Imagine you are in a store shopping for a table. You snap a photo with your phone, and an app tells you the very forest where the wood came from, and whether it was responsibly harvested. A startup called Tracy of Sweden is making that a reality — because each tree actually has a unique “fingerprint,” which Tracy’s patented algorithm can read. They worked with us at the IBM Garage, alongside actual lumberjacks, to develop a full solution on IBM Cloud that can help prevent deforestation. This algorithm can be used by any forestry professional to track and verify the origin of logs to combat illegal logging. The solution has enormous market potential in the market, ranging from tracking and tracing the origins of a log to verify authenticity, to regulating and fining illegal logging operations to protect the rainforest, to empowering a customer purchasing a wooden table to be confident in the origins of the wood and trust in sustainability claims.
How we got there
Creating long-term vision
During the IBM Enterprise Design Thinking Workshop, Tracey of Sweden, alongside us from the IBM Garage team, prioritized many opportunities and focused in on a single one: how can we improve the tree identification and tracking experience in Sweden (i.e. the process a tree goes through from origin to being cut down to sawmill processing and finally being sold)? We knew that designing this vision would be good for Tracy of Sweden’s mission and business because it would increase transparency through the process of helping to prevent deforestation by giving stakeholders factual and trustworthy information to act upon.
How we got to our Minimum Viable Product (MVP)
While the vision we created together with Tracy of Sweden was great, we knew it would take a lot of time and effort to make it real. In order to find the starting point for development, we leveraged lean startup methodology and evaluated the risks together with the team. Major risks were around the ability to move the fingerprinting algorithm to the cloud, match the images of log faces to the source in real-time, and create something meaningful for lumberjacks that incentivizes them to use the application.
To test those assumptions and at the same time create something useful, production-ready, and scalable, we chose to prioritize experiences. In order to do that, we agreed, that a mobile app for the two personas Jan (lumberjack) and Bjorn (sawmill owner), with simple enablement of capturing the log faces and matching them, would allow us to validate our biggest concerns and assumptions. That’s how we got to our MVP.
We kicked off the project and aligned with team goals on what we were going to create and how we would achieve those results. I led the design sprints and as with every project of mine, I started off with some research. Building empathy maps and personas is usually sufficient, but in our case, it was not enough. I jumped straight into sketching and quickly realized that what we were intended to create was an app with the capability of capturing photos at its core. I wasn’t getting a great feeling from the sketches, so I prototyped an interface in my iPhone just to put myself into the shoes of the end-users and arrived at an aha moment. I uncovered a range of additional risks and challenges, which weren’t uncovered at the beginning including:
Lumberjacks often use gloves. Are we going to create an intuitive and comfortable interface for gloved hands?
Is the algorithm going to work with pictures taken not ideally from the 90° angle we hope the user would?
Is this application going to slow down lumberjacks, and how would they feel about it?
I rapidly moved to prototype the application and began with a few rounds of tree face capture screen iterations. A key challenge was to figure out how to meaningfully fit everything together onto a mobile screen. During the research phase, I also found out that the whole log face has to be captured in order to do the correct matching. A circular/oval form of a log or stump face confined to the bounds of a rectangular phone screen with large enough elements to navigate with gloves — that’s the riddle I had to solve. I created a few iterations and prototyped them in InVision to validate with lumberjacks. The first version was developed on the cloud for testing purposes.
Together with the help of the client, we arranged the in-field user testing. In-field user testing enabled me to observe how the solution could be used in the real world by forestry professionals. Conducting the observation allowed me to evaluate how the application fits into the logging process and what key usability issues we needed to solve. Key learnings from this were:
The algorithm successfully works on IBM Cloud.
At an angle lower than 30°, the algorithm barely recognizes the log face. However, because taking the pictures in this position with that angle was not comfortable, this was not an issue for the lumberjack.
Because photo sizes were huge, it slowed the application’s processing speed down quite a bit, which could be a potential issue in the future. Imagine a database with 4 billion images! (That’s how many trees are cut down every year globally.)
The UI was intuitive with gloves and without gloves.
I then repeated the process a few more times to find the most intuitive design for the end-user. Having succeeded in validating the key risks at this point, the team decided to move on and scale the application. I wrapped up my key learnings and other assets for a hand over to another team. It is worth mentioning that doing this alongside development enabled us to build a strong understanding as a team of what features were most important to Tracy of Sweden users and iron out any usability issues.
All in all, it was a challenging but rewarding project in that the idea behind being able to influence and impact illegal logging was exciting and drove me and the broader team to success. Specific reasons that led to our success include:
We engaged and co-created with our end users.
Developers and architects participated in observations, helping us learn together as a team.
The client was engaged and participated in all agile activities, which was crucial for them to be a part of, so we could pivot on the go.
Both the IBM Garage team and the Tracey of Sweden team had so much fun, had a lot of laughs, and spent a lot of time outside in the Swedish forests with fantastic people.
Last but not least, the human-centric approach to solve the problems and design solution was appreciated a lot!
Bring your plan to the IBM Garage
Are you ready to learn more about working with the IBM Garage? We’re here to help. Contact us today to schedule a time to speak with a Garage expert about your next big idea. Learn about our IBM Garage Method, the design, development, and startup communities we work in, and the deep expertise and capabilities we bring to the table.
How about a new sustainable battery solution made predominantly from saltwater that could replace lithium-ion batteries? Yes, that is exactly what IBM Research‘s Battery Lab is currently in the process of testing and further developing. These new batteries are free of any heavy metals. Combining materials science, molecular chemistry, electrical engineering, advanced battery lab equipment, […]
The US Department of Energy’s Oak Ridge National Laboratory (ORNL) has deployed the world’s most powerful and smartest supercomputer built by IBM. The name of the machine is: Summit. It fights against COVID-19. Summit provides scientists with unbelievable computing power, targeting challenges in energy, AI, Human health and more. Summit is a perfect fit for the […]
The universities are currently running their Spring courses. For Claudio Christensen, IBM Academic Ambassador, this is yet another opportunity to connect to the academic world and to the students. ”It is really inspiring to work with the students. We don’t realize how much innovation we do in IBM before you start sharing this with the […]