By Cedrik Kern, Oliver Rueckert and Lauren Gibbons Paul

That last ice cream cone you enjoyed was creamy, sweet, and refreshing. It was also the product of data. This includes data about weather and soil conditions, crop protection, and livestock well-being. Data that measures improvements in productivity. Data that tracks the efficient use of resources so future generations of farmers can continue to tend their crops – and future consumers will enjoy frosty treats that originate with their harvests, like the sugar, cream (or oat milk if you want) and flour for the crunchy cone.

The journey from farm to table is a long one. It can take years for farmers to cultivate inputs (seeds, plants, livestock) and manage resources just to get their products ready for sale and distribution. Then it’s time to prepare for the next season. And then do everything all over again, only better.

More on how intelligent technologies can help the agricultural industry lift poor farmers and heal the planet.

Read “The Future of Feeding the World.”

The digitalization of farming practices is closely connected with the concept of precision agriculture. It’s a concept that has been studied for decades, starting in the 1980s with the advent of geographic information systems that tracked crop production and field conditions. Since then, the amount of data that is captured around fields and farms has increased steadily.

But while the introduction of AI and machine learning applications injects fresh promise of productivity gains in agriculture, today’s farming companies face two major challenges. The first is cultural: because farmers have been familiar with their processes for years if not generations, it can take extra effort to overcome skepticism and subsequently adopt new practices that take advantage of those growing data sources. The second challenge is data-centric and technical: there is often a lack of standards to use the data in AI and analytics applications efficiently that can help companies improve farming processes.

These challenges confront farmers in a global context that affects all industries and raises the stakes for farmers using data to improve productivity and resource management. Factors include:

• Extreme weather: The environment is more precarious than ever, with extreme weather events threatening to render time-tested farming practices obsolete at a time when the world needs productive farms more than ever.
• Conflict: When war breaks out, as it did when Russia invaded Ukraine in February 2022, it disrupts the global movement of agricultural products such as wheat, exacerbating supply shortages in regions such as the Horn of Africa, as NPR reported.
• Global hunger: In July 2023, there were 643 million people in 88 countries who did not have sufficient food available to them, according to estimates from the United Nations World Food Program.
• Population growth: UN projections note the world’s population is expected to rise by nearly 2 billion in the next three decades, to reach 9.7 billion in 2050.
• Consumer demands for transparency: In developed economies, more demanding consumers want to see where their food comes from and how it’s produced. A 2022 survey of U.S. households found consumers want food brands and retailers to share not just the ingredients in the food they buy but also information about manufacturing practices, ingredient sourcing, and company sustainability efforts, Supermarket News reported.

The better use of data and technologies like AI to inform decision-making promises to put agribusinesses on a path to address these challenges so that enterprises (many of which include their own farming operations or collaboration with other farmers) can improve their business processes. The adoption of what experts call precision agriculture is still in its early period, but there’s no question it’s spreading.

A 2023 McKinsey report found that “though agricultural technology adoption is slow, farmers are open to innovation.” Agribusinesses based in North and South America were more likely to have adopted, or have plans in place to implement, precision agriculture systems such as yield monitoring, variable rate fertilizer applications, and in-field soil sensors, the report notes. Even so, momentum is building.

But first, farmers have to trust that the data will help them.

Gaining the trust of farmers

Business process shifts that call for using data to drive decision-making are challenging by definition. Adopting AI and data-driven agricultural processes is like an aircraft pilot accustomed to flying visually shifting to instruments to guide their flight. A farmer can’t always see the data that goes into the decision-support systems. There has to be trust in the systems that indicate the next-best irrigation setting or pest-control treatment.

In our experience, this change in mindset needs to take root in different levels of the agribusiness organization. It’s not enough for the top decision-makers to buy in. The farm managers and those closest to the field need to adopt the new ways of working. We’ve witnessed large organizations that invest in producing impressive data-driven models to guide improving irrigation and fertilization programs, to improve a farm’s efficiency and sustainability. However, if managers on the ground are skeptical of this new information, and a change fails to occur, the company risks missing out.

What’s needed: a deliberate, incremental, iterative approach that brings data to agriculture in ways that mirror farmers’ experiences. Going step-by-step makes it possible to make small changes and gain support from farmers used to planting and harvesting the same way they always have. Many farming companies want to build upon the business practices they have honed over time. It’s their secret sauce. They want to ensure that the farming experience they have had for many years is one of the ingredients for future success. By using AI models that produce recommendations in discrete parts of the process – fertilization, irrigation, crop protection, and variety selection – farmers can start to adopt digital decision-support models in a non-interrupting way. The AI models can be tested, and farmers can build confidence in the systems’ recommendations. Feedback loops and continuous improvement are another crucial success factor.

At the Royal Swatini Sugar Corporation (RES) based in the Swazi nation in southern Africa, executives have made real progress in adopting new precision agriculture systems. The 4,500-employee company produces as much as 470,00 metric tons of sugar per season for use in ethanol, fertilizer products, and beverages to customers in Africa, Europe, and Asia. It manages approximately 24,450 hectares (more than 60,000 acres) of irrigated sugarcane fields, which include both its farms and partnerships with smaller third-party growers.

As RES worked to deploy systems to manage farm and field data more consistently to automate tasks and improve decision-making, including the third-party growers was a major focus. In 2023, the company cited numerous benefits from the data-driven farming solutions: a marked improvement in farmers’ ability to plan, implement, and measure the progress of key tasks. It also provides a better understanding for growers about the right time to harvest and higher grower yields.

The acceptance of the new systems has been vital to improving operations, says Rob Coombe, Group IT manager at RES. “As well as giving us insights into their crop, it lets us learn when is the best time to harvest,” Coombe says in a short video about the company’s efforts. “It’s going to be a key component of our business going forward as it helps us to be more sustainable and efficient on the farm.”

Standardizing the use of data

The second challenge facing agribusiness companies – collecting and managing the right data so that decision-support models are accurate and valuable – is tied to the first. Getting farmers to trust what systems tell them starts with having good data.

Today, agribusiness companies are faced with an increasing amount of data collected from the field. But there is often a lack of standard processes and data management practices to make efficient use of that data, so it helps farmers improve their processes.

At KWS, a Germany-based producer of seeds for farmers around the world cultivating crops like corn, sugar beet, cereals, rapeseed, sunflowers, and soybeans, investments in systems to track conditions on the ground and then use that data to improve productivity and quality have been a priority in recent years. The company, the world’s fourth-largest seed producer, has been in business since 1856.

KWS applies analytics to aerial photos of fields to speed up the collection and analysis of data to detect factors such as fungal infestation, soil conditions (such as moisture), and chlorophyll levels. The analysis is not only more efficient but more accurate than the human eye test.

Jens Hittmeyer, head of global IT and CIO at KWS, says plans call for digitizing the company’s entire logistics value chain – from field to plant and then to backend systems.

“In the future, we will get more and more data from the fields. This will help our farming customers to optimize the input from fertilizers, water, and crop protection so that they can work as sustainable as possible,” he says.

Think again about that ice cream cone (or other favorite food) that you enjoy. It takes a lot of resources to make up the ingredients – even before it gets to the people who prepare them for consumption. With a growing population and a warming planet, every seed, every plant, and every ounce of water and fertilizer is crucial. Data can make a difference. But only if the farmers who can use that data to their advantage trust what they see in an analytics readout as much as they are accustomed to trusting what they see happening on their farms. This is not just any farms, but their farms. Industrious farmers will yield the fruit of this data harvest.