Smallholder yields in Asia are well short of both international benchmarks and yields on larger farms in the region. At least part of this gap can be attributed to knowledge and information; including on seed selection, pest identification, climate and planting timing.  

 

Digital advisory seeks to solve this problem by providing advice and information to farmers. A typical solution will provide some generic advice over a social media platform and seek to migrate farmers over to an Android application, where the farmer creates a profile, and receives more customized advice.

 

Digital advice has a range of advantages:

• Once the material and interface are created, adding each additional farmer costs very little, and is much cheaper than in person advisory.

• Advice can be highly targeted to the farmer’s specific plot, crop type and point in the crop cycle.

• Once trust is established, the business can undergo viral growth, with one farmer suggesting the advisory application to another.

 

A number of types of digital advisory have been tested including Q and A with agronomists, advice driven by satellite imagery and pest identification using image recognition.

 

An advisory solution can be funded in two different ways, though farmer subscriptions or by selling farmer data.  The farmer subscription model is proving to be difficult to scale, as the cost of building enough trust to secure each new paying farmer typically exceeds their lifetime value. Furthermore, paying farmers expect a high level of service, making this a more expensive offering to deliver.

 

Based on our observations, what appears to be more scalable is to provide the service for free; and monetize the data collected on the farms and farmers. Thus, while a subscription can be added over time as a premium offering, data sales will usually drive early adoption.

 

Much remains to be learned about how to build trust in digital advisory, what information farmers want, how to design the user interface, and leveraging existing trusting personal relationships (such as with retailers) to onboard farmers.

 

However, if data sales are going to drive revenue, considering who will pay for what data is a critical consideration. Potential data consumers include crop protection manufacturers, fertilizer companies, lenders and commodity buyers. In each case the advisory service is essentially selling a qualified lead, examples include:

​

• Lender: this farmer is three weeks from planting and expressed an interest in hybrid seeds last season.  Their potential seed buy is $100.  Satellite images show consistent, high yields for three years.

• Crop Protection Company: this farmer is four weeks into a rice crop, and the crop is inundated with Brown plant hopper.

• Buyer: this farmer is one week away from harvest, with an estimated 6 tonnes of white maize.

 

The lead’s value is driven by three factors, the farmer’s spend on the data consumer’s product or service, the data consumer’s profit margin and the likelihood of conversation. For example:

​

Farmer crop protection spend:                      $100

Profit margin:                                                   5%

Likelihood of conversion:                               50%

Indicative lead value:                                      $2.50

​

This calculation gives an indication of the value of the qualified lead to a crop protection company.  Similar calculations could be run for a lender or buyer.  The value of each lead will be low, but a particularly promising aspect of the business model is the ability to monetize the same farmer’s data multiple times with different classes of data consumer.  

​

The model raises a number of ethical and legal issues around data protection and privacy, but with social media sites already operating at scale with a similar business model, solutions should be found.