Passive calorie labels failed. Active, personal steering is the opening.
Americans eat a record share of meals out and badly misjudge what's in them. Every nutrition app still makes you log after. Palatify recommends the right dish at a nearby restaurant before you order — and gets restaurants to pay to be the recommendation.
Why now
58.5%
of U.S. food spending is now away from home — a record
USDA ERS, 2023
2 in 3
diners underestimate restaurant-meal calories; ¼ by 500+
USDA/NHANES, 2024
~24 cal
all that mandatory menu labels change per order
Peer-reviewed, 2024
70–80%
of calorie-app users quit within two weeks
JMIR review, 2026
The category's unsolved failure is logging friction and bad data — worst exactly when eating out. No commercial database even has nutrition for independent restaurants, because the FDA rule only forces 20+-location chains to disclose. That structural gap is our moat.
Market
$11.1B/year
186M U.S. adults with elevated BMI (72% of adults), nearly all of whom eat out, × ~$60 blended annual ARPU. Triangulates with the $5.5–14B calorie-tracking market growing 12–20%/yr.
CDC 2024 · USDA ERS 2023
$3.3B/year
~55.7M goal-oriented 'active dieters' who eat out often (~30% of elevated-BMI adults) × $60/yr. Consistent with MyFitnessPal's ~30M MAU as a category engagement ceiling.
Bottom-up
$40–90MARR, 5-yr
Capturing 0.5–2.0% of SAM = 0.28–1.1M paying users ($17–67M consumer ARR; 1% base ≈ $33M), plus restaurant retail-media and B2B2C covered-nutrition legs.
Modeled
Cal AI is the proof point: ~15M downloads and $30M+ ARR in under two years, bootstrapped, before MyFitnessPal acquired it (Dec 2025).
Model
1. Consumer subscription
Transparent pricing with a genuinely useful free tier (exploiting MyFitnessPal/Noom trust erosion). ~$60/yr blended, annual-plan-led to fight churn. Gate growth on LTV:CAC ≥3:1, 6–9 mo payback.
2. Restaurant retail-media
Clearly-labeled 'featured healthy dish' placements at ~$200/mo/location. 2,000 locations ≈ $4.8M; 10,000 ≈ $24M. Modeled on DoorDash's $1B+ ad run-rate. High-margin, and it solves the supply side.
3. B2B2C covered nutrition
Payer/employer contracts on broadly-covered medical nutrition therapy (CPT 97802–97804) for day-one billable revenue and 'free-to-you' acquisition. The durable, defensible third leg.
Why we win
No incumbent occupies all four of pre-meal · location-aware · independent-restaurant coverage · restaurant-supply monetization.
| Player | Strong at | Structurally absent |
|---|---|---|
| MyFitnessPal | Scale (220M registered) | After-the-fact logging; inaccurate DB; 70–80% 2-wk churn |
| Cal AI / SnapCalorie | Photo estimation, TikTok reach | No recommendation, no restaurant supply side |
| Sweetgreen / CAVA | Great healthy menus | Single-brand only — can't be brand-agnostic |
| DoorDash / Uber Eats | Restaurant graph + $1B ad engine | Zero nutrition-goal intelligence |
| Foodsmart | Payer contracts, RD-led, 2.2M members | Grocery/telehealth — not in-the-moment eating-out |
Plan
Density, not breadth
Launch 2–3 dense urban markets (Boston beachhead). Seed full chain coverage instantly (Nutritionix/FatSecret/USDA); build independent-restaurant coverage block-by-block. Acquire via TikTok/influencer targeting GLP-1 and goal-oriented audiences. Fund with a pre-seed SAFE + a non-dilutive SBIR for the estimation engine.
Turn coverage into retail-media
Convert restaurant coverage into the featured-dish revenue leg; sign marquee logos. Raise a seed/Series A timed to AI + food-as-medicine appetite (AI took 62% of 2025 digital-health VC dollars).
Land the payer leg
Sign payer/employer B2B2C contracts on covered MNT, layering reimbursement as the durable moat. Realistic exit: a tech-forward chain (CAVA/Chipotle), a delivery platform, or a nutrition-care consolidator.
Risks we underwrite honestly
- Accuracy is physics-bound (~16–25% floor). We market "fast, good-enough, easy to correct," never clinical — RAG-grounded against USDA data.
- Retention is the category killer. Plan-ahead makes a recurring habit out of every meal, not a logging chore; annual plans halve churn.
- Two-sided cold start. Solved with geographic density, not national breadth.
- GLP-1 headwind → tailwind. Position as the eating-out companion for GLP-1 users.
The ask
Raising a ~$1.0–1.5M pre-seed SAFE (~$5–6M post) to launch the Boston beachhead, build the independent-restaurant nutrition layer, and prove consumer retention + the first featured-restaurant cohort — matched with a non-dilutive SBIR for the estimation-engine R&D.
18-month milestones: density in 1 metro, >35% D30 retention, 150+ featured restaurants, LTV:CAC ≥3:1 — the proof points for a seed.
Try the productAll figures are research-backed (CDC, USDA ERS, AHRQ, peer-reviewed sources, company filings). Detailed business plan & financial model available on request.