∴ Therefore

Everyone tracks their health. Nobody experiments properly. Turn your health goals into scientific experiments through structured, AI-supported, N-of-1 protocols.

Header

Create your first protocol today: Therefore | From Health Tracking to Personal Science (therefore.bio)

Overview

My WHOOP shows my HRV dropped to 30 overnight. Google says this was because of X, possibly Y, and I should try Z. Therefore says: based on 4 studies, 3g glycine before bed increases HRV by 15%.

A scientific experiment is defined as a process used to derive answers to questions from observations. This is through formulating a hypothesis, designing and conducting an experiment to test it, and drawing conclusions based on the results.

Therefore bridges the gap between health tracking and evidence-based action. It helps you design and run structured N-of-1 experiments on yourself, powered by peer-reviewed research from PubMed, scientific journals, and academic databases.

Health trackers tell you what happened. ∴ Therefore tells you what to try next.

Features

Feature	Description
Protocol Builder	Utilize a 5-step guided flow to create a structured experiment
Intervention Suggestions	Choose from 6 AI-suggested interventions tailored to your goal and timeline
Smart Metric Recommendations	AI analyzes studies to suggest which metrics to track, with confidence levels based on scientific evidence for your specific intervention
Projected Outcomes	View evidence-based predictions for how your metrics will change
Protocol Research	Get a detailed analysis of intervention effectiveness, mechanisms of action, and optimal protocols with direct citations to summarized study details, including key findings

How It Works

AI Integration

i.e. Usage of Perplexity Sonar API

/src/lib/ai-config.ts

∴ Therefore makes use of engineered AI prompts that act like a research scientist:

Multi-Stage Research Pipeline

Goal Refinement: Perplexity structures vague health goals into testable hypotheses - not just "more energy" but "maintain consistent energy levels without significant daily dips"
Intervention Matching: Perplexity analyzes how each intervention works in the body to match it with your specific goal (e.g., This stack supports muscle relaxation and reduces stress, making it easier to wake up feeling rested and energized.)
Metric Selection: Perplexity evaluates which biomarkers actually demonstrate intervention efficacy based on studies
Outcome Projections: Synthesizes effect sizes across studies to predict personalized outcomes with confidence intervals
Protocol Optimization: Extracts optimal amount, timing, and duration across multiple study methodologies

Prompt Architecture

Cascading prompts that build context (goal → intervention → measurements → projections)
Enforced JSON schemas ensure structured, parseable research output
Domain-restricted search (PubMed, ScienceDirect, Nature, etc.) prevents hallucination

Real-Time Research Synthesis

Analyzes papers in seconds to find scientific evidence for a custom specific goal
Not a static database - discovers latest research as it's published
Cross-references multiple studies to gauge confidence in identifying consistent effects

This positions the use of Perplexity as an enabler of personalized experimentation at scale.

Usage

Production

You can get started immediately with creating your first protocol at Therefore | From Health Tracking to Personal Science.

This project is built using:

Category	Technologies
Application	Next.js, Tailwind, Radix (Shadcn), Motion, Lucide
AI Integration	Perplexity Sonar
Storage	localStorage
Infrastructure	Vercel, Cloudflare

Why This Matters

Unmet needs

Health trackers today are built to tell you what has already happened, but t has come to a point where this data needs to be actionable with the power of these wearables. Market research shows that the current size of health wearables are around $53bn, and growing at a rate of over 20% year over year. In 10 years time by 2034, this would be more than $400bn ^1.

Consumer willingness

The premium health tracking ecosystem has already validated the consumer willingness to invest in the "quantified self" for health optimizations. Monthly subscriptions of around $25-30 have proved to be feasible business models. Therefore is able to capture value at this intersection as the intelligence layer above these devices, not competing with hardware but making their data actionable.

Our users already own or have invested $350-500+ in tracking devices, possibly spending much more annually on supplements. They are already primed for a platform that creates a natural premium positioning, maximizing their hardware's ROI by connecting measurement to evidence-based intervention through scientific protocol design.

Timing of opportunity

Generic health advice in a post-pandemic world does not carry the same credibility as people realize catch-all solutions don't meet individual biological needs
Therefore democratizes the gold standard of clinical trials through self-experimentation
Personalized medicine, while emerging, continues to be behind expensive gentic testing and research institutions ^2
LLMs like Perplexity Sonar enables live research synthesis at scale

Roadmap

Direct wearable integrations with WHOOP, Oura, etc.
Social proof and public protocol sharing
Protocol templates library for common goals
Increased individual control and refinement of protocols
Automated outcome and real-time adjustment recommendations based on biomarker changes: "Your overnight HRV suggests delaying today's intervention."
Aggreggated protocol effectiveness data to power matching: "Similar users saw XY% success with this intervention."

Get In Touch

Get Started: therefore.bio
Portfolio: zsy.sh

Disclaimer: ∴ Therefore is not a substitute for professional medical advice, diagnosis, or treatment. The information provided through our AI research synthesis are not medical advice, prescriptions, or treatment recommendations. Always consult with qualified healthcare providers before starting any new health protocol, especially if you have existing medical conditions or are taking medications. Results from personal experiments may vary, and what works for one individual may not work for another.