To accurately forecast a harvest, you need to know how many fruits you have on your trees. Yield forecasts help farmers and the industry plan their harvests, resources and marketing. But until recently, the forecast process was time-consuming and inaccurate.

CropCount combines satellite imagery with targeted on-ground sampling to produce a yield forecast with better than 90% accuracy. Using a sophisticated algorithm, CropCount identifies which trees should be counted to give the most accurate picture of health and yield across an orchard.

Our challenge was to marry the science with a practical way for farmers (and sometimes their seasonal workers) to accurately tally the crop. Our design sprint unearthed insights leading to a rapid prototype dovetailing into an agile production workflow. Design and engineering worked in tandem to create the interface, branding and develop the platform backend.

We worked with farmers just as closely as we did scientists, rigorously testing early prototypes to gather valuable insights and to find out what would best fit with their farm management practices. We also had to consider devices and connectivity which was a big issue when out in the field. This meant building an app that worked effectively offline with the ability to sync when users were back online.