AI assistants and search decision
FootballGPT
A single technical user can build a useful, narrowed replacement (text-driven session planner + basic SVG animations), but reproducing FootballGPT's full product (advanced video/technique analysis, CoachOS personalised memory, polished multi-advisor UX and platform integrations) is larger and operationally heavier than a small self-build.
Visit website↗Open-source builds that already do this
Every project below is open source and already does this job today. Fork one, self-host it, or take the parts you need - the build prompt further down assumes an empty file, and this is the shortcut past that. Licences differ; check the one on each card before you ship. All FootballGPT alternatives, with the arithmetic →
What a replacement has to do
- Prompt-driven session planner that returns a structured session (warmup, main, progressions, cooldown) and a simple animated diagram for each drill.
What it still won’t have
- High-polish animated practice UX and large library of pre-built animated practices
- Advanced technique analyser and video-to-practice AI (beta features)
- CoachOS personalised methodology upload, session reflections and rich cross-session memory
- The 19 specialist advisors and Football Manager–specific tools (FM tactics, wonderkid shortlists, save clinic)
- App store mobile apps, integrations with MCP clients and packaged API add-on
What remains hard
- Proprietary data
Player Development Framework The 360TFT age-appropriate development framework. Four phases from ages 1 to 21, built from 33,000+ coaching conversations.
- Brand trust
Built by a UEFA licensed coach with 15 years in the game.
First-year cost
Keep paying
Paying is—cheaper in year one.
On cash alone, building overtakes the subscription at 7 seats.
Money you would actually spend
Time you would spend
—
What you would spend
What we assumed
The verdict above measures whether you could build it. This one is only about money.
Runnable build prompt
Build a simplified FootballGPT-style web app using Next.js (React) + TypeScript, Postgres (Supabase or managed Postgres), Stripe for billing, and an LLM provider (OpenAI-compatible). Core features in scope: 1) user auth and one-seat Pro subscription via Stripe; 2) prompt-driven session plan generation (structured JSON output: warmup, main, progressions, cooldown) with server-side prompt templates and retry/error handling; 3) render each drill as an animated SVG diagram (passes and player runs) from the structured JSON; 4) photo upload endpoint using Tesseract OCR to extract text and a small parser to convert detected activities into diagram JSON; 5) basic session library (CRUD) and export to printable session cards. Out of scope: FM-specific tools (tactic builder, wonderkids), advanced video technique analysis, CoachOS cross-session methodology upload, mobile apps, and CLA/MCP integrations. Require tests for API routes, prompt output validation, and error handling for failed OCR/LLM calls.
How we checked
How the score was reached
- Partly verdict base52
- An open-source build was found+5
- 5 cited sources+3
- Price verified on pricing page+3
- 3/3 assessment runs agreed+4
- Hard moats found in the evidence-3
- Evidence score64
The base comes from the verdict. Everything under it is a check that either happened or did not, and each one is a fact frozen in this record rather than a judgement made at render time - so the same evidence always produces the same number.
How scoring works →Cited sources · 5
Every page the run actually retrieved.
- official productFootballGPT — official product
- official pricingPricing — FootballGPT
- official docsFeatures — FootballGPT
- open sourcearc53/DocsGPT
- open sourceaingdesk/AingDesk
Integrity checks
What held up, and what did not.
