Wearables provide incredible precision, but they have a fatal flaw: only 30% of users own them. Smartphone-based biomarker collection solves the coverage problem while maintaining clinical-grade accuracy for key health metrics.
The 30% Problem
Why This Matters for Product Builders
If your app requires a wearable (Fitbit, Apple Watch, Oura, Whoop), you're automatically excluding 70% of potential users. For consumer apps, this is a non-starter. For insurance programs, it means 70% of members get zero monitoring.
Coverage Visualization: Wearable-Only Apps
100 users sign up for your app:
Green dots (30) = Users with wearables (monitored)
Gray dots (70) = Users without wearables (no data)
The Smartphone Solution
📱 Wearable-Only Approach
Coverage: 30%
User acquisition: Requires device purchase ($200-$500)
Onboarding friction: High (device setup, syncing, charging)
Data precision: Excellent (HRV, detailed sleep stages)
Best for: Enthusiast apps, high-commitment programs
📱 Smartphone-Based Approach
Coverage: 100%
User acquisition: Zero additional cost
Onboarding friction: Low (just permissions)
Data precision: Good-Excellent (biomarker-dependent)
Best for: Consumer apps, insurance, mass-market products
Everyone has a smartphone. Not everyone has a wearable.
Smartphone-based biomarker collection uses built-in sensors (accelerometer, gyroscope, ambient light) plus platform APIs (HealthKit, Health Connect) to track activity, sleep, and behavioral patterns. No additional hardware required.
What Biomarkers Can Smartphones Track?
Activity Biomarkers (10 total)
Smartphone capability: Excellent
- Steps: Accelerometer-based (±3% accuracy vs wearables)
- Active hours: Movement pattern detection
- Distance: GPS + step-based estimation
- Sedentary time: Screen time + inactivity tracking
- Exercise sessions: Activity intensity classification
Method: Built-in accelerometer, gyroscope, GPS
Sleep Biomarkers (13 total)
Smartphone capability: Good (basic) Excellent (with wearable)
- Sleep duration: Screen lock + charging patterns (good)
- Sleep start/end: Device usage + motion detection (good)
- Sleep regularity: Pattern analysis over 7-30 days (excellent)
- Sleep stages: Requires wearable
- Interruptions: Requires wearable
Method: Smartphone sensors + optional wearable integration
Body Biomarkers (8 total)
Smartphone capability: Manual input or wearable
- Weight: Manual entry or smart scale sync
- BMI: Calculated from weight/height
- Body fat %: Smart scale integration
- Waist circumference: Manual entry
Method: User input or connected devices (scales, tape measures)
Vitals Biomarkers (13 total)
Smartphone capability: Requires wearable
- Heart rate: Wearable PPG sensor
- HRV: Wearable PPG sensor
- Resting HR: Wearable overnight tracking
- Blood oxygen: Wearable SpO2 sensor
- Blood pressure: Dedicated BP monitor
Method: Wearables or medical devices (no smartphone alternative)
Reproductive Biomarkers (10 total)
Smartphone capability: Excellent (tracking)
- Cycle tracking: Manual entry + prediction algorithms
- Period start/end: User-reported
- Ovulation prediction: Algorithmic (based on cycle history)
- Basal body temp: Requires thermometer
Method: User input + pattern recognition
Behavioral Patterns (Sahha-specific)
Smartphone capability: Excellent (unique)
- Screen time: Device usage analytics
- App usage patterns: Digital behavior tracking
- Social interaction: Call/message frequency
- Circadian rhythm: Device activity patterns
- Location patterns: GPS-based routine detection
Method: Smartphone sensors + OS APIs (permission-based)
Precision Comparison: Clinical Validation
| Biomarker | Smartphone Accuracy | Wearable Accuracy | Research Source |
|---|---|---|---|
| Step Count | ±3% (accelerometer) | ±2% (PPG + accelerometer) | JMIR mHealth (2019) |
| Sleep Duration | ±15 min (pattern detection) | ±10 min (actigraphy) | Sleep Medicine Reviews (2020) |
| Sleep Regularity | ±5% (7-day average) | ±5% (7-day average) | University of Otago (Sahha study) |
| Activity Level | ±8% (MET calculation) | ±5% (HR-based MET) | JAMA Network (2021) |
| Sedentary Time | ±10 min/day (screen + motion) | ±8 min/day (continuous tracking) | BMC Public Health (2020) |
| Heart Rate | N/A (no sensor) | ±3 bpm (PPG sensor) | Circulation (2019) |
| HRV | N/A (no sensor) | ±5ms (PPG sensor) | European Heart Journal (2020) |
| Sleep Stages | N/A (requires motion + HR) | ~70% agreement (vs PSG) | Sleep (2019) |
Key Takeaway: Smartphones Excel at Behavioral Biomarkers
For activity, sleep duration, and behavioral patterns, smartphones achieve near-wearable accuracy (±3-8%). For vitals (HR, HRV, SpO2), wearables are required. The choice depends on your use case.
The Hybrid Approach: Best of Both Worlds
Sahha's Hybrid Strategy
Sahha supports both smartphone sensors and wearable integration:
- 100% coverage: All users get smartphone-based monitoring
- Enhanced precision: Users with wearables get additional vitals (HR, HRV, SpO2)
- Unified API: Same data format regardless of source
- Automatic fallback: If wearable disconnects, smartphone data continues
This approach maximizes coverage while still leveraging wearables when available.
Coverage Scenarios:
| User Type | Data Sources | Biomarkers Available | Use Case Fit |
|---|---|---|---|
| Smartphone Only (70% of users) | Accelerometer, GPS, screen time, platform APIs | Activity (10) Sleep (5 basic) Behavioral (custom) | Mental health, insurance, behavioral coaching |
| Smartphone + Basic Wearable (20%) | Above + Fitbit/Garmin/Mi Band | Activity (10) Sleep (13 full) HR, Steps | Fitness apps, wellness programs |
| Smartphone + Premium Wearable (10%) | Above + Apple Watch/Oura/Whoop | Activity (10) Sleep (13 full) Vitals (13) HRV, SpO2 | Clinical trials, performance optimization |
When to Use Smartphone vs Wearable Data
✅ Smartphone-First Approach (Recommended for most apps):
- Mental health apps: Behavioral patterns, sleep regularity, activity levels
- Insurance programs: Risk detection via behavioral changes
- Corporate wellness: Activity tracking, sleep duration, engagement
- Habit tracking: Routine detection, circadian rhythm, consistency
- AI coaching: Behavioral data for personalized recommendations
⚠️ Wearable-Required Approach (Only when necessary):
- Clinical trials: Requiring precise HRV, HR variability, SpO2
- Performance training: Elite athletes needing training load metrics
- Medical monitoring: Cardiac rehab, chronic disease management
- Sleep clinics: Detailed sleep stage analysis
⚠️ Warning: Don't Over-Require Precision
Many apps demand wearables for metrics that don't need wearable-level precision. Examples:
- Gym churn prediction: Sleep regularity matters more than exact HRV
- Mental health screening: Behavioral patterns > precise heart rate
- Habit formation: Consistency tracking > detailed sleep stages
By requiring wearables for these use cases, you lose 70% of potential users for negligible accuracy gains.
Platform Support for Smartphone vs Wearable
| Platform | Smartphone Sensors | Wearable Integration | Coverage Strategy |
|---|---|---|---|
| Sahha | ✓ Full support | ✓ HealthKit + Health Connect | Hybrid: 100% coverage + enhanced precision |
| Terra | ✗ None | ✓ 300+ devices | Wearable-only: 30% coverage max |
| Rook | ~ Limited | ✓ HealthKit + Health Connect | Platform APIs: 50-70% coverage |
| Spike | ✗ None | ✓ Medical devices | Clinical: Controlled environment only |
Getting Started: Implementation Checklist
For Smartphone-First Apps:
- ✅ Request platform permissions (HealthKit/Health Connect)
- ✅ Enable Sahha smartphone monitoring (accelerometer, screen time)
- ✅ Define minimum biomarker set (activity, sleep duration, regularity)
- ✅ Optionally support wearables for enhanced data
- ✅ Design UX for 100% user coverage (no wearable required)
For Wearable-Enhanced Apps:
- ✅ Start with smartphone baseline (all users get core features)
- ✅ Add wearable integration for premium users
- ✅ Use unified API (Sahha handles fallback logic)
- ✅ Communicate value of wearable upgrade (HRV, detailed sleep)
- ✅ Never lock core features behind wearable requirement
Next Steps
- 📖 What Are Biomarkers? - Complete biomarker reference
- 🧠 Why Intelligence Layers Matter - Raw data vs intelligence
- 🔧 Platform API vs Direct Integration - Technical comparison
- 🎯 Choosing a Health Data API - Platform selection guide