Case study

Migratory beekeepers are dealing with increasing challenges due in large part to harmful pesticide applications. With mortality rates increasing year over year and creating a financial burden to businesses, beekeepers are seeking a solution to identify locations that are most heavily correlated with higher mortality rates amongst their bees. No solution currently exists for commercial beekeeping operations that meets the described need.

Data collection team

I was part of the on-field team and our main focus was to design a tool for the apiary worker to collect data of a single hive throughout its lifetime. What we considered as important data to answer the problem were the different GPS locations, data on management practices (feeding, treatment), and the mortality location.

Following our 3 principles, we decided that the best solution would involve a mobile application and NFC tags to identify each hive.

Our solution leverages the fact that a smartphone is already a tool beekeepers use and bring to the apiary. We imagined an app as a technological tool to track each GPS location that a hive visits over its lifetime. By scanning a group of hives, the application automatically creates a yard (or apiary) making it easy to report management practices within a few taps. 

First mockups of the protoype put together for creating a new yard by scanning hives.

Step two:
Elaborating the project

After pitching our concept to Dr. Brandon Hopkins and a few beekeepers' partners, we started to get traction around our solution. With the momentum, we decided to find an external agency to develop the first version of the mobile application. As a start, we evaluated that an Android app would be compatible with most apiary workers, and in any case, we could provide phones for our early adopters.

 Before jumping into refining the interface, I designed all wireframes and user flows.

Snapshot of our core userflows: location scan (1), track deadout (2), and report management practices on an entire yard (3).

Step three:
A simple interface

As the only designer assigned to the project, I decided to reuse Material UI components (app bars, shadows, text fields, bottom bars... etc) in my designs to focus more on the user experience. It seemed like a better investment of my time to me for greater impact on the project.

Built exclusively with the beekeeper in mind, my goal was to design an experience that demands the least amount of screen time: low touch.

Additionally, I designed illustrations and icons to ensure the app could easily be understood and used by illiterate apiary workers.

During this process, we constantly looked for feedback with our partners by presenting mockups in video calls, emails and on Slack. It was important for us to avoid any surprises and be aligned before testing on the field.

Onboarding

I used this opportunity to remind the apiary worker of the next steps after downloading the app. 

Track hives

After workers stapled tags on each hive, they will go around the yard to scan each one of them. By default, the scan is set to NFC, with fallback options such as QR scan or manually input the tag number. Once completed, the app will create a new yard with a 45m radius.

Final result:
MVP on the play store

We released the first version on the Google Play Store on January 22, 2021. The app is available in both English and Spanish. We were able to test and deploy the product in time for the almond pollination season as planned. Some team members flew to California to meet and observe the deployment. BeeTrack users were successfully onboarded and were generally quite comfortable collecting data with the device.

At the time of it's release, it was too early to tell if the product could make correlations with locations and mortality rates. The goal was to start collecting data and perhaps in the near future, Nectar could start marking correlations and even predictions.

Challenges and next steps

We’ve faced numerous challenges during our testing period and we already went back to the drawing board. During the deployment, we recognized the scanning process to be frustrating for the field workers. We also observed the QR scan to be much faster than the default NFC. Another challenged we faced was our default 45m radius for yards appeared to be too big, especially when in pollination season hives are distributed in small groups close by (called drops), and as a result, created overlapping yards.

We are hoping to bring more value for field workers by introducing yards and hive history in an upcoming release. We also made some quick changes such as switching the default scan mode to QR and reducing the yard radius to 10m.

As for the next steps, the priority will be to lead user interviews and usability tests to better understand how to further improve the app alongside developing an iOS version.