The room was cool, but emotions were not. It was the last day in February, and Sorin Grama and Sam White, co-founders of Promethean Power, were in a critical meeting with their team at a dairy in the village of Karumapuram, in the Indian state of Tamil Nadu. The entrepreneurs had launched their company, based in Boston, to build the world's first solar-powered milk chilling system, which they hoped would transform the archaic Indian dairy industry. Grama and White, along with Rajat Gupta, their point man in India, and three of the company's engineers had gathered at the dairy where they were installing their first unit. R.G. Chandramogan, founder and chairman of the company that owns the dairy, was Promethean's first and as yet only customer and had just finished examining the system. He wasn't pleased. Chandramogan informed the Promethean team, assembled in a tiny, cramped office, that the system was entirely too large, too complex, and too expensive.

The rejection was a devastating blow. Grama, Promethean's CEO, was forced to take a hard look at his dream of harnessing solar energy to modernize India's dairy industry. He realized the company faced urgent questions if it was to survive. Grama would have to either somehow dramatically reduce the cost of building the hybrid solar-electric chilling unit—a long shot, at best—or change direction entirely by offering a simple, cheap chiller that would run only on electricity. The second option carried a stiff emotional price. "We had developed very good solar technology, and I did not want to give up without a fight," says Grama. "I had to ask myself: Am I going to be as passionate about this business if it is not about renewable energy?"

The finger pointing began. "We were asking, How did we miss this market signal?" Grama says. "The engineers were looking at me as if I misled them, asking if we should have made adjustments earlier." Meanwhile, the team's cohesion seemed to be crumbling. "You didn't take into account the feedback I gave you—you didn't listen to me," Gupta complained. White agreed that the company was on the wrong path. "We shouldn't be trying to shove something down the customer's throat," he said.

As the gripe session progressed, one of the engineers voiced the obvious, which Grama wasn't ready to hear. If cost was such a critical issue, why didn't Promethean, which had six employees at the time, simply strip out the solar piece of the system? After all, although that component provided a key feature for users who operated in areas with outage-prone electrical grids, it greatly added to the complexity and cost of the system. Removing it and making other modifications could bring the cost down from $12,000 a unit to about $3,000. And those in areas where electricity was not available or unreliable could simply use generators.

At midnight, with the town's restaurants and bars all shuttered, a dejected White, Grama, and Gupta sat on the curb outside their hotel debating what to do next. Gupta, a Harvard M.B.A., argued the low-tech route was the only solution. He had already heard from a number of other dairy producers in India who had told him the unit's price was too steep and that the system's capacity was not large enough. Promethean simply needed to dump the solar technology, and the sooner the better. Only then could it come up with a new product that would give dairies what they needed: cheap, reliable chillers with lots of capacity.

But Grama still wasn't convinced. Coming up with the new system, he argued, wouldn't be as simple as just pulling off the solar-energy components. The chiller would need to be radically redesigned. And the company didn't have a lot of cash to burn.

Grama also worried that both customers and investors would be less interested in the stripped-down version. Building the company around solar products, he argued, was the entire premise of Promethean. It was also his passion. Raised in Romania, Grama came to the U.S. when he was 18 and earned a master's degree in engineering and management from the Massachusetts Institute of Technology. At MIT, Grama, along with White, who was working for a Cambridge start-up, became interested in clean technology. They took second place in an MIT business-plan competition for a solar-powered turbine project. That led to their launching Promethean in 2007.

India's milk industry depends on a constant round of pickups by truck from rural collection stations that have no refrigeration. Promethean's chillers would allow for storage at those stations, preserving the quality of the milk for several days. The company's killer app was its solar-energy system, which would allow operators at the collection stations to chill milk without having to rely on India's notoriously unreliable electrical grid. Grama and White recruited Gupta in 2010 to oversee operations in the country.

By early 2010, the team, thanks to $500,000 in venture capital, had a prototype ready for installation. Chandramogan had agreed to buy the system and install it in his company's factory in Karumapuram, outside the city of Salem. And he said the company would buy 10 of the units if the prototype worked out. Grama and White knew the unit was still a bit expensive, but they figured as they ramped up production, they would find ways to lower the price.

They would never get the chance to find out. In their meeting with the dairy producer, who had first seemed so enthusiastic, Chandramogan balked when he realized installation costs would push the total price to close to $20,000 per unit. Promethean's would-be game-changing innovation suddenly seemed dead on arrival.

The next day at breakfast, though, White ran into an operations manager at the plant where the prototype was installed. He asked the manager what he would think of a cheaper, less-complicated system that lacked solar power. The manager liked the idea. Over the next several days, other managers chimed in, supporting a simpler, less-costly alternative.

About the same time, White checked in with some prospective Indian investors, and they, too, were on board with the shift. The only one who wasn't willing to see the way out of his company's dilemma, it seemed, was Grama. He was in his hotel room in Salem doing yoga, trying to relieve the stress of the past few days, when White called to relay the news about the likely backing from new investors. Even so, Grama told White he was still wrestling with giving up on solar energy. Would he on some level feel like a failure if he abandoned his dream? White countered that if Promethean's first product was a hit, the company would then have resources for R&D and could add a solar component later. "This could be a bridge," White said.

The Decision In early March, Grama left Salem to meet with Gupta in Promethean's offices in Pune, an hour's flight away. Grama still wasn't on board with the stripped-down chiller. But over the course of five days of discussions in the office and while on the road for business meetings, the two men began going over ideas for a new system. Gupta's sense of urgency and enthusiasm for the redesigned chiller finally persuaded Grama to defer his dream of solar power. "I saw that Rajat was very gung ho about the new approach and ready to take the lead on development," says Grama.

With only about three months of cash left, the company now had to quickly start over and redesign its product. It became clear that the only way to do so was by taking the painful step of letting go the company's two U.S.-based engineers and shifting the job to less-expensive workers in India.(Grama and White helped the engineers—including the one who first suggested ditching solar power—land new positions quickly.) That freed up enough cash to keep the company going for another six months.

The new unit was to be completed by late summer. And Chandramogan told Grama that given the new price of $3,000 per unit, he expects his company will buy 10 units as soon as they are available.

For Grama, the shift, although painful at first, now feels right. "I realized there are things you have to do and things you want to do," he says. "In this case, we have to do this cheaper system. I may want to create a solar system. And I think we will get there at some point. But this company has to survive and get a product on the market."

The Experts Weigh In

Listen to Your Customers
Just because the technology is feasible doesn't mean it should be in the first version of a new product. Customers may not be ready for it. Going forward, it is critical Grama and his team get feedback from those first users. Once customers are using it for a while, they may have new ideas that you haven't even thought of in terms of what future versions should look like. Small companies in particular can struggle with this, because they play things close to the vest and don't want to share information. But what users want may change as they start using the product.
Melanie Turieo
| principal | Cambridge Consultants, Cambridge, Massachusetts

Know Your Market
The advantage start-ups have over large companies is that they are nimble and have the ability to change direction quickly. But the downside to moving quickly is that you are constrained in the amount of time you can spend thinking about alternatives. It is possible there is an alternate set of customers or an alternate market in which the solar technology could work. But it is probably better to live to fight another day and have the potential to incorporate that solar technology down the line than to try to get up to speed on a completely different market.
Charles Eesley
| assistant professor | Stanford University, Palo Alto, California

Leverage Government Support
They had no choice but to go with the product their one customer wanted. If they decide to revisit incorporating solar into their product, there are a couple of variables to consider. First, there is the demand from the customer. There may be markets in which there is no electrical grid at all that may be more suited to a solar solution. The other side of the equation is what support there is from the regional or national government. Certain countries have very strong policies for helping companies get past the learning curve with new technologies. They should identify where there is strong government support and leverage that.
David Miller | executive managing director | Clean Energy Venture Group, Brookline, Massachusetts