Syngenta started off publishing open data to increase transparency but soon realised, with our help, that it could help them shift to a more open, or networked, approach to operating.
Syngenta is a global agriculture business that helps farmers make better use of their available resources, primarily through agrochemical and seed production. In order to continue to advance crop productivity, it invested more than $1.4bn in research and development (R&D) across 150 international sites in 2014.
Syngenta has a history of using publicly available data in its R&D. Alongside data related to land, weather and soil conditions, it uses biological data that has been published openly to build a detailed understanding of crop (and pest) traits – like tolerance of environmental pressures and resistance to viruses. Derek Scuffell, Data Strategist for R&D Information Systems at Syngenta, describes how:
We certainly have made use of a mix of satellite data that we buy and public satellite data. In R&D, what we make really heavy use of is huge amounts of public molecular biology data, which in Europe is mainly hosted by the European Bioinformatics Institute.
In 2014 Syngenta funded the European Bioinformatics Institute (EMBL-EBI) to extract bioactivity data from a large number of academic journals. The data covered insecticides, fungicides and herbicides, including more than 40,000 compound records related to crop protection. It was made available as open data through the ChEMBL database of bioactive molecules – Syngenta’s first experience of making data available for anyone to access, use and share.
Publishing Syngenta’s own open data
A year later, the company decided to publish its own open data as part of its Good Growth Plan. The plan addresses the long-term challenge of ensuring global food security for a rapidly rising global population. It focuses on the sustainability of Syngenta’s business and seeks to support transformation and change within the organisation by setting out six significant commitments.
In April 2015 Syngenta collaborated with the ODI to make six open datasets available related to its Good Growth Plan, including descriptions of productivity, soil, biodiversity and smallholder reach. The data will be updated yearly to measure the plan’s actual performance against its stated commitments. It is collected by external companies as well as Syngenta. The reporting process, its quality controls and evidence is independently assured by PwC and the datasets achieved Silver level Open Data Certificates, verified by the ODI. One of Syngenta’s key motivations for collecting and publishing data in this way was to develop external trust in the plan, which is essential for strengthening collaboration between private and public stakeholders for global food security. Making the data available has brought about an unprecedented level of transparency for the company.
Moving beyond transparency to wider use
As well as enabling anyone to assess the performance of the Good Growth Plan for themselves, publishing open data has had an instrumental role in the plan’s promotion and communication. Head of Data Sciences Graham Mullier explains how the open data release caught public attention in a way that wasn’t expected:
We’ve done publicity around the Good Growth Plan in the past when it was launched. The interest and engagement that we got from a much lower-key open data press release, all on its own, triggered similar levels of interest to some of the more expensive media-driven pushes we have done before. We were surprised at the way in which people picked up on the story.
Syngenta is now actively exploring other potential uses of the data. For example, its productivity dataset includes agricultural efficiency indicators for over 3,600 farms in 41 countries across Europe, Africa, Latin America, North America and Asia Pacific. Elisabeth Fischer, a development economist at Syngenta, hopes that this data will eventually be used by farmers around the world for resource efficiency benchmarking:
This is data that tells you how much input farmers needed in that year to produce a certain output. The data can be very interesting in countries where such data doesn’t exist to provide a reference point. For example, if I farm my land and I need significantly more input than the benchmark, then maybe I have an opportunity to reduce my input needs and maybe it helps me find that solution.
In this scenario, a solution could then be developed by the farmer, together with their agronomic advisors, in a way that meets the commitments of the Good Growth Plan itself. The company is looking at ways the data could be improved and made more useful, such as including farming practices or other details. As Elisabeth Fischer explains:
[We see a need to] create a platform where farmers can get access to this information and benchmark themselves against what others are doing. To make it more relevant, we work to increase the scope of the data.
Eventually we want this data to help empower farmers to make better decisions on what works best for them and their land in order to feed a growing population. But this needs more than just our data and our offer. It’s a piece of a puzzle and our contribution to developing a collaborative and unbiased approach.
Adopting a more collaborative approach
For Syngenta, opening up some of its data represents a wider, ongoing shift to a more collaborative business model in which data plays an increasingly significant part. It is estimated that there will be more than 20.8bn Internet of Things (IoT) connected devices by 2020. As the quantities of data collected by these devices in the agriculture sector grow, Graham Mullier believes that Syngenta will need to work in tandem with others to meet the challenges laid out in the Good Growth Plan:
The rate of data generation is increasing. The granularity is getting finer and finer all the time. All of that gives us loads more data to work with and the chances of any one organisation being able to generate the data, host the data, analyse the data and come up with brilliant answers all on their own seems vanishingly small. We have to find ways of collaborating.
Through its own open data release and contributions to the Global Open Data for Agriculture and Nutrition (GODAN) initiative, Syngenta is playing an active role in bringing about this collaboration. Ultimately, it is acutely aware of the costs of not taking this approach and not opening up to others in the sector. Derek Scuffell explains his view that:
… if we don’t have an open data approach then Syngenta will miss out on opportunities