Tiny Prompts to build Big Relationships
Overview
Over a course of 5 days, we text prompted our participants from the diary study to contact the relative they hoped to keep in touch with more often. We decided to focus on college students who want to keep in touch with their family members on a more regular basis.
Pre-Interview & Setup
As part of our pre-intervention interview, we asked participants for when they had downtime this week (so we can text them during downtime) as well as 5 reasons they hoped to stay in touch with their family member (so we can remind them of their words in our text prompts). For the sake of consistency across participants, we pre-wrote encouraging text prompt templates for each day.
Synthesis
Despite the varying nature of relationships (grandparent, sibling, difficult relationship), we were able to see a significant increase in contact frequency due to this intervention. As you can see in the Data & Key Findings charts below, every single participant increased contact frequency with their selected relative. This led to other feedback loops in our System Map, such as relatives beginning to initiate contact to reciprocate.
Highlights
Participants found the personalization really effective – by prompting them during their downtime, using messages that they helped generate (they sent us reasons why they wanted to keep in touch at the beginning of the study), and adding content suggestions (we would ask them to text their relative a photo of Lake Lag, for example), we minimized friction and helped them feel like the prompting was a human, custom-made solution that worked. In one case, one participant felt particularly rewarded when their sister initiated contact first at the end of the study, meaning their efforts had paid off in just a short period of time.
Lowlights
Some participants found the frequency of every day prompting to be overwhelming, as their initial goal was to contact their relatives at most once a week. We did daily prompts due the short nature of this study, but for the actual app we’d likely adjust the frequency based on user preference – potentially at 2X their desired frequency (2 prompts a week rather than 5 prompts a week for weekly users) to make sure they’re amply reminded but it doesn’t get annoying.
Other Considerations
A confounding variable was Lunar New Year, as the holidays is traditionally a time for cross-border family members to call and catch up with each other. This is interesting for our study because 2 participants cited the timing as an impetus for them to begin keeping in touch with their family members regularly. This is also similar to our findings in our longitudinal study, which suggests that the timing of our product may matter to deliver the most effective results.
Data & Key Findings
Here is an overview of our participant data from each day.
Here are our key findings from doing the intervention study and our post-study interview surveys.
System Map
Here are behavioral loops we found.
Next Steps
We’re going to further analyze this data and see how we might continue to emphasize positive feedback loops while reducing negative ones. We’re also considering a few product name ideas, such as InTouch and Closer, and perhaps using technologies such as Twilio to help automate our prompting since it ended up being a high manual load on the team this week.
