Run Fun and Serious ("Run Fun," "the app") is a gait analysis and running coaching app published by Rooney Industries LLC. This policy explains what data the app collects, where it is stored, who else processes it, and what choices you have.
A plain-English companion describes the same system in friendlier language. If the two disagree, this version wins until corrected — please report disagreements to colin@rooneytech.com.
The following data is created and stored locally on your iPhone. It is not transmitted to our servers unless you explicitly choose to share or export it:
When you use cloud-backed features, the following data leaves your device:
runfun-video-uploads) for staging, then to a Cloud Run service running on Google Cloud Platform (us-central1) for pose estimation and overlay rendering. RunPod GPU containers are retained as a fallback if Cloud Run is unavailable. The job is associated with a randomly generated job ID and your profile UUID.imu/) and passed to the same GPU container (GCP Cloud Run or RunPod) as an additional input. This is automatic when those sensor recordings exist for the clip you're submitting; to avoid sensor uploads, decline cloud analysis, turn off "Send motion data with cloud analysis" in Settings → Privacy, or run the app without the Watch / AirPods paired. This motion data is kept in durable storage (see Data retention below).iPhone17,1) and app version, sent to our Cloudflare Worker for capacity planning, crash diagnosis, and performance debugging. No video, audio, or PII content. Apple-side crash reports captured by TestFlight / App Store Connect are governed by Apple's privacy policy (see the third-party-processors table below).When you use the voice coaches (Kai, James, Minna, Simon, Krishna):
ElevenLabs's privacy policy is at elevenlabs.io/privacy-policy.
We have opted OUT of ElevenLabs using your conversation audio and transcripts to train their AI models (the account-level "Improve the models for everyone" setting is turned off). If we ever change this, we will update this policy and tell you in advance.
We do not currently sell your conversation data, and per our reading, neither does ElevenLabs. If our position ever changes, we will tell you in advance.
If you grant HealthKit access, the app reads (union of phone + Watch requests):
The app may write back: workout records (which include the GPS route the Watch recorded), walking/running distance, active energy, heart rate, swim distance + stroke count.
HealthKit data is read on your device. Data the app reads is included as text context in coaching sessions sent to ElevenLabs (see "Sent to ElevenLabs" above). The raw HealthKit values are not separately uploaded to our servers as a structured dataset. Per Apple's App Store rules, HealthKit data is never used for advertising or sold to data brokers.
| Processor | Purpose | Data handled | Their policy |
|---|---|---|---|
| Amazon Web Services (S3) | Video staging | Video clips with random job IDs | aws.amazon.com/privacy |
| Google Cloud Platform (Cloud Run) | GPU video processing (primary as of 2026-05-28) | Video clips with job IDs and profile UUIDs | cloud.google.com/terms/cloud-privacy-notice |
| RunPod | GPU video processing (fallback) | Video clips with job IDs and profile UUIDs | runpod.io/legal/privacy-policy |
| Cloudflare (Workers, D1, KV) | Application backend | Conversation logs, telemetry, OAuth tokens, FIT data | cloudflare.com/privacypolicy |
| ElevenLabs (third-party AI) | Voice AI coaching | Audio + transcripts during sessions, plus the run/gait metrics, motion data, profile, recovery/sleep summaries, and prior conversation context sent as part of the session. We have opted out of ElevenLabs using this audio/transcripts to train their AI models. | elevenlabs.io/privacy-policy |
| Sentry (Functional Software, Inc.) | Crash + error diagnostics | Crash reports and diagnostic logs (device model, OS version, app version, error stack traces). No names, emails, or phone numbers. | sentry.io/privacy |
| Apple App Store / TestFlight | Distribution | App downloads, optional crash reports | apple.com/legal/privacy |
Equal protection of your data. We require each third party listed above — Amazon Web Services, Google Cloud Platform, RunPod, Cloudflare, ElevenLabs, and Sentry — to provide protection of your data that is equal to or greater than the protections described in this policy, under their terms of service and data-processing commitments.
| Surface | Retention |
|---|---|
| On-device data (profiles, IMU, video, conversation history) | Persists until you delete the profile or the app |
AWS S3: source video clips (uploads/) | 1 day |
AWS S3: motion/sensor (IMU) data (imu/, imu-durable/) | Durable — kept so the analysis and coach can use it; you can turn this off in Settings → Privacy ("Send motion data with cloud analysis") |
AWS S3: opt-in research contribution — video (contrib/) + de-identified gait/pose data | Persists per your contribution consent (default OFF; revocable in Settings → Privacy) |
AWS S3: rendered overlays + freeze frames (styled/, freeze-frames/) | 7 days |
Cloudflare KV: parsed FIT data (fit:*) | 1 year |
Cloudflare KV: raw FIT bytes (fit-raw:*) | 30 days |
Cloudflare KV: per-user training synopsis (fit-synopsis:*) | Persists by design (your training history; user-facing state) |
Cloudflare KV: auth records (auth:*) | Persists by design (your account) |
| Cloudflare KV: Strava OAuth tokens | Persists until you disconnect Strava |
| Cloudflare D1: conversation logs | Persists by design (multi-session context) |
| ElevenLabs servers | Per ElevenLabs's policy and our agent-level configuration |
Settings → Privacy → "Delete Account & Data" removes your profile's cloud records in one step — conversation logs (D1), FIT synopses, routines, rendered videos, your durable profile-keyed motion data (imu-durable/), and any opt-in video contribution (contrib/) — disconnects Strava (revoking our access at Strava), AND wipes all on-device data, returning the app to its first-launch setup. (Source video staging auto-deletes within 24 hours regardless.) De-identified gait/pose data you have contributed for research is retained de-identified (keyed by a per-device salted hash, not your name or email) per your contribution consent.
Run Fun can improve its gait models with contributed data. You can opt in — via a dedicated, declinable consent screen in Settings → Privacy — to contribute de-identified gait/pose and sensor data. It is off by default, you control it from Settings → Privacy, and you can turn it off anytime. Contributed data is de-identified (keyed by a per-device salted hash, not your name or email). To remove research data you have already contributed, email colin@rooneytech.com.
Run Fun is not directed at children under 13. We do not knowingly collect data from children under 13.
This policy continues to apply to data collected before any change in ownership or corporate structure. If Run Fun is sold or restructured, we will notify you in advance through the app and give you the opportunity to delete your data before the change takes effect.
We may update this policy as the app evolves. The "Last Updated" date will reflect the most recent revision. Material changes will be communicated in-app at least 14 days before they take effect.
Colin Rooney
Rooney Industries LLC
Denver, CO
colin@rooneytech.com