Zambian tech startup building products for farming sector: Talking business with CEO of AgriPredict

With the AgriPredict app, a farmer takes a picture of the suspected diseased plant, and the system will give a diagnosis, options for treatment if required, and location of the nearest agro-dealer.

Cassandra Mtine-Makumbi, co-founder and CEO of Zambia-based AgriPredict, talks to How we made it in Africa about building a platform to give farmers instant information regarding crop disease, weather patterns and market access.

Give us an overview of what AgriPredict does, and how the product works?

AgriPredict is a platform that helps farmers access a variety of information either via a smartphone app or a USSD code-based service.

Advertisement

Our first feature enabled users to diagnose crop diseases by taking photos of diseased plants with their smartphone and comparing images to a database of thousands of other images of crops. We later added features such as weather reports and an early warning system for external threats such as pests.

In October last year, we began to roll out a marketplace for buyers and sellers and we will soon be giving farmers access to agricultural guides and resources.

What inspired you to get started?

The majority of agriculture in Zambia and across Africa is small-scale. During the 2016-2017 farming season, the country was invaded by the deadly pest, fall armyworm, which affected hundreds of thousands of hectares of maize fields in all provinces, negatively impacting people’s yield and income.

This showed us the need for a management tool that farmers could use. We’d seen how these farmers faced a number of challenges and weren’t getting adequate solutions. From our initial research, we identified diseases, pests and weather as some of their biggest obstacles.

How has the business since developed?

We spent time in the field in 2019, speaking to farmers, and got a lot of feedback that influenced AgriPredict’s direction. Owing to travel restrictions, we were not able to spend as much time in the field in 2020, so we focused on product development based on the feedback and fieldwork.

When we ran a pilot service in 2019, we found a lack of smartphone penetration within our target audience and so we built up a USSD service that can be used via any phone. We have since established a partnership with MTN to provide smartphones to farmers. At this stage, the vast majority of our users still interact with us via non-smartphones. We anticipate more smartphone usage as they are made available to farmers.

Cassandra Mtine-Makumbi, co-founder of AgriPredict.

The pilot also emphasised the need for a wider range of services beyond mere disease protection. We realised that even if farmers increased their yields, they still had issues with access to market, which is what we are addressing with the buy-and-sell feature we rolled out last October.

Explain AgriPredict’s business model.

Our goal is always to maintain a free version of the product for farmers (although we may at some point offer a paid, premium service). The value is in the data generated: geolocational data, crop health, yield predictions and market trends.

Data is a huge gap across the region and we are confident about data as a revenue generator. There is a demand across the private and public sectors; we have received interest from NGOs and microfinance institutions interested in paying for access to data. We also anticipate interest from research organisations, investors, agricultural suppliers and support service providers.

The more people we add, the better our data becomes. Already, we have over 50,000 farmers on our platform, out of an estimated 1.5 million across Zambia. There are millions more across the wider region who equally need the service, whom we will be looking to reach.

What are your growth strategies?

Partnerships are our biggest strategy. We value them more than direct investment. Working with bigger and established organisations allows us to tap into large on-the-ground networks of farmers and pre-existing knowledge bases while offering the other party benefits as well. Some of these partnerships are revenue generating, while others are simply mutually beneficial.

AgriPredict provides tools to help farmers manage risks such as droughts, pests and diseases.

AgriPredict provides tools to help farmers manage risks such as droughts, pests and diseases.

Tell us about your biggest challenges.

Team building is always one of the big challenges; finding the technical skills in the local market in particular, because tech is still a relatively nascent industry and there isn’t a huge and established talent base within the country.

Collecting enough data is another hurdle; the artificial intelligence algorithms are data-hungry and collecting the data itself is resource-intensive. For each disease we map, we need to collect thousands of images. Again, we tend to use partnerships where another organisation can provide access to existing stores of data, rather than starting from scratch ourselves every time.

What are some of the entrepreneurship lessons you’ve learnt?

Whatever time or money you think it will take, double it. It’s not a smooth journey and things always take longer than you anticipate. A strong vision is important because, at some point, interest alone is not enough to keep you going.

Don’t limit yourself based on money as there will never be enough. We have maintained a lean mentality, bootstrapping all the way, which has taught us many lessons about how to prioritise spending.

Finally, mentorship networks are important. Find a mentor and plug into groups of like-minded people. On days when you want to give up, realising other people are in the same boat can keep you going.

James Torvaney is a business consultant and financial advisor specialising in West Africa. He has worked with clients across a number of sectors, including technology, manufacturing, consumer goods and hospitality.