Martin Fisher, co-founder and CEO of KickStart International, went to rural Kenya in 1985 as a Fulbright scholar to study poverty.

Intending to go to Africa for ten months, he stayed for 17 years.

At first, he worked on programs funded by governments and foreign aid agencies that would dig water wells and donate pumps and equipment to poor farmers.

But those programs had a major drawback.

"The first time the pump would break down, no one would fix it, because nobody owned it," says Fisher, who earned a  Ph.D. in mechanical engineering from Stanford in 1985.

Worse still, there would be nowhere to buy a replacement, because the local suppliers had been put out of business by the influx of free equipment. 

Fisher soon realized that the answer to lifting people out of poverty was to help people develop a business idea, then find them the equipment they needed to run it. 

"People in rural Africa are no different than people anywhere else... what they need more than anything is a way to make a living," Fisher says.

So KickStart sells low-cost, human powered irrigation pumps and other equipment to farmers that allow them to transition from raising subsistence crops that are harvested once or twice a year to growing commercial crops grown year-round.

Farmers can triple their income in this way, and KickStart has helped 65,000 families pull themselves out of poverty. 

For his efforts, Fisher earlier this month received a $100,000 award, the Lemelson-MIT award for sustainability.

Fisher is now raising $25 million to lift another 400,000 people out of poverty.

The money is used to create a market and an industry where none has existed through extensive marketing, advertising and education programs. 

Fisher points out that even in highly-developed, wealthy societies, it takes a long time for new technologies to become widespread.

Eventually, KickStart's goal is to create a locally-based, viable market that doesn't require subsidies.

"We can walk away from that market," then move to another region or country and start the process again, Fisher says.