Scaling seed innovations in the Ganges region

Across nine villages in India, Nepal and Bangladesh, the experience of the Rupantar project shows why agricultural innovations do not scale on technology alone. They scale when the system around them makes them timely, usable, inclusive and economically meaningful for farmers.

In Unishbisha, India, farmers face a narrow three-week window after the aman paddy harvest before the boro season begins. For years, much of this land stayed fallow. Today, more fields are being planted with mustard. The change was not driven by seed alone. It happened because a local NGO was active; scientists were reachable in person and through WhatsApp; the zero-till seeder, seed, herbicide and fertilizer arrived on time; a farmer producer company supported uptake; and a neighbour had already tried the practice successfully.

This is the lesson Rupantar brings to the CGIAR Scaling for Impact (S4I) program: scaling is not simply about expanding reach. It is about understanding the relationships, incentives, services, institutions and market links that allow an innovation to move across places and social groups.

Across the Ganges region, farmers are testing zero-till mustard, zero-till maize, poly-mulched chilli, multilayer arecanut farming, improved mustard, brinjal cultivation, dairy, goats and native poultry. The pathways differ, but farmers rarely begin by talking about the technology. They talk about someone local they trust, whether inputs and advice arrive on time, and whether there is a buyer, market or clear return.

With S4I support, the team behind the Rupantar project (which stands for 'Transforming Smallholder Food Systems in the Eastern Gangetic Plain') has begun to make these scaling conditions more explicit. The program’s contribution has been to bring scaling science into the project’s learning and decision-making, including through Scaling Scan training. This helped translate “scaling an innovation system” into a practical process for diagnosing bottlenecks across livelihood diversification pathways.

The Rupantar project is led by the International Maize and Wheat Improvement Center (CIMMYT) and partners. It focuses on enabling smallholder farmers to diversify crops, adopt climate-resilient technologies, and improve livelihoods across India, Bangladesh, and Nepal.

Using the Scaling Scan, Rupantar's Monitoring, Evaluation and Learning team examined local partners, service providers, market actors, input suppliers, machinery operators, and farmer organizations. The process identified where stronger coordination, incentives, or institutional support were needed for each pathway to move beyond a pilot. Rupantar now plans to use the Scaling Scan in pathway review workshops with farmers and partners, so decisions are grounded in a shared understanding of system constraints.

This is where S4I adds value. It is not promoting one technology over another. It is helping teams ask better scaling questions: What conditions must be in place for this practice to spread? Who needs to act differently? Which incentives are missing? Which groups may be excluded? What needs to change before replication makes sense?

The same logic is visible beyond Unishbisha. In Morang, Nepal, zero-till maize is moving from farmer to farmer because custom hiring, coordination, technical support and neighbour learning are in place. In Jhapa, Nepal, multilayer arecanut farming is gaining ground because training, planning and institutional support help farmers manage orchards differently. Scaling happens when the enabling environment reduces the frictions that usually stop adoption.

Inclusion is also part of the system. In Sunsari, Nepal, the dairy pathway grew from male migration leaving many women managing crops, livestock, household work and childcare. Improving dairy was not only about milk yields. It was also about reducing women’s daily labour. Similar lessons appear elsewhere: goat rearing, native chicken rearing and poly-mulched chilli are all shaped around women’s labour, mobility and schedules. These are design questions that determine whether a pathway fits people’s lives.

Rupantar also shows that scaling requires adaptive learning. Some promising pathways still face practical constraints. Poly-mulch may remain out of reach where credit or input supply systems are weak. A zero-till seeder may miss the season if calibration and operator support are not available. Varieties may not spread if farmers cannot access them reliably. Dairy pathways can lose momentum when cooperative payments are irregular. Advisory systems can miss farmers when information is not delivered in ways they use and trust.

Viewed through the S4I lens, these are not failures. They are scaling signals. They show what the innovation system still needs to provide before a pathway can move beyond replication. For Rupantar, S4I has helped surface bottlenecks earlier, discuss them with partners, and adapt scaling choices more deliberately.

Three lessons stand out. First, scale the innovation system, not the practice or the innovation alone. Technologies move when services, institutions, and market connections move with them. Second, build inclusion into the architecture from the start and not as an afterthought. Women’s time, mobility and decision-making are design inputs, not afterthoughts. Third, invest in the enabling environment. Input logistics, machinery services, governance, advisory feedback loops and credit instruments often decide whether a practice moves or stalls.

For S4I, Rupantar offers a practical example of scaling science in action. It shows how the Scaling Scan can be embedded into monitoring, reflection, and partner engagement, rather than treated as a separate study. It also shows why scaling agricultural innovation requires investment in the relationships, services, institutions, and feedback mechanisms that allow technologies to stay, spread, and benefit the people they are meant to serve.