# Feature Priorities

*Updated: 2026-04-05*

Prioritised by: **demo value** (can we show it?) × **user value** (does it matter?) × **build effort**.

## Tier 1: Demo-Critical (Build Now)

These are the features that make the difference between "slides about an app" and "here, try it."

### 1. Journey Walkthrough Runtime
**What:** A lightweight app that plays journey definitions as connected, animated screen sequences using existing Storybook components. Click through a complete user flow — persona selection → onboarding → daily use → crisis — with transitions and data persistence.

**Why:** This is THE demo. Everything else is supporting material. An investor watching someone walk through Kai's first crisis activation will understand the product instantly.

**Effort:** Medium — components exist, journey graphs exist, need glue code + transitions.

### 2. Fake API Layer
**What:** Static JSON endpoints (or in-memory data) that respond to journey effects. Check-in history returns 14 days of fake mood data. Safety plan loads a pre-built plan. Angel profiles have names and avatars.

**Why:** Screens need data to feel alive. A streak result that says "7 days" with a sparkline of moods is 10x more compelling than an empty state.

**Effort:** Small — just JSON files matching the API shapes defined in journey effects.

### 3. Demo Scenarios (4 scripted walkthroughs)
**What:** Pre-configured entry points for each persona with appropriate fake data:
- **Kai's first week** — onboarding → 3 check-ins → safety plan creation → first crisis
- **James accepts an invitation** — angel invite landing → learn role → set boundaries → see Kai's shared data
- **Danny's recovery journey** — onboarding → sobriety tracking → trigger management
- **Priya's circle** — connector onboarding → adding people → managing check-in frequency

**Why:** Investors need stories, not features. Each scenario is a 2-3 minute narrative.

**Effort:** Small — scripting data + entry points for existing journeys.

### 4. Screen Transitions & Polish
**What:** Smooth transitions between journey nodes. Slide/fade animations. Loading states. Micro-interactions on buttons and form elements.

**Why:** The difference between "prototype" and "product" is motion. Static screen swaps feel like a slideshow. Animated transitions feel like an app.

**Effort:** Medium — CSS transitions + orchestration in the walkthrough runtime.

---

## Tier 2: Demo-Enhancing (Build Soon)

These make the demo deeper and more impressive but aren't blockers.

### 5. REQ-001: Emotional Readiness Check
**What:** Add the decision node before trigger editing in the safety plan journey. "Before we look at your triggers, how are you feeling right now?"

**Why:** This is the single best example of Angels' research-driven design philosophy. It shows investors that the app is designed with clinical insight, not just UX best practices. It's the "wow" moment in a demo.

**Effort:** Small — component exists, need to wire the decision node and 3 routing paths.

### 6. Insights & Trends Screen
**What:** A screen showing mood trends over 7/30 days, journal snippets, and contextual reflections. Uses fake historical data.

**Why:** Shows the "so what" of daily check-ins. Without insights, check-ins look like data collection. With insights, they look like a therapeutic tool.

**Effort:** Medium — sparkline/chart components exist (Sparkline atom, FrequencyChart), need composition into an insights view.

### 7. Consent Wizard with Visual Sharing Rules
**What:** The full consent flow — choose metrics, choose who sees what, review summary. Shows the privacy-as-product philosophy.

**Why:** Differentiator. No competitor has granular consent flows with coercion detection. This is the "privacy isn't a feature, it's the architecture" moment.

**Effort:** Small — journey and components exist, just need walkthrough integration.

### 8. Crisis Flow End-to-End
**What:** Complete crisis activation → angel alert → real-check decision → post-crisis recovery timeline (days 1, 3, 7, 14).

**Why:** The crisis flow is the emotional peak of any demo. Showing the full arc from "I'm not okay" to "14 days later, here's how you're doing" is powerful.

**Effort:** Small — all components exist, need post-crisis screens connected.

---

## Tier 3: Alpha Requirements (Build for Phase 1)

### 9. Authentication (Better Auth)
Magic link login. No passwords. Session management.

### 10. Database Schema (Drizzle + Neon)
Users, check-ins, safety plans, angel connections, consent rules, crisis events.

### 11. Check-in API
POST mood + journal entry. GET streak + history. Real data persistence.

### 12. Safety Plan CRUD
Load/save safety plan sections. Version history for accountability.

### 13. Angel Invitation System
Generate invite links. Accept/decline flow. Connection management.

### 14. Push Notifications
Check-in reminders. Angel alerts during crisis. Nudges.

---

## Tier 4: Beta / Premium (Build for Phase 2-3)

### 15. Guardian Persona Backend
Guardian-specific views of shared data. Check-in monitoring.

### 16. Recoverer Persona Features
Sobriety date tracking. Trigger management. Sponsor connection.

### 17. Care Provider Dashboard
Aggregated insights. Pattern alerts. Integration APIs.

### 18. Wearable Integration
Apple Watch complications. Passive mood signals.

### 19. Data Export
Full data export in standard formats. Account deletion.

### 20. REQ-003: Streak Context
Mood trend visualisation. Journal snippets. Contextual reflections.

---

## Decision Log

| Decision | Rationale |
|----------|-----------|
| Demo walkthrough before real backend | A connected demo proves product vision faster than a half-built app. Investors fund vision + execution evidence, not databases. |
| Survivor persona first | Core value loop (check-in → plan → crisis → angel) is most complete for survivors. Other personas depend on this foundation. |
| Free crisis features forever | Ethical requirement. Also smart business: the free tier IS the viral loop. Paywalling crisis features would destroy trust and growth. |
| Therapist partnerships before advertising | Trust-based distribution for a trust-based product. Advertising mental health apps is expensive and low-trust. |
| UK first | Smaller market = faster feedback loops. NHS partnerships are accessible. Regulatory environment is clear. |