AI and data analytics enable farmers, insurers and lenders to see and act on situations faster, improve their yields and financial results, and become better stewards of the Earth

Insurance and lending companies rely on data-driven risk models to accurately price loans and policies and these are under threat. This risk assessment approach relies on dated methods, as risk models are typically built using a 10-year retrospective model. They look back at a multitude of factors, including trend behavior by crop type, yield data by crop, by region, weather patterns and as well as many other factors.
This methodology hasn’t changed much since it was established in the 1980s but extreme weather events over the past five to eight years have made it clear there’s a need for change.
Extreme weather has hammered the accuracy of traditional insurance and lending risk models. Insurers and lenders – whose profit margins have been shrinking year over year given the severity of weather and its negative impact on crop yields – are painfully aware of the problem. According to the RMA, Liabilities have doubled in Specialty crops alone in the past 12 years.
Now insurers and lenders are looking to implement more accurate datasets that consider what happens before, during and after a growing season. Such datasets, paired with sophisticated AI (Artificial intelligence) and analytics, will allow these businesses to be more efficient and predictive, produce better risk scores, and ensure they make the best decisions during the underwriting phase of a loan or policy.

As insurers and lenders embrace these new approaches, farmers stand to benefit from the results, getting better rates and policies that mirror what they are actually doing. If you are a farmer doing the right things in your field, such as limiting  water and fertilizer applications, and are a great steward of the earth, you should get rates that better align to your farming practices
The old-school method of assessing damage from severe weather is slow and inaccurate
Traditionally when extreme weather leads to a hailstorm, heavy wind or flooding, for example, a farmer will call their agent, who logs the call and tries to get an adjuster out to that area.
What happens next is typically a lot of waiting.
In some cases, it can take several weeks to get an adjuster to inspect a damaged area. When that person arrives, that adjuster is often a team of one trying to assess the damage from the perimeter or with a low precision drone.
Once that inspection occurs, it may take several additional weeks to understand the total impact that a weather event had on a crop that is several thousand acres in size.
Adjusters may also struggle to make accurate assessments given their limited access to the fields or regions.
For example, based on a peripheral assessment, it may appear that only 28% of the field was damaged. However, a massive downdraft may have left a giant hole in the center of the field, absolutely obliterating part of the crop. In that case, the exposure might be more like 42%.
AI and data analytics deliver better information faster, enabling quicker claims processing
Being able to get a complete view of what’s happening and measuring damage all the way down to the plant level very quickly and accurately enables adjusters and financial services  to understand how much of the field was impacted. That way, growers will know how much they may have lost, and lenders and insurance companies know exactly what their exposure is so that they don’t overpay or underpay on claims and they can process the claims 30%-40% faster.
Now imagine there was a big hurricane in Florida. Your insurance company tells you that it plans to fly over the citrus orchards that were likely to have been harmed by the storm.
My company, Ceres, did something just like that after a recent Florida hurricane. We quickly overflew about 11,000 acres of citrus trees using computer vision to understand the damage. In the process, we precisely identified just over 92,000 trees that had sustained damage of some sort. Our AI assessed which trees were damaged or totally destroyed and the exact location within the vast orchards.
This approach can provide that kind of detailed information in just 24 to 48 hours, accelerating and increasing the accuracy of claims processing. And when the soil dried out, the grower had the detailed information that they needed to quickly find, treat and replace the damaged trees.
The alternative was to wait until the soil dried out, have an adjuster with a clicker drive up and down each row to do a manual count of which trees are damaged, and then wait some more. This time frame is measured in weeks, not hours.
Extreme accuracy and anomaly identification are valuable all the time – in any weather
With traditional approaches, insurers and lenders also might not have accurate data on the field size or the crop that was actually planted and emerged. An existing document might indicate that a farmer owns 1,000 acres on which to plant corn. But only 800 acres of the land is actually arable. And the farmer may later decide to plant soy rather than corn to get a better price. Yet those details are not automatically updated in the insurance or loan application process.
Surveying and measuring a field quickly, and then generating a digital image that denotes the field’s footprint with the exact field boundaries down to 15 centimeters is extremely valuable.
With new digital datasets, farmers, insurers and lenders can validate the total size of the field, how much of the acreage actually gets planted and what crop ultimately emerges from it.
And computer vision makes it possible to spot anomalies emerging in some part of the field weeks before the human eye or other sensors would be able detect it. The AI then makes recommendations to farmers, who can act fast to remedy such situations.
For example, a farmer could add a fungicide or use more or less water to prevent further damage to those plants and ensure an optimal yield. That farmer’s bank and insurance company benefit too because they don’t have to cover a payment shortfall or a claim. 

AI and data analytics also help farmers maintain their sustainable and regenerative practices
Understanding what's going on in their fields at any given time allows farmers to precisely tune how much water, fertilizer and chemicals to use for optimal growth and minimal waste. 
The world’s growing population is increasing the need for food, making high yields more important than ever. Heat waves & droughts have limited the water available to grow crops.
And growing crops on the same land repeatedly can leech core minerals and additives that are naturally occurring in soil, decreasing the land’s productivity or even leading to soil sterilization.
Being able to leverage AI to understand exactly what a crop needs – and when and where it needs it –empowers farmers to run their operations more sustainably. That benefits farmers, the health and wellbeing of people who consume the food that they produce, and the world at large. Such fine-tuning is critical given the importance to maximizing yield while not impacting farmer’s income.

Agricultural insurance Lending AI

Back to blog

The difference between Ceres Imaging and other technologies I've used is the help I get from their expert team.
Jake Samuel, Partner
Samuel Farms
With Ceres Imaging we can take a more targeted approach to applying fertilizer and nutrients.
Brian Fiscalini, Owner
Fiscalini Cheese Company
These flights can cover way more ground and provide more insight than a dozen soil moisture probes — and it's cheaper to implement.
Patrick Pinkard, Assistant Manager
Terranova Ranch
The average Ceres Imaging conductance measurement from its imagery over the season has provided the best correlation with applied water.
Blake Sanden
University of California Cooperative Extension