Climaider thesis

While working as CTO of Climaider, I wrote my thesis at ITU. Climaider allowed me to conduct research using large amounts of quantitative data, as a significant number of users used our app. This allowed me to draw conclusions and make decisions based on split tests with thousands of users for robust statistical measurements.

Thesis abstract

With an increased focus on climate change, society is in need of sustainable solutions that can help consumers filter out the vast amount of information into a concrete set of instructions regarding how to live sustainably, and further help them to take action where it is necessary. Climaider is an app with the purpose of making climate action accessible to everyday consumers. The app revolves around a set of gamified challenges, which all contribute to habit changes, education, and concrete actions all concerning climate change and sustainable living. It is important for Climaider that users complete as many of these challenges as to support Climaider’s primary goal of CO2 reduction/removal. At the beginning of this thesis collaboration, the first version of the app was launched for iOS and Android. This initial app version is referred to as the baseline app and will be the foundation for improvement. This thesis presents a quantitative study based on improving challenge-based gamification apps, through different methods of design. Three main areas of redesigns have been proposed as solutions. First a general set of improvements, including simplification of UX, haptic feedback, and animations. Secondarily, a Locked Flow (LF) is designed, where users must unlock the challenges in sequence. This design is based on a Black Hat Core Drives, specifically, Scarcity and Impatience presented in the Octalysis framework (Chou, 2014). Finally, a design also based on a Black Hat Core drive has been implemented, named Continues Chat Flow (CCF) which is a Rule-Based Task-Oriented chatbot that binds challenges together in a large narrative inspired by the continuity of social media as portrayed by (Jovicic, 2020). The first mentioned general improvements have been shown difficult to compare with the baseline app since a lot of independent variables are in play, but when omitting biases, the overall completion rate from the design changes has increased by 333%. On the contrary, there seems to have been a decrease in the conversion rates when comparing overlapping challenges from the baseline app to the redesign. Since there has been a drastic increase from 5 to 23 challenges between the baseline app and the redesign, it could be a result of the consumer research concept of scarcity and abundance, where an abundance of products will lessen the value of each specific item (Zhu & Ratner, 2015). LF was found to perform significantly worse without CCF than when the two concepts were used in combination. Both design concepts have difficulties dealing with complicated challenges. But there is found indications that users tend to complete close to all challenges when involved with these designs.