100,000 heads are better than one

A posting on InnoCentive.com caught Ed Melcarek's eye during one of his regular visits to the Andover-based website, which offers bounties of up to $100,000 for solving scientific problems: The challenge was to devise a more efficient way for getting toothpaste ingredients into a tube.

The Canadian engineer sent an e-mail that suggested putting a positive charge on fluoride powder, then grounding the tube. Colgate-Palmolive Co., the InnoCentive client that had posted the problem, liked the idea, Melcarek said, and he earned $25,000 for a few hours of work. ``It's a beautiful way of doing business," he said.

It's also one of the latest innovations in research and development.

A number of corporate chieftains and academic researchers argue that companies that continue to rely mostly on an in-house R&D department risk being out-innovated by more nimble rivals that use online global networks to focus the power of thousands of minds on their most nettlesome problems.

InnoCentive Inc. is one firm with such a network. Other players in this space include yet2.com of Needham and NineSigma Inc. of Cleveland.

Launched in 2001 as an independent business venture of the pharmaceutical firm Eli Lilly & Co. InnoCentive charges clients annual fees of roughly $80,000 for access to a network of scientists that stretches from Cambridge to Kazakhstan. A private company, InnoCentive does not disclose revenues.

InnoCentive describes corporate clients as ``seekers" and scientists in its network as ``solvers." Anyone can visit the site, but to submit a solution, a would-be solver must first register. Problems that seekers post at InnoCentive.com are called ``challenges." Solvers are paid only if seekers accept their solutions, and solvers must sign over the solution's intellectual property rights.

InnoCentive's network of solvers now numbers 100,000, with about 40 percent in China or the United States, said Ali Hussein , chief marketing officer.

``Clients are excited to have access to all these brains," he said.

Sometimes, the answer to a company's problem already exists; by tapping into various global networks, a company can find a ready-made answer and save itself the expense of reinventing the wheel.

That happened with Pringles potato crisps. In a recent article in the Harvard Business Review, Procter & Gamble Co. executives Larry Huston and Nabil Sakkab recounted how P&G executives thought they could boost sales if trivia questions could be printed on the crisps. But they were unsure of how to do that cost-effectively.

Instead of relying solely on its R&D department, P&G also sought solutions through global networks of scientists it uses. P&G ultimately found a professor in Italy who had devised an ink-jet method for printing images on cookies with edible dyes. By adapting that process to Pringles, P&G was able to dramatically reduce the time and expense of launching this product upgrade.

It's not an unusual outcome for companies that collaborate with such networks. Huston and Sakkab said 35 percent of P&G's new products today have elements that originated outside the company, up from 15 percent in 2000.

InnoCentive can also point to successful outcomes. Karim R. Lakhani, now a Harvard Business School assistant professor, reviewed company results and found that 30 percent of challenges not solvable in-house were solved by an InnoCentive solver.

What's important is that InnoCentive doesn't act as a matchmaker, Lakhani said. Instead, network scientists self-select the challenges they want to tackle.

One drawback to in-house R&D is that many researchers can be trained in the same discipline, and they tend to look at a problem in the same way and operate within the narrow sphere of their expertise. That uniform mindset can be an obstacle.

But a challenge posted on InnoCentive's website is exposed to a wide variety of scientists likely to approach a problem from different angles.

Such was the case with the 2004 toothpaste challenge. Melcarek, the Canadian engineer, proposed a physics solution to what others might have regarded as a chemistry problem.

One obstacle companies face in adapting to this new R&D approach is getting beyond the bias that in-house solutions are superior to an outsider's, said NineSigma vice president Paul Stupay .

``It's a cultural shift," Stupay said. ``Companies have to get past the barrier of `not invented here.' "

Few outcomes are as clear-cut as the Pringles example.

Often, seekers are looking for a small piece of a solution to a much larger puzzle.

And challenges are posted anonymously, so as not to tip off a client's competitors.

As a result, many solvers never know the identity of the client/seeker or what happens to their solutions.

Colgate, for example, declined to comment on whether it was ever an InnoCentive client or on Melcarek's account of improving toothpaste production. InnoCentive officials named Melcarek as a scientist who had successfully solved a client's problem.

Another solver identified by InnoCentive was David Bradin, a North Carolina patent attorney with a master's degree in organic chemistry. In 2002, while visiting InnoCentive's site, he spotted a challenge seeking a more cost-efficient way to mass-produce a certain acid. Based on a reaction he had observed while he was an organic chemist, Bradin thought he knew the answer. A few months later, InnoCentive informed him that his e-mail solution had been accepted.

``It was the easiest $4,000 I ever made," he said.

But Bradin never learned the identity of the seeker company.

Describing the range of talent that InnoCentive puts at P&G's disposal, Huston mentioned during a telephone interview that one InnoCentive solver of a P&G challenge was a North Carolina patent attorney.

``He's a lawyer by day and a chemist by night," Huston said. ``He did chemistry while his wife read romance novels."

The description matches Bradin's. But citing proprietary reasons, P&G officials declined to confirm Bradin was the solver.

Such secrecy is needed to protect clients and the intellectual property rights they might obtain, InnoCentive argues.

But such secrecy may also hinder greater innovation, said Eric von Hippel , a professor at MIT's Sloan School of Management and the author of ``Democratizing Innovation."

``Potential solvers don't see one another's answers," he said.

``One advantage to this method is that it keeps the intellectual property clean so they [seekers] know who has come up with an acceptable answer.

``The downside is that Fred might be able to help Joe come up with a better answer if they were allowed to collaborate during the solving process.

``InnoCentive is developing a very interesting new model for solution search and doubtless will improve it further over time."

Chris Reidy can be reached at reidy@globe.com.