This was an in-depth interview with Zach Yadgari, a 17-year-old entrepreneur making over $1 million/month with his app Calorie AI (Cal AI), which uses AI to estimate food calories from photos.
Key topics included Zach’s entrepreneurial journey starting from age 7, building and selling a gaming website, how Cal AI was ideated, and the influencer-driven marketing strategy used to scale rapidly.
The discussion covered company structure, marketing tactics, product design, and Zach’s views on scaling, product virality, and future ambitions.
Zach shared actionable insights on influencer dealmaking, launching app studios, and creating tools for influencer campaign management.
Action Items
(None explicitly mentioned with deadlines, but notable next steps inferred from context)
Early December – Zach: Launch the influencer marketing platform Viral.tech publicly.
Zach: Continue development and scaling of new health and fitness apps for the app studio.
Zach: Assess and adjust Cal AI’s marketing and product features based on ongoing testing (e.g., Thanksgiving campaign, onboarding AB tests).
Zach: Begin influencer-driven Thanksgiving campaign (app free for the day), track results and user engagement.
Zach: Continue refining and publicly sharing internal influencer management tools and strategies.
Zach: Start payout to founders as Cal AI’s App Store cash flow cycles allow (expected in the next 1–2 months).
Zach’s Entrepreneurial Journey and Background
Began programming at age 7, inspired by a desire to create video games.
Built and monetized an unblocked gaming website at age 13, reaching 5 million users and $660k/year revenue; sold it for $100,000 at age 16.
Transitioned into mobile apps after seeing other young founders’ trajectories stagnate.
Co-founded Cal AI with three others (two teens, two adults).
Development and Product Ideation for Cal AI
Identified outdated, complex calorie tracking apps as a market gap—Cal AI was conceived as a simple, viral alternative.
Used ChatGPT technology to enable instant food calorie scanning via photo.
Focused app only on tracking calories, carbs, protein, and fat to keep it simple and user-friendly.
Decision to be a premium-only app with a $30/year or $10/month subscription and three-day free trial; all users go through an onboarding sequence tied to higher conversion rates.
Influencer Marketing Strategy and Tactics
Early growth driven by Zach directly DMing thousands of fitness influencers and iteratively refining outreach.
Uses a CPM (cost per thousand views) and RPM (revenue per thousand views) model to negotiate and ensure profitable influencer partnerships.
Offers bundled deals (e.g., paying upfront for four videos/month) to secure recurring influencer content at better rates.
Manages and tracks influencer campaigns using internal tools (soon to be released as Viral.tech), including attribution by correlating app downloads and video views.
Employs tactics like pinning comments, community engagement in the comment sections, and planting comments to improve discoverability and conversion.
Scaling, Team, and Future Vision
Cal AI has 12 team members (4 co-founders, 8 staff); company is entirely bootstrapped and runs at ~50% profit margin.
Core team includes high schoolers and older co-founders.
Plans to beat MyFitnessPal by scaling revenue and market share; sees potential for acquisition or significant exit.
Building an “app studio” leveraging their influencer network to quickly launch other inherently viral health/fitness apps.
Launching a public SaaS platform for influencer campaign management (Viral.tech).
Scaling decisions informed by investment in team, design, marketing, and new channels (e.g., paid ads, TV ads).
Experiments with virality mechanics (e.g., Thanksgiving free day, encouraging story shares, viral content prompts).
Insights on Product Development and User Psychology
Key lessons: simple onboarding with intentional, even “fake” questions increases conversion.
Education and user support are critical as many expected AI accuracy beyond current capabilities (e.g., “x-ray” food scanning).
Paid Apple Store search ads are used as a benchmark for product-market fit (profitable ads indicate a superior product).
User retention and monetization are tracked closely; app is cash flow positive and reinvests earnings continuously.
Reflections on Personal Growth and Philosophy
Zach values social experience in high school despite business success; is undecided about attending college, planning a gap year and moving cities.
Views entrepreneurial journey as a stepping stone toward larger, impact-driven goals (e.g., “iPhone for your brain,” neural-link style innovations).
Emphasizes that anyone, regardless of age, can start and succeed in tech entrepreneurship, especially with new AI tools.
Decisions
Cal AI will remain a premium-only app — Supported by higher conversion rates during onboarding and clearer monetization results.
Bundling influencer deals and recurring monthly retainers as primary marketing strategy — Proven most cost-effective for acquisition and long-term engagement.
App studio focus will remain on health/fitness apps with inherent viral potential — Maximizes the existing influencer network and proven go-to-market strategies.
Internal influencer marketing platform (Viral.tech) to be launched publicly — To capitalize on widespread industry need and first-mover advantage.
Reinvestment of all profits during growth phase — To sustain scaling and marketing momentum, with plans for eventual payouts as cash flow stabilizes.
Open Questions / Follow-Ups
Will the Thanksgiving free campaign meaningfully affect user acquisition or retention? (to be evaluated post-campaign)
What are the long-term LTV and retention metrics for Cal AI, given its youth (launched 6 months ago)?
How effective will scaling into additional paid channels (TV, Google, etc.) be as compared to current influencer-first strategy?
How will the new public influencer marketing tool (Viral.tech) be adopted by other app marketers?
Will launching multiple apps in parallel dilute focus or complement Cal AI’s growth as planned?
What product improvements or new features may be necessary as user expectations around AI accuracy and feature set evolve?