My Role
Lead Product Designer
Project Duration
Ongoing
Platform
IOS
Team Members
2 Software Engineers
Industry
CleanTech
Overview
CarbonCruise is a mobile app that modernizes forest data collection by replacing manual tools such as tape measures, clinometers, and paper notes with a single, mobile experience.
When I joined, the app was already in development, and while core functionality existed, product decisions were being made without a clear understanding of real forestry workflows, leading to investment in technically ambitious features that had limited impact on efficiency.
My role was to bring clarity by identifying where time was actually being lost in the field, and redirect product strategy toward solving the highest-impact problems first.
My Role
As Lead Product Designer, I was responsible for:
Defining product direction and priority areas.
Translating real-world forestry workflows into clear product requirements.
Partnering closely with engineering to guide technical tradeoffs.
Designing AR-first and offline-first experiences that function in real forest conditions.
The Challenge
Forestry data collection is repetitive and physically demanding. A single day can involve measuring hundreds of trees in dense forests with poor lighting, uneven terrain, and little to no connectivity.
Digital tools only succeed if they reduce effort compared to manual methods. If a tool slows foresters down or fails under real conditions, it won’t be adopted.
How do we design a digital tool that outperforms manual methods in real forests, not just ideal conditions?
Understanding real forestry fieldwork
To ground the product in reality and gain a better understanding of
professional forester workflows, I conducted:
Workshops
Ran a series of 3 workshops with professional foresters in the field.
Iterative testing
Tested the app
in active forest environments.
Comparative
Analysis
Compared and measured the app directly against traditional tools.
Workshop sessions with foresters surfaced key workflow gaps that directly informed measurement and prioritization decisions.
What I learned
CarbonCruise's current diameter measurement workflow was slower & less reliable than manual methods
Diameter measurement is performed repeatedly during plot collection and has a disproportionate impact on overall efficiency. The app's existing LiDAR-based approach introduced processing delays and required users to move around each tree, which performed poorly in dense stands and often made digital measurement slower than manual tools.
“If it’s slower than a tape measure, it’s not an improvement.”
Vicky, Professional Forester
Planned Species ID feature did not align with user priorities
Early product planning included significant investment in an AI-based species identification feature. In practice, foresters rely on experience, genus-level identification, and manual notes, as automated species identification tools are unreliable in dense or mature forests and often require additional verification.
“Speed matters more than automation. I’d rather log what I know than double-check an AI guess.”
Derek, Professional Forester
Common Field Conditions Were Not Adequately Supported
Forestry work often occurs in low-light environments and areas with limited connectivity. Offline authentication blocks, limited feedback, and unclear recovery paths disrupted data collection and reduced confidence in the tool.
"You don’t get a second chance to measure once you move on, so it's important to get it right the first time."
Todd, Terrestrial Ecologist
Tested an early version of the app in a dense forest, with foresters and engineers to evaluate core measurement workflows and compare how digital measurements perform against traditional tools.
Solution
Based on field research, we refocused the product on improving speed, reliability, and trust during the most common field workflows. This required narrowing scope, redefining success criteria, and reallocating engineering effort toward areas with the highest impact.
Refocusing on Diameter Measurement
Given its frequency and impact, diameter measurement became the primary focus of the product. I worked closely with engineering to move away from LiDAR point-cloud scanning and toward an AR-first, automatic edge-detection approach.
This allowed users to:
Measure tree diameter from a single standing position
Avoid circling trees in dense or uneven terrain
Receive faster and more consistent results in real forest conditions
This shift significantly reduced time-on-task and eliminated many of the failure cases observed during field testing.
Expanding access beyond LiDAR-enabled devices
Diameter measurement is a required step in forestry data collection and was the only workflow dependent on LiDAR, effectively limiting the app to iPhone Pro users.
To broaden access, I designed a secondary, non-LiDAR diameter measurement flow inspired by familiar tools like Apple Measure. While slower than automated measurement, this approach enabled full app use on non-LiDAR devices, allowing more users to collect data reliably in the field.
Designing for offline-use
Offline capability was treated as a core requirement rather than an edge case.
I designed flows that allow users to capture and edit data fully offline, continue working without authentication interruptions, and sync automatically once connectivity is restored. Clear system feedback communicates data, sync and connectivity status, reducing uncertainty during fieldwork.
Improving feedback and error handling
To build trust in the tool, I introduced clearer feedback around AR accuracy and environmental limitations, including contextual messaging when lighting or conditions affected measurements and clear recovery paths after interruptions.
In-app guidance was also added to support new users and volunteers.
Establishing a scalable foundation for future development
Rather than removing advanced features entirely, the product was designed to support future automation once core workflows were reliable. By stabilizing measurement and data capture first, the system now provides a stronger foundation for future enhancements, including AI-assisted species workflows, without compromising usability in the field.
Outcome
Re-established the product’s core value: faster, more reliable field data collection
Redirected engineering effort toward the highest-impact workflow
Increased measurement speed while reducing troubleshooting time
Expanded the potential user base beyond LiDAR-enabled devices
Created a scalable foundation for future automation and AI features
Next Steps
With core measurement workflows established, the next phase focuses on refinement and scaling the product:
Improve edge-detection accuracy across different lighting conditions and tree shapes
Strengthen measurement feedback and offline reliability
Continue field testing across additional regions and forest types
Expand species directory to account for regions outside of North America
With a stable foundation in place, future work can explore:
Selective automation and AI features where they clearly reduce effort
Expansion into additional forestry data collection workflows such as canopy measurement
Broader adoption across municipalities, conservation groups, and volunteer programs
