Journey
In 2017, I designed Journey around a real model of emotion instead of the single good/bad slider every mood-tracking app defaults to. Mood-logging is built on Plutchik's wheel of emotions, tested with real users, and independently validated by practicing psychologists who told me they wished it were a real, working app so they could use it in their practices. It never made it past a validated prototype, and I'm now scoping a focused native rebuild of the part that earned that reaction.
Highlights
- Rejected the industry-standard good/bad mood slider and built mood-logging around Plutchik's wheel of emotions — a real clinical framework with eight primary emotions and independent intensity, not a binary stand-in.
- Tested three input methods (linear slider, 8-track slider, circular 8-path target) and chose the one that shares its actual shape with the underlying model — validated in testing as the fastest, least confusing part of the whole prototype.
- More than one practicing psychologist, unprompted and years apart, told me they wished this were a real app they could point clients to — the exact audience the model was built to serve, arriving at the same conclusion independently.
- Never shipped past a validated prototype, an honest gap, not one I'm dressing up, which is why the current plan is a focused native rebuild of just the part that earned that reaction from psychologists.
01Project Brief
Journey began as a task-management app before research redirected it entirely. Nearly every mood-tracking competitor modeled emotion as a single good/bad slider, which isn't clinically useful, so I built Journey instead on Robert Plutchik's wheel of emotions: eight primary emotions, each with its own range of intensity.
Project Brief
Journey began as a task-management app before research redirected it entirely. Nearly every mood-tracking competitor modeled emotion as a single good/bad slider, which isn't clinically useful, so I built Journey instead on Robert Plutchik's wheel of emotions: eight primary emotions, each with its own range of intensity.
Journey began as a task-management app, a better to-do list and habit tracker. Research showed people don't want to log tasks in their free time, and redirected the whole concept toward mood, habit, and self-affirmation tracking instead.
Nearly every competitor I looked at modeled emotion as a single good/bad slider, which isn't clinically useful. Journey is built instead on Robert Plutchik's wheel of emotions: eight primary emotions, each with its own range of intensity, not a single axis pretending to capture something as layered as a mood.
02From To-Do List to Mood Log
The earliest surveys killed the to-do-list premise outright — over half of testers abandoned a task app within two weeks. Follow-up research across psychologists, therapists, project managers, and educators reframed the whole question toward mood, habits, and self-reflection.
From To-Do List to Mood Log
The earliest surveys killed the to-do-list premise outright — over half of testers abandoned a task app within two weeks. Follow-up research across psychologists, therapists, project managers, and educators reframed the whole question toward mood, habits, and self-reflection.
It started as a to-do list and habit tracker
The original concept was a better to-do list. Research showed this wasn't worth solving.
Research redirected the whole concept
My earliest surveys killed the to-do-list premise: over 50% of testers abandoned a task app within two weeks, nearly 80% within a month. Follow-up surveys reframed the question toward mood, habits, and self-reflection instead.
Interviews across psychologists, behavioral therapists, art therapists, project managers, educators, and designers confirmed something specific: people are genuinely curious about their own patterns, but existing tools were either clinical and inaccessible, or shallow wellness apps that didn't hold up to that curiosity.
03An App-Shaping Discovery
Mapping every mindfulness and mood app on iOS and Android showed nearly all of them relying on the same shallow good/bad slider. The real turning point came from literature, not app research: Robert Plutchik's wheel of emotions, which became the backbone of the whole mood-logging feature.
An App-Shaping Discovery
Mapping every mindfulness and mood app on iOS and Android showed nearly all of them relying on the same shallow good/bad slider. The real turning point came from literature, not app research: Robert Plutchik's wheel of emotions, which became the backbone of the whole mood-logging feature.
Every competitor used the same shallow model
I tested every mindfulness and mood app on iOS and Android, and mapped effort against value. Nearly all of them used a binary good/bad slider. The turning point came from literature, not app research: the Dalai Lama and Paul Ekman Group's Atlas of Emotions, and the actual breakthrough, Robert Plutchik's wheel of emotions — eight primary emotions plus intensity, not a single axis. That wheel became the backbone of the mood-logging feature.
Personas grounded in more than intuition
I built three personas, Meghan, Robert, and Samantha, from NIMH data, academic mindfulness and self-affirmation research, and my own survey and interview results. Each represents a different real relationship to self-tracking: crisis-driven self-monitoring, habit-driven self-improvement, and reflective, low-stakes curiosity.
04Key Decisions & Tradeoffs
Four decisions shaped the product: testing three mood-input methods and choosing the one shaped like the model itself, scoping a fuller product than mood-logging alone, choosing color for its effect on a nervous system rather than a mood board, and assembling 86 screens into one coherent, testable prototype.
Key Decisions & Tradeoffs
Four decisions shaped the product: testing three mood-input methods and choosing the one shaped like the model itself, scoping a fuller product than mood-logging alone, choosing color for its effect on a nervous system rather than a mood board, and assembling 86 screens into one coherent, testable prototype.
1) Three ways to log a mood, one clear winner
I built and tested a linear slider, an 8-track slider, and a circular 8-path target. The linear slider lost the nuance the whole model was built around. The 8-track slider preserved the data but was slow and fussy to use. The circular 8-path target won clearly, and it's not a coincidence: it's visually the same shape as Plutchik's wheel itself, so the input and the underlying model are the same object.
2) Mapping a fuller product than mood-logging alone
I scoped three connected features beyond mood-logging: activity logging with color-coded sliders, goal tracking with sub-goals, and self-affirmation with a card-based carousel. All three were fully mapped in a task-flow diagram and PRD before any high-fidelity screens existed, more than the idea strictly needs, but a reasonable scope for demonstrating the full product.
3) Color chosen for what it does to a nervous system, not a mood board
Blue calms and reduces stress; a deeper blue signals intellect and consciousness; purple carries emotion and reflection. I moved the palette to a deep lavender-blue deliberately, for what it does to a person looking at it, not because it looked nice. Logo and iconography draw on Tibetan mandala art, Japanese mon emblems, and Western sigils.
4) Assembling 86 screens into one coherent product
The tested prototype ran to 86 individual screens in Sketch, covering every state across mood-logging, activity tracking, goal tracking, and affirmations.
05Testing With Real Users
A full InVision click-through prototype went through several rounds of testing with a few dozen users, deliberately varying how much guidance they were given. Mood-logging through the circular target was consistently the fastest and least confusing part of the whole experience.
Testing With Real Users
A full InVision click-through prototype went through several rounds of testing with a few dozen users, deliberately varying how much guidance they were given. Mood-logging through the circular target was consistently the fastest and least confusing part of the whole experience.
A click-through prototype, tested repeatedly
I built a full InVision click-through prototype and ran it through several rounds of testing with a few dozen users, deliberately varying the level of guidance given. Mood-logging via the circular target was consistently the fastest, least confusing part of the whole prototype, the clearest validation of that specific design decision.
06Outcome & What's Next
Journey never shipped past a validated prototype, an honest limitation I'm not interested in dressing up. More than one psychologist, unprompted and years apart, said they wished it were real. The current plan is a focused native rebuild of just the mood-logging feature, with three connected views onto the same dataset.
Outcome & What's Next
Journey never shipped past a validated prototype, an honest limitation I'm not interested in dressing up. More than one psychologist, unprompted and years apart, said they wished it were real. The current plan is a focused native rebuild of just the mood-logging feature, with three connected views onto the same dataset.
Where it stands
Journey never shipped. It stayed a Sketch and InVision prototype, validated but never built as production software, an honest limitation, not one I'm interested in dressing up.
Real-world validation, unprompted
More than one psychologist, in casual conversation, years apart, told me unprompted that they wished it were a real app they could point clients to.
The rebuild: stripped down, and native
The planned rebuild is a native iOS and Android app, cut back to just the emotion log, the part that actually earned that reaction. Three views onto one dataset:
- Calendar heatmap. A macro view, days colored by dominant emotion, GitHub's contribution graph, but for your inner life.
- Chronological timeline. A micro, journal-like view.
- Aggregate Plutchik wheel. An existential view: a weighted emotional center of gravity over a period. Nobody else is building this view.
Three views, one dataset, a single toggle between them.
What This Demonstrates
I chose a real clinical model over the industry default because the default doesn't hold up to what people actually feel. That decision then had to earn its way through three rounds of input-method testing before the interface matched the model instead of just gesturing at it.
This case study is also an honest one. Journey never shipped, and I'm not dressing that up as something it wasn't. What it did do was get validated, by real users and, unprompted, by the exact clinical audience it was built to serve, which is why it's worth rebuilding rather than shelving.