There is a specific moment in most agricultural technology projects that nobody plans for: the month after the external support ends.
Up to that point, everything works. Attendance is good, the equipment functions, the reports are positive. Then the visits stop, and within a season a meaningful share of the adopters have quietly returned to what they did before.
Adoption is not a moment, it is a maintenance cost
We tend to measure adoption as a binary event — the producer received the technology and used it. But every technology carries an ongoing cost that is invisible at handover: calibration, consumables, repair, the mental overhead of remembering a new step, and the social cost of doing something the neighbours are not doing.
If that recurring cost exceeds the recurring benefit as the producer experiences it — not as the project calculates it — the technology gets dropped. Not from resistance. From arithmetic.
Three questions worth asking before the pilot
Who repairs it in month fourteen? If the answer involves the project, the technology is on loan, not adopted.
What does the producer stop doing? New practices are almost always added to a workload that was already full. If nothing is removed, something will be.
Who else in the community can explain it? A practice that lives in one person’s head disappears when that person migrates, gets sick, or simply loses interest. Adoption becomes durable when it becomes local knowledge rather than individual knowledge.
What this changes about design
It moves the design question from “is this technology good?” to “is this technology maintainable here?” — which frequently favours the simpler option. The most sophisticated tool I have seen abandoned was abandoned because it needed a part that took six weeks to arrive.
I would rather deploy something that captures seventy percent of the benefit and can be fixed with what is available locally. That is not a compromise on ambition. It is what ambition looks like when the timeframe is measured in years instead of project cycles.