As Marr (2015)1 points out, the real value associated with big data is not in the vast volumes available but what can now be done with it. He also discusses the danger of being bamboozled by the proliferation of technology and getting lost in a sea of data that delivers no value whatsoever.
Marr believes that whilst most businesses are already data rich, they are insight poor and that without a clear vision of the insights that an organization is seeking to exploit, significant resources will be wasted trawling through vast arrays of irrelevant data for non-existent insights. Therefore, it is critical that big data is leveraged in a systematic way that delivers business value. This white paper examines the application of big data to identify some key lessons for UK agricultural firms.
Focus on value, not volume – as alluded to above, it is not the volume of data that is important, it is the value that can be derived which is the critical issue. At a recent Agri-Tech East event2 which discussed the implications of Defra’s intention to release 8,000 datasets during 2016, it was highlighted that many datasets are currently non-comparable and exist in silos. In addition, these datasets suffer from a serious lack of inter-operability and connectivity which means that their potential value is not being realized. This presents a challenge for many UK agricultural firms that do not have the skills necessary to derive value from such datasets but yet would benefit greatly from the insights that could be gained. Furthermore, as David Flanders, CEO of Agrimetrics, points out a key challenge for precision agriculture offerings is that a significant proportion of farmers still need to be persuaded that it can be of value to their businesses. It is therefore vital that firms ask, what value big data offerings will provide to their organization, their customers and other supply chain stakeholders.
Clearly defined strategy and business objectives at outset are essential – Marr puts forward “The SMART Model” as a framework to realize the potentially substantial competitive advantages that big data could offer. This model advises to “start with strategy” to clarify business objectives and specify what one is trying to achieve before delving into the data. This is a sentiment echoed by Robert Allen of Greenvale AP, a UK potato supplier, who counsels that organisations must have a clearly defined strategy as to why and how they use data which is essential to extracting maximum commercial benefit. He cites crop agronomy and meteorology data usage within the potato sector as critical for yield forecasting, stating that the “commercial yield of a crop is determined by the gross yield, tuber size distribution and crop quality. In-season yield forecasting models provide valuable insight into final yields which can then be used for adjusting procurement and factory operational plans”3.
Be clear on which part(s) of big data value network to specialize in – according to a PA Consulting report4, there are five elements which are essential for creating value from the digitization of agriculture. These are:
ALGORITHMS: To calculate and evaluate key data parameters.
PLATFORMS: That provide common ground to link hardware, software, producers and consumers.
SMART FARMING HARDWARE: Such as computer driven seeding machines.
COMMUNICATIONS: Including Apps.
TECHNOLOGY HARDWARE: Such as drones.
Based on conversations with a number of industry experts, it appears that whilst having access to all of the elements listed above is critical, it is worthwhile to identify key areas where an organisation can develop its expertise and add value.
According to Srini Sundaram CEO of Agvesto, a provider of risk management solutions to businesses, it is also worth deciding on whether a firm should prioritise focusing on data sourced from within the farm gate (e.g. farming operations or enterprise performance data) or from outside the farm-gate (e.g. satellite-based remote sensing analytics, product usage data, and macroeconomic conditions etc.). This approach would allow a firm to directly link the data sources to the key risks firms face with respect to their profitability and sustainability. Such decisions would help organizations to make sense of the opportunities associated with big data and the areas that they need to focus on to add value.
Transparency concerning the usage of farmers’ data – is highlighted as a major issue and the question of data ownership is a key challenge facing the sector5. Farmers are understandably wary of sharing their data if it leads to them becoming disadvantaged when it is combined with other datasets to generate new insights (e.g. allowing financial traders to manipulate market prices).
Works best when it is innovative, impactful and can be integrated into existing datasets – these are some of the key criteria that Sainsbury’s uses when allocating big data R&D grants to agriculture6. Within a firm, data can be considered innovative when it fills a data gap and is often used in conjunction with existing data to deliver new insights and meaning.
Collaborate with external partners where appropriate – it is evident that with such vast volumes and varieties of data potentially available, even large firms find it very challenging to become in-depth experts in all aspects of big data. Where it is prudent for businesses to do so they should seek to collaborate with others to develop smarter solutions or to access offerings already available from other providers which enable them to access the solutions they need in a cost effective manner.
- Marr, B. (2015), “Big Data: Using SMART Big Data, Analytics and Metrics To Make Better
Decisions and Improve Performance”, Wiley. Accessed via: http://eu.wiley.com/WileyCDA/
- Agri-Tech East (2016), Special Interest Group Report: “Big Data and Remote Sensing and
Monitoring – Defra Data Opening Up the Treasure Trove”.
- PA Consulting (2015), Digitising Agriculture: Unlocking the potential in the agricultural
value chain, accessed via: http://www.paconsulting.com/our-thinking/digitisingagriculture/