Designing a data-based dashboard for decision making in animal farming: Traffic lights vs. Sequential colours

• Digital dashboards support data-driven decision making in agriculture. • Design choices may lead to biased decision as it offers a frame of reference. • Traffic light colour schemes enhanced information retention and certainty. • Simultaneously, traffic light colours for mortality may be misinterpreted. • Important nuanced output may be missed in the presence of more salient output. To support data-driven decision making, aquatic and terrestrial farmers increasingly use digital dashboards that visualise data collected on farms or offer decision support through the provision of information about relevant metrics. However, the design of the dashboard, particularly colour schemes, may shape users’ perceptions and decision making in unintended ways. This study investigates how the different colour schemes “traffic light” and “sequential,” and alert level configurations including regional discrepancies may influence users’ recall, perceived risk, certainty, and actions in salmon health management. For this, online experiments were conducted with 1482 members of the public in the United Kingdom and 98 European students in veterinary medicine, animal science, and fish biology, serving as proxies for real-life decision makers. Participants interacted with simulated data tools that visualised salmon mortality and harmful plankton levels at both the farm and region level. Alert levels on the participants’ own farm significantly influenced risk perceptions ( p < 0.001, η 2 = 0.50-0.80) and decision making ( p < 0.001, Cramer’s V = 0.63-0.66) in both samples. However, colour schemes had only small effects in that the traffic light system enhanced information retention ( p < 0.001, Cramer’s V = 0.21) and, in high-alert scenarios, led to greater decision certainty ( p = 0.015, η 2 = 0.01). Nevertheless, caution is warranted due to potential misinterpretation or bias when using traffic light colour schemes. Stakeholder-informed, iterative dashboard design, meaning involving stakeholders in the step-by-step development of digital dashboards, is essential to balance visual simplicity with accurate risk communication.