πŸ”¬ Function Health Product Analysis

Deep Dive: Reports, Onboarding, Disclaimers & AI/ML Usage
CSO Intelligence Report | February 2, 2026

🎯 Chairman's Hypothesis: CONFIRMED

Function Health uses clinician-reviewed AI-generated insights, not individual physician consultations. The "clinician notes" are templated responses based on biomarker patterns, with human review for flagging β€” NOT personalized doctor-patient interactions. This is exactly what WellWalla can replicate with Einstein.

πŸ“¦ What Customers Actually Get

1
Lab Tests
100+ biomarkers at Quest Diagnostics
2
Dashboard
Visual results with optimal ranges
3
Clinician Notes
~8 pages of AI-generated insights
4
Action Plan
Lifestyle/supplement recommendations

The Dashboard Experience

The "Clinician Notes" (~8 pages)

Key insight: These arrive 4 weeks after testing. They are NOT real-time physician consultations. The notes are templated based on which biomarkers are out of range, with personalized variable insertion.

πŸ“‹ Onboarding Process

Step Details WellWalla Equivalent
Signup Pay $499, bypass "waitlist" (growth tactic) Simpler pricing tiers ($149-299)
Intake Survey Health goals, reasons for joining Similar + medical history
Schedule Labs Book at Quest Diagnostics (2000+ locations) Labcorp or Quest integration
Blood Draw 2 visits, 10 vials each (~35mL total per visit) Single visit panels possible
Results Roll in over 1-7 days Real-time as available
Clinician Notes 28 days later Einstein: instant AI analysis

βš–οΈ Disclaimers & Legal Structure

What They Claim

What They Disclaim

Critical: Function explicitly states they do NOT replace your doctor. No diagnosis. No treatment. No prescriptions. Results are "informational only."

The Legal Shield

Function positions itself as a health information platform, not a medical practice. This is the same model WellWalla can use. Key elements:

πŸ€– AI/ML in Their Stack (Chairman's Hypothesis)

"I would be surprised if there's actually a doctor reviewing each of those and going through all their data on an individual basis. More likely they had machine learning, and now AI, generating these reports and recommendations."

β€” Chairman Dave Desai

Evidence Supporting AI-Generated Reports

Their Likely Architecture

  1. Rule-based flagging: Biomarkers outside optimal range trigger specific content blocks
  2. ML pattern recognition: Correlate multiple markers (e.g., liver + inflammation = specific guidance)
  3. LLM text generation: Natural language reports personalized with user data
  4. Human QA layer: Clinicians review flagged edge cases, not every report
  5. Critical result escalation: Urgent findings trigger phone calls (claimed)
WellWalla Advantage: Einstein can provide INSTANT AI-powered analysis β€” no 28-day wait. Real-time conversation about results. Ongoing coaching, not one-time report.

πŸ’‘ Strategic Implications for WellWalla

What We Can Replicate

Where We Can Beat Them

What They Have That We Don't (Yet)

πŸ“Š Sample Report Structure (Reconstructed)

Based on user reviews and TIME article, Function's clinician notes follow this pattern:

Section Content
Summary X of Y biomarkers out of range, biological age, key concerns
Heart Health Cholesterol breakdown, risk assessment, action items
Metabolic Blood sugar, insulin, A1C analysis
Hormones Thyroid, sex hormones, cortisol patterns
Nutrients Deficiencies, toxins (mercury, lead)
Inflammation CRP, immune markers
Organ Function Liver, kidney, thyroid specifics
Action Plan Diet changes, supplements, lifestyle mods, "see your doctor" for X