Building good decision support tools: advice for designers

Decision support tools, usually considered to be software-, app-, or web-based products, offer the ability to provide evidence-based decision-making advice to practitioners. Such tools lead users through clear decision stages based on evidence, ultimately allowing practitioners to make a more informed, productive, and efficient decision. However, previous research in agriculture and other areas has found that uptake of decision support tools in practice is low. In other words, researchers from universities and industry have used resources to build tools which were seldom used by their desired users. This is a major problem for two reasons, 1) lack of uptake means that the aim of improving the evidence base for decisions is not realized, and 2) resources are wasted.

As part of Defra’s Sustainable Intensification Platform, the University of Cambridge (alongside other partners ADAS, Glasgow Caledonian University, University of Nottingham and the University of Exeter) led a piece of work to find the reasons why farmers and advisers use, or do not use, decision support tools. Through the use of focus groups, surveys, and interviews, farmers across six sectors (Cereals, General Cropping, Dairy, Mixed, Lowland livestock, LFA livestock), as well as arable and livestock advisers, were asked about whether, and why, they used tools to inform their decisions. An open access paper published in Agricultural Systems, presents the results from this study. The research team argues that fifteen factors are influential in uptake and use of tools and present a checklist to inform the future design of tools. By paying attention to the fifteen points on the checklist, which include performance, ease of use, and peer recommendation, it is hoped that tools will be designed that make an impact in practice. In order to create the working environment in which these fifteen factors are considered in the design and delivery process, it is suggested that designers consider adopting the five-stage process below:

1. Tools should look and behave differently for different audiences e.g. a tool for a farmer should be different than one for an adviser or policy-maker. Different groups have different skills and diverse questions to answer. A tool that doesn’t know its audience is likely to fail. Other user factors to consider: age, cashflow, workflow, computer literacy, embedded habits, farm scale.
2. A designer needs to consider whether the tool is better than current ways of making decisions. Does it beat farmer/adviser knowledge or existing guidance or tools? Does using it lead to tangible benefits?
3. An interface needs to be user-friendly and provide information quickly.
4. A designer needs to consider whether there are barriers to widespread adoption e.g. lack of rural connectivity.
5. Even the most fantastic tool needs marketing. Designers could make use of trusted farmer and adviser networks to enable peer recommendation and knowledge exchange.

Designers of decision support tools in other areas could also benefit from a similar process.

For further information about this work, please contact the lead researcher Dr David Rose at dcr31@cam.ac.uk