ai-enabled fitness for googleContext
Google’s At-Home Fitness Program, internally known as Motus, began as a product team at the start of its incubate stage in Area120. The team positioned itself as a fitness community offering personalized, interactive workouts accessible to people of all health and wellness levels. Their goal was to solve a gap in the market for at-home, boutique-style fitness classes without the expensive hardware.
Problem
The gap for accessible, at-home boutique-style fitness classes only widened at the onset of the COVID-19 pandemic, as more users came online looking for virtual fitness classes. This presented the Motus team with a unique opportunity to rapidly scale its product offerings to meet the needs of a growing audience of end users. In particular, we were tasked with understanding core barriers to onboarding and adoption as fitness classes scaled online.
Process
Usability Testing: I worked closely with engineering to scale several rounds of evaluative usability testing through the team’s three-, six-, and twelve-month funding reviews with Area120 VC Partners. I recruited and conducted on-site usability tests with 50+ internal testers.
Unmoderated Interviews: I also partnered with the Design Lead to launch 50 remote unmoderated interviews to increase the richness of qualitative insights about our early target population. I kept all of my research findings and relevant recommendations updated in a rolling review deck that the Product Manager used in product reviews and pitch meetings. Storytelling was key to underscoring the product’s rapid growth over time, so I took care to emphasize the weekly quantitative improvements (e.g., CSAT scores, ease of use ratings, etc.) in all of my summaries.
Impact
Onboarding: My usability study findings revealed that users struggled to calibrate their movements during the onboarding process. Participants expressed discomfort and frustration with their inability to align fully with the tall, thin human form represented on the product setup screen. The team leveraged my findings to quickly iterate on the calibration process to expand it from a human form to a more inclusive bounding box that cued users when they were in the correct position. This feature update, alongside additional improvements (e.g., new audio cues, visual improvements like larger font, added color boxes/key points, and enhanced camera scale), reduced friction from the onboarding process, increasing end-user CSAT rankings by +1.4 pts during early beta testing.
Adoption: Users reported high levels of camera fatigue at the start of the pandemic, invalidating an early assumption that end users would want to participate in workouts on camera with instructors and other students to increase accountability and community while using our platform. I led a series of rapid usability tests, generating findings that suggested users wanted their cameras off during their at-home workouts to maintain a sense of privacy and to get a break from being on camera all day. These insights influenced how the team scaled their social features, including a shift to enable community through chat engagement. At their 15-month review, the team found that Monthly Registered users who engaged with the chat feature returned the following month to register for additional classes.
Reflections
Where do I begin? This product was my gateway to user experience research at Google and remains one of the most gratifying projects I’ve worked on to date. Quite literally overnight, I went from holding back-to-back usability testing sessions onsite at Google’s UX Lab in downtown San Francisco to moderating experiments remotely at the onset of COVID-19 lockdowns in early 2020. Solving the pain points our users were sharing with us—an inability to engage in at-home workouts that required loud movements, a need for form correction and proper technique, connection to others around a shared interest in a time where people were more physically isolated than ever—felt human-centered in every meaning of the word. While the research sprints were demanding at the time, I loved seeing repeat users return to the product week after week and submit improved CSAT scores as the team worked tirelessly to address their feedback in earnest. This experience taught me how incredibly insightful, curious, and HUMAN users can be when invited into the design process. Thanks to them, this product has seen so much success and remains a staple of Google culture today.
“Every instructor at Motus has their own unique style. Most of the instructors did a callout for emojis, and that worked really well for us. At the end of every set, we would send out these little emojis and others in the class would send out theirs, too, and that gave us a feeling as if we’re in a real group.”
