the equivalent of venture capital without taking an equity. This was the SBIR (Small Business Innovation Research) program, and they started accepting applications from US scientists for commercialization grants. They were giving serious money for free: from $200,000 to $3 million.

When the National Science Foundation (NSF) approached me, I realized that the results from this program were terrible. Scientists turned out to be horrible at commercialization. But no one was measuring it, and that’s why the program kept running. The guy behind the program at the time had the mandate to do something to improve this.

He said to me, “We think you have invented the scientific method for commercialization of technology. We believe that scientists will get this immediately.” That’s when I designed a class for them and that became Innovation Corps (I-Corps) program. This program exceeded the expectations of all those involved in the SBIR, because we trained thousands of scientists and many of them went on to form businesses.

So you have trained more than 500 teams of scientists and that have gone on to create 261 startups. But they have raised only $49 million in funding. Why so little?

The reason for not raising as much money is that most science is pretty esoteric. These scientists are not working on a social media network. They are not working on problems that are of great interests to VCs.

The other problem was that they might have had an interesting academic paper but when they go in front of an investor and are asked “tell me about your customers,” they are left silent. When asked “tell me about your revenue model,” they don’t understand the question. Our training helped overcome some of that, but there was also a problem on the VC side.

Why do VCs not fund such science-oriented startups?

Because VCs are people who goal-seek for the largest return. They are a financial-asset class, and they are not interested in optimizing other things. They are driven by financial incentives, and I don’t mean this in the pejorative sense. Even if some of these startups are the gold mines that VCs seek, they are not investing in them because their time and space is limited. The lead time for science startups is much greater. Why should VCs spend more efforts when there are millions to be made on fart apps?

In a perfect world, we would be allocating money and giving investors incentives to invest in things that are, say, good for health or the country or the world. But that’s not how venture capital works. The market has priced science out of the most valuable activities. There are corner cases, but most ideas are at best $50 million companies, even if they might prove to be lynchpins in the future of humanity. There are plenty of people working on the next health IT company or in finding the next blockbuster drug, but there are hundreds of ideas which aren’t being funded even though they deserve to be funded. And that’s because there is only a finite amount of venture capital chasing other high-return investments.

There is, however, enough academic work going on which is of considerable interest to VCs: artificial intelligence, genetic testing, digital health and such. So if money is not the only metric by which you measure the success of I-Corps, what is?

You’re right, but there’s more to this story. Errol Arkilic, the NSF program director, wanted to use the I-Corps program to change the mindset of scientists. They may not seem to be entrepreneurs, but they are in one way. They are running their own labs, which employ junior scientists and vie for funding. So the training that the program would give them could help them for the rest of their career, not just for commercializing one piece of technology. This would affect an entire generation and how they look at science.

What’s more is that the VCs who have come to teach the I-Corps class, which is something I ensured was a feature, have changed too. For instance, one of them thought that the only way to do a medical therapeutics company successfully was to raise millions and millions of VC funding, but he realized that he is wrong. He now believes that you can get a therapeutics company off the ground if you partner with Big Pharma.

So given how radical this approach is, the world hasn’t changed overnight but it is changing much faster than it would have. What’s heartening to me is that more companies than ever before are being run by scientists, and not by MBAs. It is something I hope to see more of, because scientists are definitely capable of running companies and they totally deserve to run them.

What do you think is a change that a scientist must go through to become an entrepreneur?

The “must” is to realize that just because you are the smartest person in the building does not make you capable to run a company. That is a big idea, and not all scientists are comfortable with it. It’s one reason we keep the I-Corps program admission process so rigorous. We need people to recognize that, apart from making the technology better, people must accept that they don’t yet have the commercialization insights they need. All they have right now is a bunch of untested hypotheses.