Two self-help groups, six months of training, and then nothing. That's the short version of what Hemant Kumar told the room about his own PhD fieldwork in northern India: he and his colleagues had spent half a year teaching women how to run a sanitary pad-making machine, the production line, the bookkeeping, the basic marketing. When he went back to check on them, both groups had stopped.
He didn't blame the machine. He listed three reasons instead, each one unrelated to engineering. The men in the community had stopped the women from working at the production unit, because making sanitary pads still carried real stigma. Infrastructure was unreliable: electricity cut out for seven or eight hours a day, exactly the hours women would otherwise have been free to work. And the raw material travelled two thousand kilometres to reach them, heavy and expensive to ship, which meant the finished pads cost more than the branded versions already sitting on shop shelves.
"On the one hand there is a success for the innovator," Kumar told the group. "On the other hand, it's failed." He didn't pick a side. He left it open, and the rest of the session never quite closed it either.
Borrowed Tools, Borrowed Time
Before any of the history or the case studies, Kumar asked the room a few direct questions. Who had ever solved something with a low-cost workaround? Modified a tool, repurposed an object, fixed something with whatever was lying around?
The answers came quickly. One participant described hanging a shelf without a drill, just a hammer, because that was what she had. Another teaches English in Vietnam and shows up to class with whatever is in her bag: a cooking pot that becomes a goal for throwing paper balls into, dog toys that turn into props for practising vocabulary, a baseball bat and a balloon on days when the lesson needs energy. Over time, she said, she'd realised that most of what was already in her house worked better in the classroom than the expensive materials she'd started out with.
Kumar thanked the group and moved on: none of this happens only in workshops and labs. It happens constantly, outside any formal system, whether anyone names it or not.
What Gandhi Already Knew
Kumar took the group back nearly a century, to Gandhi and Tagore. Every society has its own tradition of innovation, he argued, and grassroots creativity has long been part of how communities develop.
Gandhi wanted a spinning wheel that village families could build and use themselves: cheap, simple, productive, and comfortable enough for a woman to operate for long duration. He ran a design competition to improve it. When a serious engineering firm finally submitted something, the new version was faster, yes, but also heavier, more expensive, and less durable. Better by an engineer's standards. Worse for the person actually meant to use it. Gandhi rejected it and designed his own version himself.
Kumar took the group through fifty years of history at speed, and the conclusion was the same each time: well-meaning people kept trying to improve things for communities by bringing in solutions from outside, and it didn't work.
It took until the late 1980s, when Prof. Anil Gupta and the Honeybee Network in Gujarat started building a network of researchers, farmers and organisations to document what ordinary people were already building and solving on their own, for anyone to take a different principle seriously: that innovation had to come "at, for and from" the grassroots itself, not arrive from outside dressed up as help.
To show what that looks like in practice, Kumar had chosen four cases with care. Not success stories, not failures, but four innovations that each raised a different, unresolved question about what it actually means for a grassroots idea to work.
When Help Makes Things Worse
The cotton-stripping machine put that warning to the test almost immediately.
Its inventor had no technical training. He worked as an electrician and noticed how brutal cotton processing was for the women and children doing it by hand. His first prototype, built with family and friends, worked roughly, though the parts wore out fast. His second, two or three years later, solved that and introduced new problems. Eventually an industry contact connected him to India's National Institute of Design, which sent an intern from France to help redesign the machine.
What came back was unusable. The intern, working from formal training the innovator had no way to follow or argue with, produced something heavier and more complicated than any farmer could actually take out to their fields and use. Exactly the trade Gandhi had rejected a century earlier: faster and more sophisticated, but worse for the person who actually had to carry it. The innovator rejected it flat out, too: it fixed one issue and created a worse one in its place.
Seven prototypes and nearly ten years later, he had something that worked, until it caught fire after seven or eight hours of continuous use, a fault that took him another year to trace. No bank would lend to a homemade machine, so he borrowed from friends and early grassroots support organisations, filed a US patent with their help, and later let it lapse when the annual cost outweighed what it was protecting.
The machine still runs, and farmers in the region still use it. But it's built for one specific local cotton variety, and as hybrid seeds spread, fewer farmers grow that variety at all. Ten years of work, a fire, a patent gained and dropped, and the problem it was built to solve kept changing shape.
The Upgrade Nobody Could Afford
Two brothers in northeastern India needed to lift water from a river below their fields. Bamboo was the one material they had in real supply, so they built a windmill out of it, scrap parts filling in the gaps, for roughly seven to eight thousand rupees. Neighbours wanted one too, and the brothers started selling copies for ten to twelve thousand rupees apiece.
The Indian Institute of Technology was assigned to fix the design scientifically: the motor wasn't aligned properly, the whole design lacked any grounding in actual wind mechanics. They offered to help, with a patent and wider distribution as the eventual payoff.
The brothers tried to keep up their end of that bargain, travelling two to three hundred kilometres to the institute and sometimes waiting a day or two, time they couldn't spare without a steady income. Eventually the trips stopped, and the final "scientifically sound design" was finished without them: bamboo replaced by iron, the motor properly engineered at last.
The new version was better in every technical sense, and ten times more expensive: sixty to seventy thousand rupees against the original seven or eight. The farmers who had wanted the bamboo version stopped buying it. They simply couldn't afford what it had become. The people the thing was built for were no longer in the picture.
It didn't disappear entirely. A corporate social responsibility programme eventually placed the redesigned windmill with the Agariya salt farmers of Gujarat, a community that needed exactly this kind of pump and could reach it through corporate funding the original village farmers never had. Kumar put the paradox to the room without resolving it himself: is this a success, because the technology found people who genuinely needed it? Or is it a failure, because it severed the windmill completely from the place and the people it was built for in the first place?
When the Innovation Reaches Home
The third case brought the question of success and failure closest to home, in the most literal sense.
The inventor of the sanitary pad machine had no formal education and no resources to speak of. He started by asking his own wife why she wasn't using the pads already available in stores. She told him it was none of his business. He kept asking. Eventually she explained that between the women in the household, they simply couldn't afford them. He took a pack apart, found the raw material cost almost nothing, and spent years trying to build something cheaper himself. He tested early prototypes with materials borrowed from a butcher's shop, was turned away by medical students who told him men had no business asking women to test such products, and lost his standing in the village along the way, before finally finding the right material through a contact in the packaging industry.
His model was simple: sell the machine to women's self-help groups, who would then produce and distribute the pads locally, generating both affordable sanitary products and income at the same time. Not just a product, but a livelihood.
Kumar had tried to replicate that model himself, during his PhD. The result was the story this article opened with: two self-help groups, six months of training, and both shut down, for reasons that had nothing to do with the machine.
The inventor is celebrated and award-winning. The model, in many places, still isn't working.
The One That Just Worked
Pochampally Ikat is a traditional weaving technique from Telangana in southern India, where yarn is folded, bound and dyed in precise patterns before it is woven, so the design emerges from within the fabric rather than being printed onto it. A single sari can take eight or nine hours of work a day for two weeks to a month, with the body held in one sustained position the entire time, the kind of work that wears down knees and backs over years.
One innovator watched his own mother go through exactly that. He left his job as a city labourer, came home, and spent two or three years building an electronic device that compressed the yarn-preparation process into a single day. No technical schooling, same as nearly everyone else in this presentation.
Set against the other three cases, this one had no asterisk attached. No abandoned patent, no community priced out, no group that quietly shut down. Just a problem someone had watched his own mother live with for years, and solved.
Into the Breakout Rooms
After the four cases, Kumar sent the group into four breakout rooms, one per case, with questions he was careful not to answer himself: what challenges did this innovation face, what made it frugal, and can you call it a success?
The group working on the cotton-stripping machine noticed something Kumar hadn't raised. The machine doesn't replace the people doing the work: it still needs someone to feed it, clean it, and collect from it. In a context where labour is one of the few resources available, that's a feature rather than a flaw. Conventional mechanisation eliminates jobs entirely, making poverty worse. The group also suggested earlier collaboration with people who had electrical or mechanical knowledge might have shortened the ten-year process, without the formal partnership that went wrong with the French intern.
From the bamboo windmill room came a phrase that stayed in the air after they'd finished speaking: the temporal nature of success. The windmill had been genuinely successful for the context it was built for. The problem wasn't the technology, it was that the conditions of success changed the moment anyone tried to scale it. What makes something work in one place, for one group of people, at one cost, stops working the moment any of those conditions shift.
On the sanitary pad machine, the feedback cut across the technical discussion entirely. Even after a working solution exists, thousands of women are still using cloth. Awareness, trust, and social permission matter as much as the product itself. A machine that works is not the same as a problem that's solved.
The fourth room had been given the least information of the four, so they looked it up themselves, and found practical questions about maintenance and running costs that hadn't come up in the presentation. Those details led somewhere broader. Looking at all four cases side by side, the group noticed that every single one of these innovations had taken many years to develop. What happens, they asked, if the context the innovation was built for has already shifted by the time it's ready?
Whose Success Is It?
Kumar closed with two points he wanted the group to carry forward, both of them unresolved.
First: policymakers keep applying the same rules to grassroots innovations that they use for the formal sector, including intellectual property rights. But a patent only works if you can stop others from copying your design. Many grassroots innovations are simple enough that anyone nearby can replicate them just by watching. Filing a patent costs money every year, and if the design can be copied anyway, what exactly does it protect?
Second: there's no template here. Four cases, four entirely different timelines, motivations, and relationships to the people they were built for. Even partnering with serious scientific institutions, the kind of help that sounds like it should always be good news, sometimes ended up stripping an innovation of the exact thing that had made it work. Context matters in grassroots innovation, more than any design does, and supporting it takes approaches that stay close to local conditions, knowledge systems, and community needs.
Who gets to decide whether a grassroots innovation has succeeded? Kumar left that question open, on purpose.