Evolving the Headspace coaching service
Design manager + IC | 4 month project | Team of ~20
Key results
Increased coach satisfaction
Improved member engagement
New service offering
Clients and members strongly desired video coaching within the Headspace app, but it was unsupported. We aimed to integrate video in a way that would boost member engagement and coach satisfaction, while helping us learn for the future.
Context & challenge
In 2023, we learned from market research and an off-platform service experiment that video was the number one way members sought to connect with a behavioral health coach. It built a needed sense of credibility, connection and trust in the service. In addition, competitors that offered video were eating into our market share.
Background
In 2024, we decided to pilot video coaching as an offering to a select number of D2C + B2B Members using a hybrid model, where members would receive 1 video session per month (starting with their initial consult) followed by additional chat appointments.
Our goal was to figure out what structure would work best for coaches as they balanced running video and chat-based sessions for new and existing members.
Context
Adding video coaching into the current operational mix had many (foreseen and unforeseen) implications. Unlike the 2023 service experiment where a small group of participants all started video sessions on the same day, the new service pilot schedule had to account for staggered start dates, more weeks of care, coaches that were new to video coaching, and of course - the service had to function within our current product infrastructure.
Challenge
Launch a hybrid service model of video and chat-based coaching within an updated platform and service architecture, in order to test and learn what it will take to support a broader commercial rollout in 2025.
Findings from the 2023 service experiment
My role + responsibilities
I drove the coach-facing design for Headspace’s video coaching pilot, uncovering and addressing key workflow challenges through co-design, and delivering a viable MVP within four months. By balancing ambitious ideas with the realities of a tight timeline, collaborating across functions, and mentoring designers, my team and I helped shape a pilot that boosted coach satisfaction and paved the way for Headspace to scale video as its primary coaching medium.
Understanding coach needs
Given that the service experiment in 2023 had been conducted off-platform in a research tool called dscout, we had a lot to learn about creating a video service model within our existing infrastructure. Our first step was to understand coaches top concerns, priorities and questions when it came to video coaching.
We specifically wanted to learn:
Routines & breaks
How do coaches prepare for their sessions and use their breaks?
How do they envision using their time during breaks or prep time in a video world?
What, if anything, do they need to support them in those moments?
What might they need or want in Care Hub?
Session mix & scheduling
What is the ideal ratio of text to video sessions for coaches in a day?
What is the ideal time-blocking strategy for text and video coaching?
Tools, workflow & environment
What, if anything, do they anticipate needing to change during their sessions to support video appointments?
How might their experience differ with writing notes, completing checklists, and scheduling follow-up sessions? How could Care Hub support them here, if at all?
How might their physical (and virtual) environments need to change to best support video sessions?
Methods
We established weekly working sessions with a group of 15 coaches who volunteered to participate in the pilot. I coached and led many of these workshops with our UXR/Service Designer. Below are a sampling of some of our workshops.
Needs gathering workshop
Asked coaches open-ended questions about what they might need before, during, and after video sessions. We also discussed their physical setup at home and what else they may need to be successful.
This set the stage for understanding a breadth of coach needs and the XFN teams that would need to address those challenges.
“Design your schedule” workshop
Invited coaches to design their ideal weekly schedule, balancing chat and video sessions within their current shifts.
Our intent here was to capture a schedule that felt sustainable and rewarding to coaches to help inform Care Operations decisions.
Insights and recommendations
Following discovery workshops, I created a deck to share insights from our research with Product and XFN leads - recommending a coach-backed, “Split Day” schedule, for Care Operations, Monetization, Product and Engineering to continue to explore in collaboration with UX. It also recommended an investigation into coach requests around training, equipment, existing caseload management, and ongoing support.
Ideation & strategy
Based on our research, we summarized coaches high-level needs as “how might we” statements to broaden our thinking as we moved into the ideation phase.
How might we….
...help coaches get a sense of their day and what to expect?
...support coaches in maintaining eye contact while taking notes
...make distinguishing between video and chat sessions easy?
...reduce the strain of note-taking for coaches?
...help coaches prep mentally and physically for a session?
...draw on familiar processes to reduce cognitive load?
Co-design workshop
We ran a co-design workshop to test how adding a Zoom video window would affect coaches’ workflows in Care Hub. Coaches shared how they’d orient their screens to maintain eye contact and trust, highlighting the limited screen space available. This insight led the design team to advocate for external monitors with care leadership. The session also surfaced usability issues in Care Hub’s note-taking, which we addressed during design execution.
Concepting
Next, we created lo-fi mockups of potential solutions, centered around identified coach needs.
I coached our UXR/Service Designer in how to create mocks on top of existing screens or screenshots to get their ideas across quickly, but effectively.
Ideas ranged from small tweaks, such as adding clickable “Join Session” buttons to large initiatives like including AI summarization and auto-population of the notes and summary sections.
Feedback from coaches and engineering
We polled coaches on the need/desirability of each our concepts and collected feedback from engineering on their ability to deliver for MVP.
Many of the larger, significant LOE features, like AI summarization and embedded video, had to be cut for MVP due to our aggressive launch timeline.
We focused design efforts where we could make the biggest impact: providing coaches with visibility into their day, accessing a video session and scheduling video sessions.
Design execution
During the design phase, we iterated on key scenarios, working hand-in-hand with coaches, care operations, design, engineering, and the PM to iterate on and refine our solution. Our collaborative work style with the member-facing UX team also helped uncover additional requirements around member eligibility for video coaching that coaches needed visibility into. This became a part of the final solution.
Video of final prototype
Outcome & impact
Our solution touched almost all aspects of the business and required extensive XFN coordination. When the pilot ended in January 2025 after 3 months, coaches reported feeling greater job satisfaction and fulfillment with their work. Member activation remained low - possibly due to marketing and awareness challenges.
Despite a low pilot member count however, the company decided to fully embrace video as the primary medium for coaching and is further scaling capabilities to support the new service model.
Lessons learned
Even if you build it, they may not come
In our rush to get the pilot operational, we lost sight of partnering with our clients to create strong member awareness campaigns. External marketing and robust onboarding flows are essential to complement new feature/service launches - even for pilots.
Operational research and design is critical to product success
Investing early in co-design activities with coaches and prioritizing the creation of sustainable schedules, ensured the product’s user experience was built on top of a strong service foundation from the get-go. If we had ignored this step, we would likely have designed a UX that supported a poor service.
Pilots should include at least one bold differentiator
Cutting AI-assisted notes and automation got the pilot shipped, but missed an opportunity to test a high-impact differentiator. Experimenting, even with an imperfect version, may have allowed us to learn more about what it would take to successfully scale the service.