Skip to the case study

Published research on how often these states are missed: 98% of stress related adjustment disorders go undetected; 71% of distress in cancer patients goes unrecognised; 86% of burned out teachers also meet criteria for depression.

I joined Vinyasa in 2024 as its product designer. People weren't filling in our forms, and the team kept chasing them. I led the fix. This is the rebuild.

A quick check in, a little room to reset, no account required. Pin it to your home screen.

About Vinyasa

Vinyasa helped people reach the right healthcare professionals, including therapists and psychiatrists.

It was working on two problems: people being misdiagnosed, and people never being directed to the right mental healthcare.

Brief: Vinyasa's users weren't filling in its forms

I was tasked with solving one problem: new users weren't finishing the onboarding questionnaire, however often the team followed up.

Project Overview

Client
Vinyasa, Divya Shah, Founder
My role
In house Product Designer and Builder
Year
2024
Duration
21 days
Platform
Progressive Web Application (PWA)

What was going wrong

Every new user had to fill in a long onboarding questionnaire before Vinyasa could help them. Almost every mental health platform works this way, BetterHelp included, so I understood why we did too.

Then I read the therapy session transcripts and talked to people in therapy on Vinyasa. The form looked fine. Its format was the problem: people in real distress often lack the mental capacity to type out long answers.

That was enough to convince the founder to try something else.

How I found this out

I asked the founder for access to Vinyasa's therapists, users, session transcripts and team. Then I asked users to sit down with me, one on one.

  1. I read the transcripts of the therapy sessions

    To see what people were telling us in therapy.

  2. I listened to the sessions, not just the words

    One on one video calls with the same people, about what they were going through.

  3. I heard what the words missed

    Their voices carried feeling the transcripts never showed. The check in grew from that.

Three insights that shaped the design

  1. Long forms ask too much of people who are already struggling

    The people with the least to give were asked for the most. Every extra field was one more reason to stop.

    In Vinyasa: Questions you answer with a tap, and two short spoken tasks instead of an essay.

    Across 93 mental health apps, about 4 in 100 users were still active after 15 days.

    Baumel et al., 2019

  2. Voices carry what words miss

    Transcripts showed me what people said. Their voices on our video calls showed me how they felt.

    In Vinyasa: I suggested we read emotion from people's voices, not only from their answers.

    In studies, voice patterns help tell apart speakers with and without depression. It's a signal, not a verdict.

    Briganti & Lechien, 2025Martínez-Nicolás et al., 2025

  3. Trust has to be designed

    People should know what a score means and where their voice goes. A calm interface isn't the same as a trustworthy one.

    In Vinyasa: Plain words for every score, a clear “not a diagnosis” label, no account and nothing stored.

    Vinyasa's seven areas are its own. They cover similar ground to common questionnaires but aren't one.

    Novopsych

Problem: forms lose the people who need help most

Long forms caused people who were already mentally distressed to drop out of onboarding, so Vinyasa never got the signals it needed to direct them to the right help.

More than 75% of new users dropped off during onboarding. That left users without help, therapists unable to support them, and the business struggling to move forward.

How might we

make a mental wellness check in easy enough to finish and useful the moment it ends?

Solution: a short check in that listens

Vinyasa became a short mental wellness check in that uses audio, a person's own voice, to understand how they're doing. A few taps and two short spoken tasks replace the long questionnaire, so nobody has to type long answers.

This rebuild goes one step further, into sonic therapy. After scoring the check in, it plays calm sound, from soft keys to rain, picked to ease stress and anxiety and give some rest and relaxation (R&R).

  • Describe a picture: a screenshot of the Vinyasa app.
  • The report: a screenshot of the Vinyasa app.
  • Your next step: a screenshot of the Vinyasa app.
  • During a session: a screenshot of the Vinyasa app.

User journey map

  1. 1Arrive

    Doing
    Opens Vinyasa in a browser.
    Risk
    Doesn't know what it will ask of them.
    Design answer
    No account or download. “How Vinyasa works” is one tap away.
  2. 2Rate and consent

    Doing
    Rates their stress from 0 to 10.
    Risk
    Worries about where their voice goes.
    Design answer
    One plain sentence: Gemini hears it, Vinyasa keeps nothing.
  3. 3Check in

    Doing
    Answers 5 short questions.
    Risk
    Typing long answers feels like too much.
    Design answer
    One tap per answer, and Next waits until they're ready.
  4. 4Speak

    Doing
    Reads a passage aloud and describes a picture.
    Risk
    Feels awkward, or the microphone is blocked.
    Design answer
    Both optional, 30 seconds at most. Skip is always there.
  5. 5Report

    Doing
    Reads 7 areas in plain words.
    Risk
    Takes a score as a diagnosis, or panics at a high one.
    Design answer
    Labelled “not a diagnosis”. Very high scores point to a doctor, therapist or helpline.
  6. 6Listen and recheck

    Doing
    Plays a calm sound, then rates their stress again.
    Risk
    Leaves with nothing to do next.
    Design answer
    One sound picked for them, and the change in their own rating.

Research: what Vinyasa was able to uncover

Vinyasa's view of mental wellness came from countless studies, research and the practical experience of its therapists. Together they showed that parts of mental wellness can be tracked and scored through sound.

Human voices carry feelings and emotions, and a trained model can pick those signals out of audio. In the original MVP I used the Thymia API to do this and score users against standard mental health criteria.

The three studies that matter most

  • Music and stress

    • 104

      randomised controlled trials

    • 9,617

      participants

    • d = 0.545

      Psychological stress outcome

    Music eased stress, more in how people felt than in their bodies. d = 0.545 is an effect size, not a 54.5% drop.

    Read the study: de Witte et al., 2020

  • Music therapy and stress

    • 47

      studies

    • 2,747

      participants

    • d = 0.723

      Outcomes related to stress

    Music therapy, a structured approach, had a medium to large effect. An app playing music is not the same thing.

    Read the study: de Witte et al., 2022

  • Why results vary

    • 14

      experimental studies

    • 706

      participants

    • not statistically significant

      Overall effect in the short term

    Lab studies found no clear overall effect of music on stress recovery, so Vinyasa never promises one.

    Read the study: Adiasto et al., 2022

More findings, with their sources

The first three are the figures that open this page.

  • Stress: 98%

    GPs detected only 2 of 110 cases of adjustment disorder, a diagnosis linked to stress. 3,815 primary care patients, Catalonia.

    Fernández et al., 2012

  • Distress: 71%

    Doctors recognised only 29% of patients whose GHQ-12 scores suggested psychological distress, so about 7 in 10 went unrecognised. 2,297 patients, 34 UK cancer centres.

    Fallowfield et al., 2001

  • Exhaustion: 86%

    86% of teachers classed as burned out met criteria for a provisional diagnosis of depression on a standard screening form. 1,386 US teachers.

    Schonfeld & Bianchi, 2016

  • Sleep

    Music improved sleep as adults with insomnia rated it, but not clearly when sleep was measured.

    Jespersen et al., 2022

  • Depression

    With a qualified therapist, music therapy added to usual care eased depression in the short term.

    Aalbers et al., 2017

  • Mental fatigue

    In small experiments, upbeat music helped with mental fatigue more than relaxing music did.

    Ding et al., 2025

  • Natural sounds

    Natural sounds lowered some body measures of stress, but people didn't report feeling less stressed.

    Fan & Baharum, 2024

  • Voice

    In studies, voice patterns help tell apart speakers with and without depression. It's a signal, not a verdict.

    Briganti & Lechien, 2025Martínez-Nicolás et al., 2025

  • Drop off

    Across 93 mental health apps, about 4 in 100 users were still active after 15 days.

    Baumel et al., 2019

  • Short forms

    A short PHQ-9 form of three questions had 98% sensitivity and 76% specificity, not yet checked against interviews.

    Regenstrief Institute

  • Seven areas

    Vinyasa's seven areas are its own. They cover similar ground to common questionnaires but aren't one.

    Novopsych

Prototype: the rebuild, two years after the original

From the 2024 MVP to this rebuild

The original MVP, built in 2024, ran on the Thymia API and a HIPAA compliant language model. It was later acquired by a US mental health startup.

Two years on, I rebuilt it from scratch and made it free to try. Every screen below is from this rebuild. It runs on free tools, so its voice readings are less accurate than the original's.

Information architecture

Vinyasa

  1. Welcome

    • First name, if they want
    • How Vinyasa works
    • Menu: privacy, terms, case study
  2. Check in

    • Stress rating and voice consent
    • 5 questions, one tap each
    • Read aloud (optional)
    • Describe a picture (optional)
  3. Results

    • Report: 7 areas
    • Scorecard to save
    • Your next step: one sound
  4. Listen

    • 22 calm sounds, made in the browser
    • Stress rating again
    • Check in again, or finish
  • Need help now?

    Helplines, one tap away on every screen.

  • Crisis language

    If the voice step hears it, helplines come before the report.

Six Moments in the User Journey

  1. Consent and baseline: a screenshot of the Vinyasa app.

    Consent and baseline

    A 0 to 10 rating of how stressed you feel, and a note on your voice.

  2. Quick check in: a screenshot of the Vinyasa app.

    Quick check in

    Tap an answer, then press Next. Skip is always there.

  3. Read aloud: a screenshot of the Vinyasa app.

    Read aloud

    An optional reading of up to 30 seconds. Skip is one tap away.

  4. Describe a picture: a screenshot of the Vinyasa app.

    Describe a picture

    Describe a calm picture in your own words. Skip anytime.

  5. The report: a screenshot of the Vinyasa app.

    The report

    Seven areas in plain words, all from your own answers.

  6. Your next step: a screenshot of the Vinyasa app.

    Your next step

    One recommended sound, with other sounds a tap away.

Edge cases

The rebuild runs in a browser, on free tools and a tight timeline. Those limits decided what I handled and what I traded off.

Accounted for

  • Microphone blocked or missing

    A plain message, and Skip still leads to a result.

  • Silent or very short clip

    Caught before anything is sent. Try again, or skip.

  • Slow or failed voice analysis

    Two AI models race. If both are slow, the report uses the answers alone.

  • An unexpected AI reply

    Checked against a strict format. The AI's own words are never shown.

  • Crisis language

    Helplines come first, and the report stays one tap away.

  • Skipped questions

    Scored from what was answered, with “Not enough to say” for the rest.

Traded off

  • Time
    Crisis phrases

    English and basic Hinglish only, not yet reviewed by a clinician.

  • Budget
    Voice accuracy

    Gemini's free tier is less accurate than the Thymia API the 2024 MVP used.

  • Budget
    Free quota

    When it runs out, the report falls back to the answers alone.

  • Platform
    Locked iPhone

    iPhone may pause browser sound when the screen locks. The app says so.

  • Platform
    Screening needs a voice

    A skipped voice step can't be screened, so “Need help now?” stays one tap away.

Outcome: what changed, and what a session leaves

The impact of the 2024 MVP

  • Onboarding drop off
    Before
    75%
    After
    17%

    of new users dropped off during onboarding.

  • Check in
    Before
    A long form
    After
    About 3 minutes

    of taps and two short spoken tasks, nothing to type.

  • Build
    Before
    A brief
    After
    An MVP

    designed and built end to end in 21 days.

  • Business
    Before
    Struggling
    After
    Acquired

    by a US mental health startup.

Before, during and after

Screens from one session made with sample data, not a real user.

  1. Before: a screenshot of the Vinyasa app.

    Before

    A 0 to 10 stress rating before listening sets the baseline.

  2. During: a screenshot of the Vinyasa app.

    During

    A calm scene and simple controls keep the sound in front.

  3. After: a screenshot of the Vinyasa app.

    After

    One more rating shows the change. Your own rating, not proof.

What I took from it

A check in only works if it asks very little. A product only works if people trust it. Both came from listening before I designed anything.

This rebuild doesn't claim to fix mental health issues or act as a treatment.

It's a free rebuild of the original MVP that lets people check in on their mental wellness, so you can try the solution yourself.

Go Back to Portfolio