The founder of a beloved productivity app thinks the startup model is broken - here's how he's trying to keep the tech industry from 'making the same 10,000 mistakes over and over again'

• Phil Libin, a longtime entrepreneur and the founder of Evernote, thinks the technology industry isn't doing a good job of promoting innovation.
• He's particularly critical of the startup model, because its focus on building companies leaves out worthy ideas and talented technologists.
• Libin new startup, called All Turtles, is hoping to disrupt that model by funding the development of ideas and turning them into products before their developers teams have to worry about building companies around them.
• All Turtles is intended to promote innovation globally; it already has offices in three countries, with more planned.
• Some of the first ideas it's fostered will effectively graduate from its process this fall.

---

If you ask Phil Libin, the tech industry doesn't do a very good job these days at promoting innovation.

They may seem a strange thing to say, given that the industry has a reputation for being the most innovative in the country, if not the world. But Libin's not just some crank.

He's been in the industry for more than 20 years, founding a succession of successful startups - including note keeping app Evernote, the one he's most known for - and working as a venture capitalist. He's seen how the business works from the inside.

"The structure for making innovative tech products … is old and inefficient and broken in lots of different ways," Libin, All Turtles' founder and CEO, told Business Insider in an interview earlier this month.

Libin's solution is a new kind of company called All Turtles that he's designed specifically to address the problems he sees in tech innovation, particularly in the area of artificial intelligence. If his bet is right, the result could be not just a bunch of new AI-based products and services, but a better way to foster and develop new technologies.

"If this works, you can copy and spread it" widely, said Abhishek Nagaraj, an assistant professor at the Haas School of Business at the University of California, Berkeley, who focuses on entrepreneurship and innovation. He continued: "The potential is tantalizing."

Neither of tech's two innovation models is working well

The way Libin sees things, the tech industry has two basic models for promoting innovation - and each has its own shortcomings.

The first, he says, is through the research efforts of the giant tech companies, including Facebook, Microsoft, Google, and Apple. The problem with that model is those companies tend to be very conservative, focusing on small-bore innovations that won't disrupt their core businesses. When they are more forward thinking - such as Google with its X lab - they tend to be focused on moon shots that, by definition, have will likely have little near-term impact on the real world.

The second model, by his reckoning, is built around startups. While entrepreneurs are often more willing to focus on disruptive technologies than the tech giants, innovation in the startup world is thwarted by the very focus of trying to create new companies. That fixation is fundamentally flawed.

By necessity, that model self-selects for founders who are good at fundraising and managing, rather than those skilled at solving important problems, Libin said. By contrast, technologists with good ideas or product development skills can fail simply because they have no talent for building companies.

What's more, because entrepreneurs and startups are frequently working by themselves, they often don't and can't learn from the mistakes of others. Instead, they end up reinventing the wheel over and over again at the expense of time, money, and often the ideas themselves.

"It's a total s--t show," Libin said. "Like everything's on fire, no one knows what they're doing, you have to scrounge around for resources and everyone makes the same 10,000 mistakes over and over again."

But there are other problems with the startup model, Libin said. Again, because it's so focused around companies, it tends to exclude innovative people and ideas from areas of the world outside of Silicon Valley that don't have the kind of ecosystem found there for funding and supporting startups.

And with the growing size of funding rounds, Silicon Valley has become increasingly focused on ideas that have the potential to be billion dollar companies or larger. That's started to leave out practical innovations that are worthwhile but don't have that promise.

"Not everything has to be a multibillion dollar outcome," he said.

Libin thinks he has a new and better way to jump start innovation

Libin thinks there's a better way. With All Turtles, which he launched last year, he aims to bring together talented technologists working in AI and help direct them to solve particular real-world problems with the technology. All Turtles will fund the technologists, assist them in developing their solutions, and then help them figure out how to go to market with them.

It's already a global effort. In addition to its San Francisco headquarters, All Turtles has offices in Tokyo and Paris to fund projects in those cities. It plans to open a Mexico City office next year. Within five years, Libin would like to have outposts in eight cities, including ones in Eastern Europe and Africa and potentially in India and China.

Libin plans to have each All Turtles office oversee 10 local projects at a time. Each year, the offices would each graduate about three projects and replace them with three new ones. Libin wants to keep the total number of projects per office fairly small to keep things manageable and to maintain quality control.

All Turtles is flexible when it comes to both originating ideas and eventual exits

The projects will come from any of three places.

They could be ideas generated in-house by Libin and the All Turtles team. They could come from outside corporations who are looking for help in developing a specific idea or solving particular problem. Or they could come from individual technologists or teams who are looking for help attacking a particular problem. Libin expects that within each office, new projects will roughly be split evenly among the three different sources.

In most cases, All Turtles will fund the development of projects by itself. In other cases, it will bring in projects that already have some venture or corporate funding. Regardless, the plan is to develop the projects in-house to the point where there's an actual product that's launched and has traction in the market before worrying about where the project will end up.

All Turtles establishes all the projects as separate companies within its corporate structure so it can easily spin them off as standalone startups with their own sources of external funding, as circumstances allow. But All Turtles is intentionally designed to accommodate other possible outcomes for ideas that make it past the development stage.

They could be acquired by outside companies, for example, particularly those that partner with All Turtles to develop the ideas. helped fund them. Or All Turtles could keep them in-house and run them itself, offering their products and services as its own.

"We're indifferent toward companies," Libin said. "If some of these products then become independent companies, fine, that's the best outcome, but they don't have to."

He continued: "We don't have the startup fetish that permeates the industry."

One of its first products is a chatbot designed to combat sexual harassment

The company already has "several" projects that it expects to graduate from its program in the fall and get outside funding, Libin said. One of them, a product called Spot, is a chatbot designed to make it easier for victims of workplace sexual harassment to document and report their experiences. The system relies on natural language processing and an interaction model that's intended to encourage victims to recount their experiences as accurately as possible.

Launched earlier this year, Spot is free for individuals to use. All Turtles charges corporations who want to build it into their human resources processes.

"By the time Spot gets the first penny of external funding, it will have a launched product with thousands of users, with actual, paying large companies, with revenue, with traction, with tons of media exposure," Libin said. "It will be a real thing before we ask anyone for money."

Although All Turtles is designed to fund innovative tech ideas, it's set up as an operating company, not as a venture fund. Ironically, while Libin is trying to pioneer a different model for innovation, All Turtles is itself a venture-funded startup. It's raised $20 million in financing in a series A round, with much of the money coming from Salesforce Ventures.

And while Libin thinks there's a need to fund and develop technology innovations that aren't necessarily going to turn into the next billion dollar companies, he has big ambitions for All Turtles. He's hoping All Turtles itself will eventually become a public company worth billions of dollars.

All Turtles has plenty of company in trying to jump start innovation

Libin isn't the first person, of course, to recognize that the tech industry has a problem in funding and developing new ideas and technologies. In the 1990s, entrepreneur Bill Gross set up Idealab as an incubator to develop ideas for web-based companies, many of which he later spun off or sold.

Later, Y Combinator and 500 Startups were launched to help entrepreneurs establish themselves and get their startups off the ground. More recently, Android founder Andy Rubin launched Playground to help incubate innovative ideas, particularly in technology hardware. Meanwhile, numerous venture funds and organizations have launched with the express purpose of trying to help entrepreneurs - including those in other countries - build companies and develop their ideas.

Neil Cohen has had a front row seat to this movement. In the 1990s, he cofounded Camp 6, an early startup incubator that fizzled out with the dot-com bust. More recently, he's worked with other incubators and accelerators from around the world.

"I can appreciate the process and the thought [Libin's] putting in to be better at it," said Cohen, now an independent marketing consultant. He continued: "Great for him to have this organization that's going to try to create and accelerate these ideas. But I would say that there other places doing great work."

Separating product development from corporate development could prove problematic

What's more, Libin's idea of trying to promote innovation outside of the typical corporate structure could prove troublesome. Products are often shaped by the business model and goals of the companies that develop them, noted Haas' Nagaraj. A company that was looking to compete with Facebook head-to-head would likely design its product very differently than a company that aspired to be acquired by Facebook.

"How successfully you can decouple [product from business model] is a question for me," he said.

For his part, Libin recognizes that there have been other efforts to try to overhaul technology innovation - and that his might not work out. The company's name reflects that realization, he said. It refers to a tower of turtles. Libin's company is building on the efforts of those who came before it, and others will build on his and learn from his mistakes.

"There's a rich history of people trying to change the way innovative products get made and make who gets to make them more inclusive," Libin said. "We're the next level up from that."