Health, home and travel decision
Run Plan
A competent developer can reimplement a usable subset (plan generation, Strava ingestion, UI) in a few weeks, but matching the paid product's continuous model tuning, polish, and reliable Garmin sync at scale is non-trivial; prior open-source coaching projects exist but don't provide full parity.
Visit website↗$6.58/mo
$79/yr
Read off the official pricing page.
$100one-off64 h to build
$50/mo6 h/mo upkeep
On cash alone, building overtakes the subscription at 9 seats.
No open-source build does this yet
Nothing published replaces this one, so a replacement starts from an empty file. Here is what it would have to cover.
What a replacement has to do
- Ingest running history from Strava/Garmin, generate and adapt a multi-goal training plan via an LLM + domain logic, present plan/calendar and analytics in a web UI, and (optionally) push daily workouts back to Garmin.
What it still won’t have
- Polish, UX refinements, and continuous product improvements from a dedicated team
- Customer support, subscription management, and refunds handling
- Potential performance and edge-case handling for device sync at scale
- Ongoing model tuning and dataset aggregation used to improve plan quality over time
What remains hard
- Product polish and ongoing maintenance
First-year cost
Keep paying
Paying is—cheaper in year one.
On cash alone, building overtakes the subscription at 9 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 minimal AI-powered running coach web app using Node.js (Express), Postgres, Next.js React frontend, and OpenAI-compatible LLM API. Core features in scope: Strava and Garmin OAuth + activity ingestion; user onboarding to set goals and race schedule; an LLM-orchestration service that generates a season-long plan from user history and goals and exposes endpoints to adapt the plan on events (injury, missed workouts); a calendar UI showing per-day workouts and basic analytics (weekly volume, pace trends); a scheduled worker to push workouts to Garmin Connect; authentication, subscription stub (no payment processor integration required for MVP). Out of scope: payments processing, mobile native apps, large-scale multi-tenant observability, and training a proprietary model. Include error handling for API failures, retries for sync jobs, unit tests for plan generation logic, and integration tests for the Strava/Garmin flows.
How we checked
How the score was reached
- Partly verdict base52
- 2 cited sources+1
- Price verified on pricing page+3
- Evidence score56
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 · 2
Every page the run actually retrieved.
- official productRun Plan — product page
- official pricingRun Plan — pricing
Integrity checks
What held up, and what did not.
