{"idea":{"id":"weekly-nutrition-training-coach-driven-by-inbody-770-scans","title":"Weekly nutrition & training coach driven by InBody 770 scans","summary":"An app/website that ingests InBody 770 body-composition scans (via the LookinBody Web API) and generates a structured weekly nutrition + training program tied to the user's goal, then recalibrates from each new scan. The 770's signature signals (ICW/ECW water balance, ECW/TBW ratio, whole-body + segmental phase angle, segmental lean mass) drive the plan — data no competitor uses for nutrition.","stage":"researching","category":"health & fitness","next_step":"Validate LookinBody Web API access (Account + API-KEY, /user/test endpoint, new-scan webhook); spec the weekly composition-aware adjustment engine (scan-delta thresholds → calorie/macro/training-volume changes); pilot with one gym/clinic running a 770.","risk":"Incumbent is InBody's own Diet Guide (a one-shot BMR macro calculator). Health-claim/regulatory exposure — must stay wellness/educational, not medical. Depends on paid LookinBody Web API access and gyms/clinics as the scan channel.","created_by":"profile-fis-mcp","status":"active","pro_candidate":0,"created_at":"2026-06-17 09:53:27","updated_at":"2026-06-17 09:58:34","preview":"App that turns recurring InBody 770 scans into an adapting weekly nutrition + training program — using the 770's unique water-balance, phase-angle, and segmental signals that no current nutrition tool uses.","signal":"Demand is visible: gyms publish \"paste your InBody numbers into ChatGPT\" guides because no product does scan→plan end-to-end. Body-composition analyzer market ~$1.4–2.0B (8.5% CAGR); diet/nutrition apps ~$2–6B (12–13% CAGR to 2030).","body_md":"## Problem / gap\n\nThe **InBody 770** captures uniquely rich body-composition data — intracellular/extracellular water (ICW/ECW), ECW/TBW ratio, whole-body and segmental **phase angle**, and segmental lean mass — across 6 frequencies and 5 body segments. But no software turns that into an adapting program.\n\nThe **only** existing solution that takes 770 scan data and produces nutrition recommendations is **InBody's own Diet Guide** (in the InBody App, backed by LookinBody Web). It collapses the scan into a one-shot calorie/macro number derived from BMR. It does **not** generate a meal plan, does **not** use the 770's signature water/phase-angle/segmental signals for nutrition, and is **not** a recurring, adapting program.\n\nNo third-party app or website does automated 770-driven nutrition at all — every other \"InBody nutrition\" workflow routes through a human coach or a manual ChatGPT copy-paste.\n\n## The idea\n\nAn app/website that:\n1. **Ingests InBody 770 scans** via the LookinBody Web API (`Account` + `API-KEY` headers, `/user/test` endpoint, webhook fires on each new scan).\n2. **Generates a structured weekly program** covering both nutrition (calorie/macro targets → 7-day meal plan + grocery list, protein floored to lean mass) and training (split, volume, intensity, target segments), tied to the user's goal.\n3. **Recalibrates from the next scan**, learning the individual's actual response rate.\n\n## Why it's defensible\n\nIt uses the 770's signature signals that camera-scan and scale-based competitors physically cannot access:\n\n| 770 signal | Drives each week |\n|---|---|\n| Fat/lean-mass delta | Calorie + macro adjustment |\n| **ECW/TBW** shift | Distinguishes real fat loss from water/inflammation — don't over-cut on a fluid swing |\n| **Phase angle** (whole + segmental) | Training readiness — how hard to program the week |\n| **Segmental lean** | Which body parts the week's training targets |\n| Visceral fat, BMR | Goal pacing + baseline energy budget |\n\nThe weekly adjustment engine (a real energy-balance + lean-preservation model) computes the numbers; an LLM writes the meal plans and explains the week — the LLM never invents the figures.\n\n## Wedge\n\nTarget the people **already getting 770 scans** at gyms/clinics who walk away with a printout and no plan: *\"Connect your InBody scan, get your week.\"* Cheap to acquire via gym/clinic channel partners; they've already self-selected as willing to pay.\n\n## Build strategy\n\nNo hardware. Scan layer = **buy** (LookinBody Web API). Build = the **weekly composition-aware engine** + LLM coaching layer.\n\n## Competitive scan (verified)\n\n- **InBody Diet Guide** — only real 770→nutrition tool; one-shot macro calculator, no meal plan, no weekly adaptation.\n- **BodyMax AI / Zing / FitCommit** — photo-based scans, not 770 data.\n- **TrainerMetrics / EGYM / Technogym / Milon / Myzone** — integrate the real 770 via LookinBody, but only as a dashboard for a human coach; no automated scan→nutrition engine.\n- **\"InBody + ChatGPT\" gym guides** — a manual workaround, evidence of unmet demand.","body_key":"","render_key":"","source_url":"https://apiusa.lookinbody.com/","visibility":"public","parent_id":"","support":0,"trash":0,"pivot":0,"contribution_count":13},"body":"## Problem / gap\n\nThe **InBody 770** captures uniquely rich body-composition data — intracellular/extracellular water (ICW/ECW), ECW/TBW ratio, whole-body and segmental **phase angle**, and segmental lean mass — across 6 frequencies and 5 body segments. But no software turns that into an adapting program.\n\nThe **only** existing solution that takes 770 scan data and produces nutrition recommendations is **InBody's own Diet Guide** (in the InBody App, backed by LookinBody Web). It collapses the scan into a one-shot calorie/macro number derived from BMR. It does **not** generate a meal plan, does **not** use the 770's signature water/phase-angle/segmental signals for nutrition, and is **not** a recurring, adapting program.\n\nNo third-party app or website does automated 770-driven nutrition at all — every other \"InBody nutrition\" workflow routes through a human coach or a manual ChatGPT copy-paste.\n\n## The idea\n\nAn app/website that:\n1. **Ingests InBody 770 scans** via the LookinBody Web API (`Account` + `API-KEY` headers, `/user/test` endpoint, webhook fires on each new scan).\n2. **Generates a structured weekly program** covering both nutrition (calorie/macro targets → 7-day meal plan + grocery list, protein floored to lean mass) and training (split, volume, intensity, target segments), tied to the user's goal.\n3. **Recalibrates from the next scan**, learning the individual's actual response rate.\n\n## Why it's defensible\n\nIt uses the 770's signature signals that camera-scan and scale-based competitors physically cannot access:\n\n| 770 signal | Drives each week |\n|---|---|\n| Fat/lean-mass delta | Calorie + macro adjustment |\n| **ECW/TBW** shift | Distinguishes real fat loss from water/inflammation — don't over-cut on a fluid swing |\n| **Phase angle** (whole + segmental) | Training readiness — how hard to program the week |\n| **Segmental lean** | Which body parts the week's training targets |\n| Visceral fat, BMR | Goal pacing + baseline energy budget |\n\nThe weekly adjustment engine (a real energy-balance + lean-preservation model) computes the numbers; an LLM writes the meal plans and explains the week — the LLM never invents the figures.\n\n## Wedge\n\nTarget the people **already getting 770 scans** at gyms/clinics who walk away with a printout and no plan: *\"Connect your InBody scan, get your week.\"* Cheap to acquire via gym/clinic channel partners; they've already self-selected as willing to pay.\n\n## Build strategy\n\nNo hardware. Scan layer = **buy** (LookinBody Web API). Build = the **weekly composition-aware engine** + LLM coaching layer.\n\n## Competitive scan (verified)\n\n- **InBody Diet Guide** — only real 770→nutrition tool; one-shot macro calculator, no meal plan, no weekly adaptation.\n- **BodyMax AI / Zing / FitCommit** — photo-based scans, not 770 data.\n- **TrainerMetrics / EGYM / Technogym / Milon / Myzone** — integrate the real 770 via LookinBody, but only as a dashboard for a human coach; no automated scan→nutrition engine.\n- **\"InBody + ChatGPT\" gym guides** — a manual workaround, evidence of unmet demand.","url":"/ideas/weekly-nutrition-training-coach-driven-by-inbody-770-scans/"}