So, how are farms capturing and using this data?
As new technologies are created, data methodologies are sure to change and adapt. For the time being, there are three levels of analysing the data; descriptive analytics (identifying what’s happening), diagnostic analytics (identifying what’s been done about it) and predictive analytics (identifying the outcome). All of these have a place in farming and, as analysis methods become more sophisticated, data can be interrogated differently, for the benefit of everyone.
Most new combines are able to record some level of yield data and many have a GPS receiver, so you know exactly where the bad places are. Samples are taken more frequently and modern day management means information is stored on computers and tablets rather than scribbled illegibly on to notepads. This had made data more accessible and easier to share than ever before. The technologies probably lend themselves best to arable farming. Growers can monitor fields on a square metre basis, applying nutrients only where necessary and treating disease problems only in areas that require it.
Farming by satellites and algorithms, rather than experience and knowledge, could potentially take the office-bound farm manager to a new level of efficiency. In terms of livestock, thanks to automated data collection, it is now possible to gather information in real time on flock health. However, few modern poultry farms can collate and truly analyse this data because of insufficient time, lack of theoretical knowledge, or other obligations. As a result, valuable information that could provide farmers with insight goes unused. This often means producers missed important warning signs that something was going wrong, which could result in significant damage and loss.
There is a vast range of sensor technologies currently available to farmers, and many are still developing, meaning even more data will be in hand in the coming years. With it will come huge opportunities to drive efficiencies through the supply chain, if that data is taken advantage of.
Even at current levels of technology, large and economically exploitable yield gaps remain in many places. In sub-Saharan Africa, in particular, there are indications of yield gaps whch could be exploited with given varieties and with known practices. There are many reasons why wield gaps exist. One is that farmers do not have sufficient economic incentives to adopt yield enhancing seeds or cropping techniques. This may be explained by numerous factors, including lack of access to information, extension services and technical skills. Poor infrastructure, weak institutions and discouraging farm policies can also create huge obstacles to the adoption of improved technologies at farm level. Nonetheless, access to and proper interpretation of data can make a massive difference for farmers who have exploitable yield gaps. Farms are not short of data creation.
The volume and range of data types generated by farms and their supply chains is vast and is generated at farm, service provider, processing and packaging, as well as retail level. Farms generate a huge amount of data and this volume is growing exponentially. This makes the flow of data between different areas of the sector a difficult process as the utility of the varying data types are not always immediately visible. This will undoubtedly change as time goes on though.
Despite a long history of data sharing or pooling in agriculture, it has not developed rapidly or kept pace with other industries – even though a huge transition has seen the industry go from being data-sparse to data-rich in a relatively short time. There have been technical issues with connectivity and compatibility as well as confusion on the legal status of data.
As such, a challenge facing the use of data in agriculture is the ability to flow and combine data with other market participants that could improve value. Furthermore, there is also a lack of willingness to combine this data. Generating insights and volume from this data will, in many cases, necessitate and include data sharing. However, farmers are understandably still cautious about sharing what, for many, is seen as their competitive advantage – how they farm their land. This is not to say that data sharing does not happen. In UK agriculture, there are good examples of successful data sharing. The sugar industry has collected data from all growers for years, giving them instant feedback on yields and sugars during the campaign, as well as market information.
One of the key barriers to knowledge gained from data in agriculture is complexity – of language, of scientific knowledge, of the format in which the data is reported. If we can take away that complexity, the base knowledge across the agricultural ecosystem can be increased. By doing that we can make the market a lot more efficient than its current state.
Information collected by farmers – yields, fertiliser use, crop rotation, rainfall and dozens of other data points – is catnip to firms such as Bayer, Syngenta, DowPuPont and BASF. The companies feed it into software that predicts combinations of seeds, fertilisers and sprays to maximise yields. That can boost sales of their products while also padding the bottom line from subscription fees farmers pay for recommendations on what to sow and when to spray. Lots of companies would like access to information about individual farms. Grain traders would pay a lot of money for real-time crop yields straight from the combine. Chemical companies would love to get an accurate picture of which weeds are prevalent on your farm before bombarding you with adverts.
In terms of governance and business innonvation, Ghana is a particularly interesting country to observe for agricultural data. The Ghanian Ministry of Food and Agriculture is tasked with surveying 24,000 farmers across 60 districts every year for their Ghana Agricultural Production Survey (GAPS). Collecting this data and entering it into a system takes over 9 months, making the reports outdated by the time they are published, limiting GAPS’ utility in the development and monitoring of effective agricultural policy.
By moving data to a tablet and data storage to the cloud, MoFA were able to conduct surveys faster and eliminate the time needed to manually enter data and fix errors, reducing the overall duration of the survey process by 82%. Timely data collection and analysis leads to responsive policy creation that focuses on economic growth as well as sustainable production and consumption.
Shifting focus on to businesses, Ghana-based firm Farmerline deploy mobile and web technologies that bring farming advice, weather forecasts, market information and financial tips to farmers who are traditionally out of reach due to barriers in connectivity, literacy or language. Sokopepe uses SMS and web tools to offer market information and farm recording management services. Another firm, AgroCenta, is an online platform that connects the smallholder farmer in the staple foods (rice, maize, millet and soybeans) value chain to a wider, online market to trade, access truck delivery services at the click of a button and also get a real-time market information delivered to their mobile phones via SMS and voice services.
Elsewhere there are companies, such as Gro, who are looking to combine big data and artificial intelligence to play a role by uncovering information and trends that send signals to the market. Such data could, for example, encourage the building of storage facilities to cut food waste at the farm level or help to price credit accurately.
Agriculture has always been an industry where numbers are important – whether it was how much nitrogen is applied, how much fuel you used or how many hours since the tractor’s last service – so it lends itself nicely to being a generator of lots and lots of data. The output of farm-related data has snowballed since precision farming practices like variable rate application and yield mapping came on the scene.
Another important point to remember is that, with consumers demanding transparency in the food chain, information on how a crop was grown and its environmental impact will be enormously valuable. Data is the foundation of better farming from both an efficacy and moral standpoint and this goes for all players involved in the value chain, not just the producers. By starting to collect data in a standardised way, a producer can gain the information required to make more informed choices about the services and supplies they choose to use and, in turn, the consumer can calm their conscience when purchasing these products.
Other key questions revolve around how private data is curated, their reliability and their statistical veracity. Data from private sources can be packaged and sold without any understanding of the quality of it. There could be big issues when it comes to determining sound premiums for insurance. Nevertheless, it appears that data has a significant role to play in the advancement of agricultural techniques and how the industry progresses.