Three Observations on Designing Systems That Work for People

For decades, governments, banks and development organisations have invested in expanding financial services for smallholder farmers. Yet despite better products, more funding and improved technology, many programmes continue to struggle to achieve lasting impact.

That was the question we encountered while working with Zambia's National Credit Programme. The ambition was clear: strengthen access to agricultural finance for smallholder farmers across the country. Like many national programmes, it brought together government, financial institutions, technical partners and farming communities around a shared objective.

The more time we spent with farmers, however, the clearer it became that access to credit was only one part of a much larger system. Financial decisions were shaped by rainfall, markets, household priorities, trust, transport, previous experiences with lenders and the realities of everyday life. Improving one part of that system without understanding the rest rarely changed the outcome.

Looking back, three observations from that experience have continued to shape how I think about public systems.

People don't experience programmes. They experience systems.

From an institutional perspective, the work centred on improving a national credit programme.

From a farmer's perspective, it was one decision among dozens they made throughout the season.

Whether someone borrowed money depended not only on the product itself, but on weather forecasts, crop prices, family obligations, access to markets and confidence that they could repay the loan. Finance was simply one part of a much broader set of considerations.

Institutions often improve individual programmes while the people they serve continue navigating interconnected systems.

Better products rarely overcome poorly understood realities.

The programme generated many ideas for improving financial products. Yet some of the most valuable insights had very little to do with finance itself.

Farmers spoke about uncertainty, timing, relationships and risk far more than they spoke about interest rates or repayment schedules.

The leadership challenge therefore was not designing a better financial product. It was ensuring that institutional decisions reflected the realities people were already navigating.

Research became valuable not because it generated more information, but because it challenged assumptions that had quietly shaped earlier decisions.

Systems change when institutions learn together.

No single organisation could strengthen agricultural finance on its own.

Government shaped policy. Financial institutions managed lending. Agricultural extension officers supported farmers. Markets influenced livelihoods. Communities adapted continuously to changing conditions.

Progress depended less on improving individual organisations than on helping them see how their decisions affected one another.

Lasting systems change rarely comes from one institution performing better. It comes from many institutions developing a shared understanding of the system they are trying to influence.