A case study on redesigning onboarding for Lightyear Health — a telehealth membership app — after research diagnosed a 58% drop-off between signup and a member's first booked visit.
Try the onboarding flow ↓Lightyear Health is a telehealth membership app: members get a dedicated care team, book virtual and in-person visits, and track health goals in one place — positioned as an alternative to the fragmented experience of juggling separate portals for insurance, scheduling, and records.
New signups were dropping off before they ever felt the product's value. Onboarding opened with a long intake form — insurance details, a full health history, emergency contacts — before a member had spoken to anyone or seen anything Lightyear Health could actually do for them. Product analytics showed 58% of signups never completed onboarding; interviews suggested it wasn't the length of the form alone, but its timing.
Redesign onboarding so members experience something valuable — meeting their care team, seeing what a visit actually looks like — before being asked for sensitive information, break the remaining intake into short, honestly-labeled steps instead of one wall of fields, and make clear exactly what health data is collected and why, before asking for it.
Health data carries more weight than most onboarding forms ask people to hand over, and it showed in the drop-off pattern: members backed out fastest at the two steps that asked for the most before giving anything back — insurance and health history. Trust had to be earned before it was spent.
A five-stage double-diamond process, run end-to-end with weekly syncs alongside one PM and two engineers.
Research & competitive analysis
Personas & journey mapping
HMWs, IA & user flows
Wireframes to hi-fi UI
Usability testing & iteration
Four opportunities emerged for closing the gap between signing up and feeling like a Lightyear Health member.
Let members meet their matched care team and preview a visit before any insurance or health history is required.
Split the long form into short, clearly-labeled steps with visible progress, instead of one wall of fields.
Explain exactly what health data is collected, why, and who sees it — before asking for it, not after.
Anyone stuck on a step can reach a real person immediately, instead of hunting for a support link.
Understanding exactly where and why signups stalled, and what "trust" needs to look like before someone hands over health information.
1:1 moderated interviews with people who started onboarding but never finished it.
Product funnel analysis plus a follow-up survey validated interview themes at scale.
A structured review of three telehealth products to map strengths, gaps, and openings for differentiation.
TDTeladoc Health |
OMOne Medical |
RORo |
|
|---|---|---|---|
| Description | General virtual urgent & specialty care, billed per visit or via insurance | Membership primary care blending virtual and in-person visits | Condition-specific care (e.g. weight, sexual health) with async messaging |
| Strengths | Broad specialty coverage, fast time-to-first-visit | Strong in-person + virtual relationship, same-day booking | Very low-friction intake for a narrow condition |
| Weaknesses | Insurance & payment collected immediately after signup, feels transactional | Membership fee is a paywall before any value is felt | No lasting care-team relationship, narrow scope only |
| Onboarding approach | Insurance form is the first screen past signup | Payment required before a care-team preview | Minimal intake, but no relationship shown at all |
The gap isn't another virtual-care app — it's one that combines Teladoc's specialty breadth, One Medical's relationship-first feel, and Ro's low-friction intake, without their respective weaknesses: transactional onboarding, a paywall before value, and no lasting relationship. Lightyear Health's opportunity is to let members feel the relationship first, and ask for insurance and health history only once trust is already underway.
Synthesizing research into shared patterns, then into two personas that anchored every design decision.
"I'll give you my health history once I know you're worth it."
"If something goes wrong, I want a phone number, not a chatbot."
Scenario: Meera signs up for Lightyear Health and books her first visit.
| Phase | Awareness | Care Team Preview | Health Snapshot | Insurance & Confirm | First Visit Booked |
|---|---|---|---|---|---|
| Action | Sees Lightyear Health recommended by a colleague | Browses matched care-team profiles, no account required | Answers a short, chunked health snapshot in under two minutes | Adds insurance once she understands exactly what's shared and why | Gets a warm, specific confirmation — not a generic receipt |
| Feeling | 🙂 Curious | 🙂 Reassured | 😌 Comfortable | 😊 Confident | 🎉 Cared for |
| Opportunity | Lead with the care team, not the feature list, in marketing | Let the relationship be felt before any data is required | Keep each step short with a visible sense of progress | Explain exactly what's accessed and why, in plain language | Confirm with specifics — who, when, what to expect — not just "Done" |
Reframing research insights as "How Might We" questions, then structuring the product around them.
→ Let members meet their matched care team and preview a visit before insurance or health history is required.
→ Split it into brief, clearly-labeled steps with a visible progress indicator, not one continuous form.
→ A plain-language "why we're asking" note attached to every sensitive question, before the field itself.
→ Auto-save every step so members can leave and resume exactly where they stopped, no restarting.
→ A persistent, one-tap "talk to someone" option visible on every onboarding screen, not buried in settings.
→ A specific, human confirmation naming the care team and next step, instead of a generic "You're done."
Flow 1 · Completing the health snapshot step
Flow 2 · Booking a first visit after meeting the care team
From low-fidelity structure to a visual language built around trust, calm, and quiet celebration.
Not every idea survived contact with users. The gamified onboarding tracker below is the clearest example — and the reasoning behind cutting it shaped the whole flow's tone more than almost any other decision.
Complete a step today to keep your streak alive!
Progress badges and streak nudges are a familiar pattern for driving completion in consumer apps, and an early sketch testing badges for each finished step scored well for "motivating" in isolation.
A plain progress bar and step count kept the "visible progress" benefit that made the streak idea appealing, without the game mechanics or the pressure to rush.
Teal
#1D5C56
Coral
#D97757
Slate
#7C8783
Paper
#EFF3F1
Rust
#B8503F
Deep teal and warm coral replace the typical blue-and-white clinical palette to feel human rather than sterile; the rust accent is reserved only for genuine warnings, never routine onboarding prompts, so it keeps its meaning. Icons use rounded strokes to match Fraunces' soft serif terminals.
High-fidelity screens from the final prototype, each tied to a research insight above.
We ask this so your care team can check for interactions before your first visit — not for insurance or marketing.
Dr. Priya Shah, Tue Aug 4 at 10:00 AM. She'll have your health snapshot before you log in.
Every element on the flagship screen traces back to a specific research finding.
Matches the "one step at a time" preference from interviews — nothing to parse a full intake form for.
Members see exactly how much is left, addressing the "how long will this take" uncertainty from research.
Basic info first, sensitive steps later — the order itself is a product of the funnel-analysis findings above.
The step in progress uses the accent color — color signals a decision point, not decoration.
All five destinations are reachable one-handed — see the reachability study below.
2026 digital health UX is judged as much on who it excludes as on what it enables. Three checks ran alongside every screen.
Primary actions, the progress card, and the bottom navigation all sit in the bottom two-thirds of the screen — the zone a thumb reaches without a grip shift, matching the mostly one-handed usage the interviews described.
Before asking for health history, Lightyear Health shows exactly what's collected, who on the care team can see it, and a revoke-access control in the same screen — not buried in a settings menu three taps deep.
This isn't a screenshot — it's a working slice of the actual flow tested in the study below. Tap a provider, then book or pick another time.
Matched based on your goals and location. 10:00 AM · virtual or in-person.
Dr. Priya Shah, Tue Aug 4 at 10:00 AM. We'll share your health snapshot with her before the visit.
Your visit with Dr. Priya Shah is confirmed at the new time. We'll send a reminder beforehand.
Providers surface based on your goals and location, with a match reason attached.
Tap in to see experience, availability, and why they were matched — never a black box.
Either path confirms a specific visit — no settings menu required.
Moderated testing to validate whether the value-first ordering and progressive intake actually reduced drop-off in practice.
8 participants (4 matching each persona), moderated remote sessions via Maze, testing four core tasks: browsing the care-team preview, completing the health snapshot step, understanding the data-use explanation before adding insurance, and booking a first visit.
Session recording · Task 2 — completing the health snapshot step
This reaction — relief once a sensitive question was explained, paired with an instinct to abandon an unexplained one — showed up in 5 of 8 sessions and directly shaped Finding 01 below.
5 of 8 participants didn't register the explanation until they reread the screen — it sat below the field instead of before it.
Moved the reason above every sensitive question, so it's read before the field, not after.Several participants expected to book a visit from the Health tab, since both showed similar progress-style cards.
Gave Visits a distinct calendar-forward icon and layout, separate from the Health metrics view.Testers trusted "98% match, based on your goals and location" more than a plain "Recommended" label with no reasoning attached.
Added the specific match reason next to the score on every care-team card, not just a badge.Validated through iterative usability testing ahead of a planned engineering handoff.
Timing mattered more than length — the intake form itself wasn't the real problem, asking for it before any value was felt was. I also learned how much a single sentence of context changes how a question lands: "so your care team can check for interactions" turned a suspicious field into a reasonable one, and participants noticed the difference immediately.
Next steps include testing the progressive intake with a larger, more varied sample, exploring a lightweight way to resume onboarding from a text message reminder, and partnering with engineering to validate auto-save reliability before this replaces the current flow.