Watch Now

Commentary: AI, machine learning generate insights on global ag for Gro Intelligence

Startup helps companies forecast more accurately

Gro Intelligence is helping companies in the global agricultural industry gain better insights and make more accurate forecasts. (Photo: iStock)

The views expressed here are solely those of the author and do not necessarily represent the views of FreightWaves or its affiliates.

In this installment of the AI in Supply Chain series (#AIinSupplyChain), we explore how Gro Intelligence, an early stage startup based in New York City, is helping companies in the global agricultural industry gain better insights and make more accurate forecasts based on the AI and machine learning models it is deploying on its data platform.

Gro Intelligence — born out of insights gained in commodities risk and trading

I first met Sara Menker, founder and CEO of Gro Intelligence, in April 2013. She had left her job at Morgan Stanley, where she rose to the rank of vice president over the course of eight years, during which she started out managing commodities risk before assuming responsibility for managing an options trading portfolio.

When I met her, she told me that it was during her time at Morgan Stanley that she realized that there is a dearth of data for certain aspects of the market for agricultural commodities. This realization provided the inspiration for what eventually has evolved into Gro Intelligence. 

I asked Menker, “What is the problem that Gro Intelligence solves for its customers? Who is the typical customer?”

She said, “Global agricultural and climate data are incredibly messy and hard to use, which means that critical business and financial decisions — ones that impact our ability to feed ourselves and tackle climate change — are made with very little data.

“Gro provides clients with a clean, structured view of this vital but heretofore inaccessible data, and uses it to build dynamic visualizations, AI-driven forecast models, and indices that provide objective benchmarks which decision-makers can use to drive action.”

Further, she explained that “Most of Gro’s clients are part of the global agricultural supply chain — food manufacturers, commodities traders and manufacturers of agricultural inputs. Increasingly, financial institutions also rely on Gro’s data, models and indices.”

Gro was founded in 2014, she said. It launched the Gro platform in 2017.

Gro Intelligence’s secret sauce

I asked Nemo Semret, CTO of Gro Intelligence, “What is the secret sauce that makes Gro Intelligence successful? What is unique about your approach? Deep learning seems to be all the rage these days. Does Gro Intelligence use a form of deep learning? Reinforcement learning? Supervised learning? Unsupervised learning? Federated fearning?”

He said, “Compared to many legacy industries, we benefit from high-quality data in some aspects of our domain. Especially on the supply side, where we are dealing largely with the physical world, there is a real wealth of data, especially (but not only) from satellites.

“All our modeling would fall under ‘supervised learning.’ Sometimes we use classical machine learning (e.g., yield models, demand models), and in some areas deep learning (like knowledge graph automation). In general, we use whatever approaches help get the best results, and we value explainability a great deal.”

In very simple terms, supervised learning is a form of machine learning in which algorithms are trained to understand the function that maps input variables onto output variables. If we call the input variable X and the output variable Y, then the algorithms learn to determine the relationship between X and Y such that Y=f(X), and so that in the future when values of X are obtained, the algorithm can determine Y. The process of training the algorithm to determine future values of Y given X, on the basis of prior knowledge of the function that governs the relationship between X and Y, is called supervised learning.

In contrast, in unsupervised learning one needs to determine values of Y given historical values of X. There is no explicit historical knowledge of Y=f(X) on which the machine learning algorithms can be trained. As such the algorithms first have to determine the nature of the relationship between Y and X before they can be applied to future values of X.

Customers and competition

While Gro is not free to name most of its customers, the company recently worked with Unilever on Unilever’s Future 50 Foods project, which aims to encourage global consumption of 50 less-commonly eaten foods, such as buckwheat, fonio and nopal cactus, by incorporating them into Unilever’s products.

Gro used its machine-learning models to help Unilever understand how much of these foods the world produces today, where these crops might be grown most readily, and how to increase global supply without adversely affecting either the planet or people.

Most AgTech startups focus on precision farming, helping farmers calibrate, control and manage yield. There aren’t many startups working on the problem that Gro is focused on, and those that are tend to be much smaller in scale than Gro. The startups that came up as being most comparable are Trellis, Agrimetrics and Indigo.

Agricultural commodities — market background

According to The State of Agricultural Commodity Markets 2020, a report by the Food and Agriculture Organization (FAO), global trade in agricultural commodities and food “has more than doubled in real value between 1995 and 2018, rising from USD 680 billion in 1995 to USD 1.5 trillion in 2018 (measured in 2015 prices).” Moreover, “The share of agri‑food trade in total merchandise trade averaged at 7.5 percent over this period.”

According to Global Agriculture’s many opportunities, a June 2015 article by Lutz Goedde, Maya Horii and Sunil Sanghv of McKinsey, “Food and agribusiness form a $5 trillion global industry that is only getting bigger. If current trends continue, by 2050, caloric demand will increase by 70 percent, and crop demand for human consumption and animal feed will increase by at least 100 percent. Meeting this demand won’t be easy: for example, 40 percent of water demand in 2030 is unlikely to be met, and more than 20 percent of arable land is already degraded.”

These aren’t issues that will only play out in some far-off and distant future.

For as long as I can remember, Menker has been talking about the risks that the world’s ability to produce enough food faces from shortages in water supply. One relatively recent example is this October 2018 keynote address she delivered during the Henry C. Gardiner Global Food Systems Lecture at Kansas State University.

Moreover, in October the Chicago Mercantile Exchange (CME) and NASDAQ announced that water futures would begin trading on Dec. 7. Water futures are a financial derivative that enable users of water to manage price risk caused by the scarcity of water.

The team at Gro Intelligence often summarizes data about developments in the market for agricultural commodities using engaging infographics.

For example, the image below summarizes how the ethanol market is coping with the COVID-19 pandemic.

Triangulating between the data from the FAO and McKinsey, respectively, one conclusion is clear: The market for reliable and complete data and forecasts about the state of agricultural commodities markets around the world is a large and growing opportunity, one that the team at Gro Intelligence seeks to own.


If you are a team working on innovations that you believe have the potential to significantly refashion global supply chains, we’d love to tell your story in FreightWaves. I am easy to reach on LinkedIn and Twitter. Alternatively, you can reach out to any member of the editorial team at FreightWaves at [email protected]

Dig deeper into the #AIinSupplyChain Series with FreightWaves.

Commentary: Optimal Dynamics – the decision layer of logistics? (July 7)

Commentary: Combine optimization, machine learning and simulation to move freight (July 17)

Commentary: SmartHop brings AI to owner-operators and brokers (July 22)

Commentary: Optimizing a truck fleet using artificial intelligence (July 28)

Commentary: FleetOps tries to solve data fragmentation issues in trucking (Aug. 5)

Commentary: Bulgaria’s Transmetrics uses augmented intelligence to help customers (Aug. 11)

Commentary: Applying AI to decision-making in shipping and commodities markets (Aug. 27)

Commentary: The enabling technologies for the factories of the future (Sept. 3)

Commentary: The enabling technologies for the networks of the future (Sept. 10)

Commentary: Understanding the data issues that slow adoption of industrial AI (Sept. 16)

Commentary: How AI and machine learning improve supply chain visibility, shipping insurance (Sept. 24)

Commentary: How AI, machine learning are streamlining workflows in freight forwarding, customs brokerage (Oct. 1)

Commentary: Can AI and machine learning improve the economy? (Oct. 8)

Commentary: Savitude and StyleSage leverage AI, machine learning in fashion retail (Oct. 15)

Commentary: How Japan’s ABEJA helps large companies operationalize AI, machine learning (Oct. 26)

Commentary: Pathmind applies AI, machine learning to industrial operations (Nov. 20)

Commentary: Chain of Demand applies AI, machine learning to retail supply chain profitability (Nov. 26)

Author’s disclosure: I am not an investor in any early stage startups mentioned in this article, either personally or through REFASHIOND Ventures. I have no other financial relationship with any entities mentioned in this article.

Brian Aoaeh

Brian Laung Aoaeh writes about the reinvention of global supply chains, from the perspective of an early-stage technology venture capitalist. He is the co-founder of REFASHIOND Ventures, an early stage venture capital fund that is being built to invest in startups creating innovations to refashion global supply chain networks. He is also the co-founder of The Worldwide Supply Chain Federation (The New York Supply Chain Meetup). His background covers the gamut from scientific research, data and statistical analysis, corporate development and investing for a single-family office, and then building an early stage venture fund from scratch - immediately prior to REFASHIOND. Brian holds an MBA in General Management, with a specialization in Financial Instruments and Markets, from NYU’s Stern School of Business. He also holds a Bachelor’s Degree in Mathematics & Physics from Connecticut College. Brian is a charter holding member of the CFA Institute. He is also an adjunct professor of operations management in the Department of Technology Management and Innovation at the New York University School of Engineering.