More than training

“We just need to do more training”

I’ve heard this phrase many times, suggesting a way to bring about all sorts of changes in agriculture, particularly when it comes to small farmers in developing countries. Sometimes it's about new technologies or techniques. Sometimes it's about more sustainable practices. Sometimes it's simply about protecting the farmer, their workers, and their family.

But it's important to maintain perspective about what training can and can't do.

I think back to when I was leading workplace quality and safety programs, training was never the whole answer, it was the starting point. Real change also needed reinforcement, practice, promotion, and audit with follow-up action plans. Training on its own rarely moved the needle.

In agriculture though, we often lean on a single short session with small farmers — and if we're lucky, their hired labour too.

I remember one of these sessions with Indonesian cocoa growers. Two farmers were chosen to model full PPE, - coveralls, gloves, goggles, half-face respirators. When fully kitted out, the room started laughing. With this social cue, I knew straight away that almost no one in that audience was going to take the kit seriously, let alone wear it when in the field.

So, what’s the answer?

In the aid/development world we see more intensive training and extension programs being put in place, alongside various interventions. These aim to achieve changes in knowledge, attitudes and most importantly, practices.

I recently reviewed a cluster-randomised trial involving 539 Ugandan smallholder farmers. It found that training followed by SMS reminders, grounded in behaviour change theory, produced greater changes in knowledge and attitude – and in some instance practices – than training alone. There’s no doubt that training builds awareness, but broader engagement is more likely to change behaviour.

Commodity certification programs tend to do even better, because training is backed by audits and real financial consequences (both positive and negative).

The problem is that development and certification programs are resource-intensive, and by their nature they'll never reach every farmer who needs them.

In the agricultural inputs space, I'd argue that a strong engagement program, based in behavioural science, and tailored to the audience and specific technology, is probably the most cost-effective solution.

It is important to remember that engagement isn't training delivery — it's the purposeful application of communication and persuasion techniques toward a specific outcome, such as risk mitigation or benefit optimisation.

There's opportunity to leverage technology - social media, video platforms, apps, and AI agents trained on local language and context — to improve engagement outcomes.