AC4D

View Original

Rapid Experimentation for One Village

Background

Over the past six months, myself and two other students have been researching the problem space of food insecurity in Austin, TX. According to the USDA, food insecurity is defined as "a lack of consistent access to enough food for every person in a household to live an active, healthy life." Research has shown that 1 in 7 people in Travis County experience food insecurity and Central Texas has a higher percentage of food insecure households than most of the country.

People experiencing food insecurity face numerous barriers when seeking assistance. Our research led us to identify the ATX Free Fridge, a Mutual Aid organization, as an effective way to directly help those experiencing food insecurity. The ATX Free Fridge reduces barriers to access by not requiring individuals to meet any qualifications to receive assistance.

While the Free Fridges are a way to directly help individuals in need and strengthen the community overall, they are often times empty. We found this is due to people dropping off expired foods, it not being convenient, and it being hard stay motivated and take the time to help.

Two Free Fridges in Austin, TX

After conducting research and brainstorming possible solutions, we developed One Village, a product that aims to help others while fostering connections within the local community.

High-Fidelity of One Village, the concept we are testing.

Next, we focused on determining if our idea would be something that people would actually want and use. To test this, we devised a simple way to gather feedback before investing significant time, money, and effort into the final project.

As we planned how to conduct rapid experiments, I felt a sense of excitement and optimism, knowing that our approach could alter the course of our project and being open to that possibility.


Experiment 1: Unveiling Interest and Engagement

Our initial experiment tested the deadliest killer assumption: is there interest in our idea? We hypothesized that people will opt in for reminders about donating to the free fridge and will appreciate education about what items are specifically needed. Depending on the results of this experiment, our subsequent experiments or product idea may change entirely.

We conducted a survey to gauge interest in receiving a weekly grocery list of items needed at the Free Fridge. We obtained feedback on this idea from the existing ATX Free Fridge network and measured the response rate as the number of people who signed up for this feature.

Initially, we anticipated receiving 25-50 responses; however, we were pleasantly surprised to receive 148 participants within 24 hours.

Survey results for how often people wanted reminders

Survey results indicating drop off preferences

Our learnings after running this experiment were:

  • People reported wanting to drop off donations themselves.

  • Of the 148 participants, 65 preferred weekly reminders and 70 preferred monthly reminders.

  • Participants preferred sharing their email address rather than their phone number.

Through this, we learned that one of our main features, drop-off locations for donations, was unnecessary at this time as people reported wanting to drop off donations themselves.


Experiment Two: Fine-Tuning Donation Triggers

Experiment two aimed to determine whether providing specified donation lists would encourage people to contribute more often. We provided a mix of specific lists (oranges, broccoli, etc.) and generic lists (vegetables, fruit, etc.). We asked people to send us texts or emails when they donated.

Midway through the experiment, we realized that familiar donors were engaging through ATX Free Fridge's Instagram rather than reaching out to us. This prompted us to adjust our tracking methods. The ATX Free Fridge would send us donation confirmations when they aligned with what was included on the list.

Seeing donations rolling in was really exciting. However, it required a lot of work to ensure that we were texting the correct people the correct lists. The organization of who was texting and emailing whom required a lot of coordination. Luckily, one of our teammates was very skilled in Google Sheets, which helped with organization.

Our learnings after running this experiment were:

  • Preference for specific lists over generic lists

  • Texts lead to more engagement over emails

  • Half of the confirmations were shared over Instagram

Through tracking donations, we found that the lists were helping people and encouraging new donors to contribute to the fridge. This success led to the decision to move forward with texting lists about what is needed for the fridge, and to look into text automation services for the future.

Texts to and from our participants— showing recommended lists and donations


Experiment Three: Cultivating Engagement Through Recipes

For experiment 3, our assumption was that people would want to connect more with those they were helping and sharing co-created recipes alongside storytelling would increase their motivation and connection to the cause.

To test this assumption, we began by asking if people would even spend extra time cooking for the Free Fridge. We provided a simple recipe and asked people if they would cook a meal this week instead of donating items.

The experiment involved sending out a request along with a short recipe, and the metric used was how many people made the recipe and sent a confirmation of doing so (text or photo). The group also conducted A/B tests, comparing recipe requests versus requests for home-cooked meals, as well as grocery list-based donations versus home-cooked donations.

As of now, we have not received any feedback about people making the home-cooked meals.

In retrospect, we feel like this was not the correct method of testing our initial assumption, as we wanted to know if this would build a connection and increase motivation, rather than simply if people would spend extra time for the cause. Moving forward, we believe we can devise a new experiment that better supports our hypothesis, potentially emphasizing the storytelling aspect more than the recipe itself.


Conclusion

Over the course of several experiments, we have gained valuable insights into how to improve our product, One Village. Through these experiments, we have learned that suggested, specific donation lists hold immense value and that text-based communication and curated, specific lists are key components that demand focus. We were surprised to learn that people do not care about having a drop-off location, which has helped us to further refine our approach.

Moving forward, we want to test our concept with people who are accessing the resources from the fridge to see how we can better design for their needs, figure out how to better concept test for how storytelling can improve motivation, and research third party, automated texting services

Overall, the process has been exciting, and we have been fortunate to have some successes along the way. Our experiments have reshaped our concept of engagement and donation strategies, and the insights have given us confidence and direction to help us to create a product that truly meets individuals needs.