Indian agriculture has always suffered from an information problem. Governments often did not know precisely what was being grown where, farmers had limited information about weather, soil and markets, and financial institutions struggled to assess agricultural risk. Agricultural policy, therefore, often had to work with incomplete, delayed or fragmented information.

That is beginning to change. Satellite imagery, remote sensing, geographic information systems, drones, digital crop surveys and other technologies are making it possible to observe agriculture at a much finer level than before. We can increasingly identify where crops are being grown, assess their condition, map soils and water resources, monitor weather-related risks and link this information with farmers, land records and markets. This is an important technological change. But its real significance lies elsewhere.

The value of agricultural information begins when it changes a decision.

A satellite image, for example, does not by itself increase agricultural productivity. It becomes economically valuable when the information derived from it leads to a better decision about irrigation, fertiliser, crop insurance, credit, procurement or logistics.

Consider water. If remote sensing can identify areas experiencing crop stress, that information can help target irrigation more precisely. The benefit then comes not from the satellite itself, but from using less water, reducing crop stress and potentially improving yields. The same principle applies to fertiliser. Better information about soil and crop conditions can allow inputs to be applied more precisely, reducing waste and costs while improving resource efficiency. But this is where the challenge begins.

Information does not automatically become action.

A useful way to look at the emerging digital agricultural ecosystem is as a chain:

Data → Information → Intelligence → Decision → Implementation → Economic outcome.

There can be a gap at every stage. The first is a data gap. Do we know what is happening and where? Which crop is being cultivated on which plot? What is its condition? What are the soil and water conditions? Who is actually cultivating the land? New digital and geospatial systems are helping address many of these questions.

The second is an information-to-intelligence gap. Having data is not the same as understanding it. A crop map may tell us where wheat is being cultivated. But can that information tells us how much wheat is likely to be produced, what the crop may require in terms of water and inputs, or where production risks are emerging? This is where analytics, artificial intelligence and the integration of multiple datasets become important.

The third is the intelligence-to-decision gap. Even when we know what is happening, does somebody act on that information? Suppose a system identifies water stress in a particular area. The economic value arises only if the information reaches the appropriate decision-maker in time and results in a change in irrigation behaviour or water management.

The fourth is the decision-to-implementation gap. A digital system may recommend a particular input, irrigation schedule or crop practice. But does the farmer have access to irrigation? Is the required input available? Can credit be obtained? Is the extension system capable of supporting the farmer? Digital information cannot substitute for missing physical and institutional infrastructure. This is perhaps one of the most important points in the debate over digital agriculture. Technology can improve information, but it cannot by itself remove every constraint on agricultural production.

The final gap is between implementation and economic outcomes. A digital crop survey may improve the government's ability to identify beneficiaries. Better crop-loss information may make insurance assessment faster. Digital land and crop records may make agricultural credit easier to process. These are valuable improvements. But they are largely improvements in administrative or transactional efficiency.

The larger economic question is whether they ultimately affect the farm economy. Are input costs falling? Are yields increasing? Are crop losses declining? Is water being used more efficiently? Is access to credit improving? Are farmers receiving better prices? Is income becoming more stable? This distinction between administrative efficiency and economic transformation deserves much greater attention.

The possibilities also extend beyond the farm.

Better information about the location, condition and likely harvest of crops can improve procurement planning, storage, transportation and processing. A retailer or food processor could potentially use agricultural intelligence to anticipate where supplies will come from, when they will reach the market and how much infrastructure will be required.

Thus, some of the economic value of agricultural intelligence may emerge well beyond the farm gate. And this raises another important question:

Who captures that value?

Better information can reduce information asymmetry. But it can also create new forms of asymmetry. A farmer may gain from knowing more about soil conditions or weather. At the same time, a large buyer may gain from knowing more about expected production and local market conditions. A bank may assess agricultural risk more accurately. An insurer may price risk more precisely. A retailer may optimise procurement and logistics.

All of these can create economic value. But the distribution of that value will depend on access to information, market structure and bargaining power. This is why the debate about digital agriculture should not stop at the question of how much data India can generate. The more important question is how far that information travels through the agricultural system.

• Does it reach the farmer?

• Does it reach the extension worker?

• Does it improve a government decision?

• Does it change the behaviour of a bank or insurer?

• Does it improve procurement and logistics? And, ultimately,

• Does it improve the economic position of the farmer?

India is, therefore, moving from an era of limited agricultural visibility towards one in which it can increasingly observe agriculture at the level of the field and, potentially, the individual plot. That is a remarkable capability. But better visibility is not the same as better outcomes.

The next phase of India's digital agricultural transformation should, therefore, focus on building the institutional and economic links between data and action. We need to know not only whether information exists, but whether it is usable, accessible and acted upon. The ultimate test should not be the number of hectares mapped, databases created or applications developed. It should be what these systems change on the ground.

The journey we need to complete is straightforward: From data to information, from information to intelligence, from intelligence to better decisions, and from better decisions to better economic outcomes. The satellite image is not the outcome. The better decision is the outcome.

And the better decision itself is not the end. For Indian agriculture, the real measure of success will be whether this new information architecture ultimately produces higher productivity, greater resilience, more efficient use of resources and, most importantly, better and more stable incomes for farmers.

(The writer is senior fellow at the National Council of Applied Economic Research, New Delhi. The views expressed are personal and don’t necessarily reflect those of Down To Earth.)

Counting India’s waterbodies

Related Stories

Now that India can see its farms, what next?

Vapour on the rise

Data centre water consumption could triple to 644 billion litres by 2030

Thirsty data centres, thin regulation: Who is watching India’s water?