Uncovering the DNA of successful start-ups

What works and what doesn't work when building a Web start-up? The Startup Genome project aims to answer those questions with its initial report, issued today.

Daniel Terdiman Former Senior Writer / News
Daniel Terdiman is a senior writer at CNET News covering Twitter, Net culture, and everything in between.
Daniel Terdiman
6 min read

If you spend much time with Web entrepreneurs or investors these days, it quickly becomes clear that "pivot" is the hottest term in Silicon Valley. It signifies a young company's shifting of focus, and everyone has an opinion about whether it's something start-ups should be doing or not.

The answer, it seems, is yes. And as long as it's done at the right pace, it can even be an extremely lucrative and important step. In fact, young Web outfits that pivot once or twice can raise two-and-a-half times as much money, see 3.6 times the user growth, and are half as likely to scale too soon than start-ups that either never pivot or that do so more than twice.

That was one of the major conclusions of the initial report of the Startup Genome project, an initiative put together by a group of Silicon Valley investors that aims to identify the DNA of successful Internet start-ups and the teams and investors that build them.

The project began last December as an attempt to "find a way to accelerate start-ups in a scalable way," said Bjoern Herrmann, one of the leaders of the Startup Genome project. It was initially a survey of about 50 start-ups, but over time, the team's ambitions grew, and as the crew set out to codify what works and what doesn't with Web start-ups, its research grew to include surveys from more than 600 companies.

Today, Herrmann and his partners on the project--Max Marmer and Ron Berman--released the first version of their study (PDF), a 59-page volume that breaks down the experiences of those 600-plus ventures and draws some very clear and potentially actionable conclusions.

The authors of the report acknowledge that there have been previous attempts to isolate the winning strategies of successful Web outfits, including Eric Ries' Lean Startup and Steve Blank's Customer Development project. But, the report's authors write, none of these efforts have given would-be entrepreneurs much more than the initial patterns of how a start-up should be built.

"Most founders don't know what they should be focusing on and consequently dilute their focus or run in the wrong direction," the report says. "They are regularly bombarded with advice that seems contradictory, which is often paralyzing. And while start-ups are now gathering way more qualitative and quantitative feedback than they were just a few years ago, their ability to interpret this data and use it to make better business decisions is sorely lacking."

And that's why, the report continues, the Startup Genome team set out to build the "flexible framework that enables the integration of the principles, methodologies, and wisdom that have been discovered about how to create a successful start-up."

The major conclusions
The full 59 pages of the report are probably going to be read only by certain investors and entrepreneurs. But for those short on time, there are a number of main conclusions, starting with the idea that Web start-ups tend to pass the same kinds of development milestones, and that those that skip those steps tend to perform poorly. "They tend to lose the battle early on by getting ahead of themselves," according to the report.

Startup Genome Project

Here are some other major conclusions:

• "Founders that learn are more successful: Start-ups that have helpful mentors, listen to customers, and learn from start-up thought leaders raise seven times more money and have three-and-a-half times better user growth."

• "Many investors invest two to three times more capital than necessary in start-ups that haven't reached problem-solution fit yet. They also overinvest in solo founders and founding teams without technical co-founders, based on our indicators that show that these teams have a much lower probability of success."

• "Solo founders tend to stick to their original concept way too long and take much longer to reach scale. Compared with a founding team of two, they are 2.3 times less likely to pivot, and take 3.6 times longer to reach the scale stage."

• "Business-heavy founding teams are 6.2 times more likely to successfully scale up with sales-driven start-ups than with product-centric start-ups. Technical-heavy founding teams are 3.3 times more likely to successfully scale up with product-centric start-ups with no network effects than with product-centric start-ups that have network effects."

• "Balanced teams with one technical founder and one business founder raise 30 percent more money, have 2.9 times more user growth, and are 19 percent less likely to scale prematurely than technical- or business-heavy founding teams."

• "Start-ups that haven't raised money overestimate their market size by 100 times and often misinterpret their market as new."

In addition, the report divvies most Internet start-ups into six discrete development stages, each with several substages. "This creates a kind of directed tree structure and allows for more granular assessment by being able to pinpoint the main drivers of progress at each stage."

Those six stages are Discovery, Validation, Efficiency, Scale, Profit maximization, and Renewal.

Startup Genome Project

Does it make sense?
It's all well and good to compile a study like this, but the work done by the Startup Genome team is clearly only as good as its data. The question is, do their results matter in the boardrooms and offices of Silicon Valley and beyond?

To be sure, the report is just out, and it's too early to reality-test its conclusions. But the team did share an early version with a number of independent successful Silicon Valley investors and entrepreneurs, asking each to evaluate the report's advice.

At least two of those who've seen the report cautiously suggest they think the Startup Genome project is on to something.

"A lot of these things make sense," said Ilya Fushman, a principal at Khosla Ventures. "Are they actionable, or are they going to get someone to stop what they're doing and re-evaluate? And I think at the end of the day, if you look at the state of start-ups now, you have a very broad gamut of founders with varying levels of expertise, and at the very early stage, this can be helpful to start-up teams, especially with teams out of university."

Some, of course, may look at the major conclusions and think that there are few surprises. But whether that's true doesn't mean entrepreneurs can't learn from them.

The conclusions are "sort of obvious once you think about them," Fushman said, "but the question is, does everyone think about them? The answer is, maybe they do, maybe they don't."

And that's the key to the report, it would seem: to get people thinking about the process of building a company and to shine a light on the things that work and those that don't, and to give would-be founders some ideas of what they might do, and what they should stay away from, if they want to be successful.

Startup Genome Project

"I buy the [report's] major conclusion, that start-ups evolve in phases and that awareness of the phase you are in helps to focus the founder's attention and increases the likelihood of success," said Soeren Stamer, a board member at the European think tank the Lisbon Council and CEO of a start-up called Yokudo. "Some of the results are very helpful [for raising] red flags for a team: pivoting too often, not at all, etc. Again, that is helpful to create awareness and constructive discussion within a founding team."

But Stamer also pointed out that start-ups should be wary of thinking that if they follow the conclusions in the report to the letter they will automatically be successful. After all, the report is merely a framework, as Herrmann put it, and its data points more to averages of what has worked in the past. There is no substitute for real experience and expertise. As he put it, founders should be careful not to conclude, "we have to pivot two more times to be in the green zone."

What's next?
The initial report is out, but Herrmann said it's just the beginning. Though more than 600 start-ups participated in the survey, there are countless others in Silicon Valley and elsewhere that could contribute their own data, and the Startup Genome project is now going to work on a second, larger survey.

At the same time, Herrmann said he thinks the report is a "pretty significant map" that could give a lot of people "a better understanding of the world of entrepreneurship."

The team is also considering repackaging the data as a free product that would give investors a dashboard through which they could assess the start-ups they're funding.

Ultimately, Herrmann said, the project is about helping to pinpoint some key elements of successful start-ups in a measurable way that could help future innovators as they get started in business.

"We've reached a critical mass of people who have been describing certain patterns of entrepreneurship," he said, "and we're seeing entrepreneurship evolving as a science. For the first time, [we've started] to develop specific repeatable patterns for entrepreneurs building high-tech start-ups."