Crossing the innovation ‘valley of death’


A growing movement is pairing researchers and farmers to bring discoveries to the field faster and to make sure those innovations meet real-world challenges.

Nicolas Tremblay is a research scientist with Agriculture Canada at Saint-Jean-sur-Richelieu near Montreal. He said most research follows a traditional model of experimentation and statistical analysis developed in the 1920s at the Rothamsted agriculture research centre in Britain.

“This was conceived a century ago and now we have realized limitations of this approach,” he said. “Also it was developed at a time when we didn’t have access to a wealth of data from the digital technologies we now have.”

Over the years, the system revealed a bias toward publishing in scientific journals rather than delivering results for producers — basically, it was farming papers. Production objectives or needs of the farmers were not needed.

“There’s been gradually a bias being developed so that research was conducted eventually for publication more than for the farmers,” Tremblay said.

The approach has often led to something called the “valley of death” for ag innovations. The latest knowledge may be published, but it takes considerable effort and funding to bridge the gap between scientific journals and farmers’ fields.

Tremblay is one of a small but growing number of researchers worldwide who are championing another approach. He is co-chair of the first conference on Farmer-centric On-Farm Experimentation coming up in October in Montpellier, France (online offerings are also being planned).

Funded through the Organization for Economic Co-operation and Development, the event aims to combine two deep pools of expertise: farmers and researchers.

The key difference with farmer-centric on-farm experimentation is the starting point: the farmer, their needs and their expectations from the research. From here, expertise and resources are selected to achieve the objectives.

“The farmer is central to the whole thing, which has not been the case most of the time in the past,” Tremblay said.

Knowledge from institution-based research, no matter how well designed, can be hard for farmers to adopt. Tremblay gave the example of his own work on nitrogen rate treatments. While they might be perfectly valid for his Ag Canada research fields, they may not apply well to any given commercial farm, with their different soil types, climate and agronomic practices.

“We need to address the context as well as the results,” he said. “To be accepted by farmers, we need to operate in their context.”

The approach is rather new. There is some activity in Australia and France. Agriculture Canada’s Living Labs initiative is designed to do similar work in this country.

The approach has raised a few eyebrows both of intrigue and skepticism. Tremblay said questioning the established order has created a bit of tension, but this has become further impetus for the conference to explore and compare best practices so far around the world.

“That’s an opportunity to see what we need to keep in the old tradition and what we need to change,” he said.

Obviously, farmer participation is key to the success of the Montpellier conference and the farmer-centric research movement as a whole. To this end, organizers are hoping for strong producer attendance as well as participation in four online workshops coming up in May. These are organized by four themes: value creation; people and processes; data and analytics; and policy linkages. Tremblay said all sessions will be in English, although due to the location, much of the “hallway chatter” is likely to be in French.

More information is available at