Energy conscious eating and grocery shopping
Product & Program Management
Design of Experiment
The project started with the question - how can we drive energy-conscious behavioral changes for people who care but are not sure what to do? We focused on grocery shopping and eating, hoping to bring awareness to people to adopt a plant-rich diet.
Project Partner: Jeff Kassab
At the beginning of our design process, we prototyped two simple deployments to understand how much people know or care about energy consumption of the foods we purchase and eat. The first deployment is sending sassy energy related messages with AI voice to our friends, such as the one below, and see how they react. The key takeaway is that people do not enjoy being peached to do/not do something, but enjoy random fun facts. The second deployment took place in a local grocery store where we put carbon footprint of foods on sticky notes. When showing to people, none have any idea of what units like BTU or kg of CO2 means, and they are more interested in comparative values than absolute.
We Zoom interviewed with people before and after they do grocery shop about their shopping patterns, and learned that people do not like feeling judged for what they buy, especially when they consume a lot of meat. They like to feel encouraged and be given an actionable solution. Reliable and personalized data for their own use case is also important to them.
We did landscape analysis on energy conscious behavior apps that are in the market, as well as diet apps that are influencing the way people eat.
There aren’t many easy to understand food energy ratings out there so we made our own based on research data.
Based on earlier deployments and interviews, we journey mapped four steps of the grocery shopping journey, and created a prototype that introduces carbon footprint of foods via various touch points and promotes more energy conscious grocery shopping and eating behaviors.
From the grocery prototype, we decided to further zoom in to study what elements might promote behavior change in the context of eating via a food study. We recruited 8 non-vegetarian Stanford students, and asked them to send us pictures of what they eat for breakfast, lunch, dinner and snacks for an entire week. We gave them 1000 carbon points at the beginning, and after they send us data, we calculate what’s the carbon footprint of the food they eat, and deduct points from 1000. At the end of each day, we tell them how much points they’ve used for the day, and how much they have left.
As we mentioned, the goal of the study is to study which intervention elements would trigger behavioral change towards food. We specifically test out two variables, competition and encouragement. So for some participants, besides sending them their carbon points expenditure and remaining, we tell them their ranking among all competitors, like you are among the top 20%. For some, we send them encouraging notes, acknowledging they are making great foot choices, if we noticed they consumed soy milk instead of dairy milk, or eat chicken instead of beef, and we also send them sustainability data of these foods.
Above is a portion of the pictures we got. It’s really interesting to see such personal information of people, and learn about their habits, patterns, and culture. We learned a lot about who they are and what their lives are like through these images and interactions. We found that people are competitive. They would like to perform well when knowing they are being compared with other people, though knowing who they compare with matters a lot. If they are compared with people they are related with, they would take the game more seriously, though not doing well in the game would result in abandoning the game completely. People enjoy positive encouragement, but the message would need to be personalized to feel genuine and trustworthy.
Here are a few non-obvious insights we found.
From the insights above, we did a brainstorming and came up with a few next steps. We are super excited for what comes next. This is an ongoing project, will be continuously updated here.