Open Data and Food
Towards a smarter Agric industry
A lot goes into turning tiny seeds planted in soil to the food we get on our plates. From inputs to production, transport and logistics, trade, processing, waste handling, distribution up till retail and purchase, a lot of decisions are made and each step is carried out by several actors along a value chain, ending with the consumer. All these actors, like in any other sector need access to sufficient information if they are to make smart trade and investment decisions.
For example, the more informed a farmer is about the weather, the different types of seeds he could plant, the soil conditions and so on, the better he is able to grow crops more successfully — with minimal losses. The farmer should ideally also have enough market information to decide who to sell produce to after harvest or whether to even sell at all, rather than perhaps store crops for maximum profit later.
Similarly, input and commodity traders need to have information about price trends and supply/demand for their products at their fingertips. Processors must have sufficient market data about inputs, suppliers, distributors, consumer prices and any other market activity that could affect their business. Investors and commercial finance institutions on the other hand are interested in the profitability of the agricultural businesses in order to quantify risks and provide required loan and insurance services. This entails having crucial details about the farmers/traders business activities and other market projections.
These days, consumers too are caring more and more about where their food comes from. Details about the nutritional content of the food in their plate including environmental and health concerns like how much pesticide was used, type of packaging, animal welfare, CO2 neutrality or even ‘fair’ costs of production labor are taken into account by some consumers.
Agricultural data is also closely tied to policy making. As a primary industry, agricultural activities impact our natural and environmental resources. It also provides jobs for a huge part of the world population and quite literally ensures our survival through food security. A lot of trade, environmental and welfare policies therefore revolve around agriculture and it’s commodities and so, analyzing certain data points as key indicators can help policies be better formulated and evaluated.
Clearly all these activities require various types of information from multiple data sources.
Where does this data come from?
Naturally, there’s a huge gap between the needed and the available agricultural data — especially in developing countries. Because farming is typically carried out in rural communities, it’s harder to collect accurate and up-to-date farm data regularly. It’s also traditionally harder for these farmers to get the right data and information they need — or to even know they need it!
We also need a lot of quantitative and qualitative data about the environment, financial and socio-economic data when thinking about the food system as a whole. Typically, industry data sources include governmental bodies, academic research institutes and private organisations who do market research. Of these, Governmental and academic organisations usually have more incentive to provide open data in order to facilitate industry growth — however in most developing countries this isn’t the case. Most industry actors in developing countries have to rely on Intergovernmental agencies such as FAO and World bank for broad, high-level data and private market researchers for more granular data. For example, I’m a foreign manufacturer of mid-priced tractors looking to expand my market to Nigeria. I would want to know which areas have the largest farms (by landmass) or at worst a high concentration of clustered farmlands where mechanized farming is feasible. What is the average revenue of each of these farms and so on and forth. Needless to say, this is quite problematic and poses limits to growth in the sector.
It’s one thing to have data and another to actually use it effectively. A lot of agricultural data is not properly collected for the purpose of re-usability. And according to the CGIAR Big data platform, there’s so much agricultural data on the web which is simply lost due to non-findable metadata. I’d add also to this the use of hard to reach data storage formats like storing data tables as PDF reports, or using non- standardized indicators.
Another big issue is access: Even when this data exists, they end up in silos without achieving much of their potential purpose. Where can people easily go to find this type of data in a format that’s easily understood and relevant to them? How can we design more solutions that bring this data closer to the end-users?
A good example I came across is Vechain; block-chain powered supply chain tracking. Or imagine if farmers could simply scan a barcode and track the entire history of a seed/input, including information about its nutritional content. I was excited to read about a proposed project like this.
One of the ways to make data ‘digestible’ for end users is by visualizing it using charts and dashboards. E.g this mock-dashboard I created using Tableau, showing some basic trends in Nigerian agriculture. Visuals help us communicate the essence of data and promote usability and distribution to those who actually need it.
We should be looking to create open, FAIR and presentable agricultural data platforms that puts the power in the hands of all actors in the food chain to make better, well informed decisions.
