BloopBack privacy policy
Effective 17 November 2026
This policy covers BloopBack, the iPhone app. Signalfern is operated by Gilad Bonjack.
Your audio stays on your phone
Recording, reversing, effects and playback happen on your device. We do not upload your recordings, waveforms, speech, transcripts, filenames or saved-sound identifiers to Signalfern or PostHog. The app does not require an account.
Accepted sounds are saved in the app's local storage. Temporary processing and export files, including reveal videos, are generated and stored locally. Reveal videos do not use a camera. Your device's normal backup settings may include saved sounds. Deleting a sound removes the app's local copy; it does not recall copies in backups or copies you have shared.
Mystery Games in TestFlight
Mystery Games is being tested in our internal TestFlight build. Team names, scores and your imitation recordings stay in its separate local game storage, which is excluded from device and cloud backups. Removing that local game can remove the exact free game available for replay. We do not upload this game content.
To verify an eligible app installation and keep the first-game offer separate from beta testing, the app sends Apple App Attest verification material and a signed Apple app transaction for verification by our Signalfern service. The Apple transaction and DeviceCheck token are processed for the current request, not retained or sent to PostHog. The on-device Apple device-verification identifier is not sent to our service.
The service retains an attestation key identifier, public verification material and the Apple attestation receipt, an assertion counter, and the operational free-game request and signed grant. The request includes a random retry token, audio language, number of teams and catalog revision; the grant selects a game deck without revealing your recordings, names or guesses. These records are used for app functionality and offer-abuse prevention, separately from analytics. They are retained while needed to operate and recognize the offer; there is currently no automatic expiry for accepted key and claim records. Temporary verification challenges expire after two minutes. Abuse-prevention rate buckets use a hash of the action and attestation key identifier, alongside shared request-limit buckets. They contain no IP address or raw key identifier, and are removed after one hour when the service next handles a request. You can contact us about these records using the privacy contact below.
Beta and public-distribution claims are separated. The current TestFlight service does not consume or change public first-game eligibility. Apple and our hosting provider receive network connection information when processing requests. These integrity checks do not add advertising tracking or an account.
The optional unlimited-games unlock uses Apple's StoreKit purchase and restore interfaces. Apple handles purchase and payment processing. We do not receive your payment-card details or Apple account identity, and do not send purchase receipts or transaction identifiers to PostHog. The app checks the current Apple entitlement on your device. TestFlight purchase testing uses Apple's sandbox, separately from any public purchase.
Feedback you choose to send
When you send feedback, we receive your written message, app version and build, and an email address only if you choose to provide one. No recordings are attached. We store feedback privately to handle your request and improve BloopBack, and send an owner notification using Cloudflare Email Service. You do not need an account or a mail app. The in-app receipt confirms storage, not that the message has been read or email delivered.
Feedback is available in our service for up to 90 days; expired records are removed when feedback processing or the private recovery view next runs. Email copies may remain in our support mailbox until deleted as part of handling the conversation. You can request deletion at hello@signalfern.com. A random submission token prevents duplicate sends; it is not an installation identifier and is not sent to analytics.
Selected usage analytics
We use PostHog's EU-hosted service to understand app use and improve the recording, reverse-speech and sharing experience. The app sends selected events when you open the app, enter a play mode, start or complete a round, pass the phone, swap roles and record the next turn, record, play a sound, select an effect, reopen a saved sound, prepare an export, request app sharing, finish the system share sheet, open or submit feedback (success or a bounded failure category), or request Apple’s native rating prompt. A rating request does not tell us whether Apple displayed the prompt or a rating was submitted.
Events contain the play mode, bounded round action or failure type (including Mystery game starts, effective hint or imitation use, judgments, skips, resume, completion and replay), relevant take or playback type, effect choice, broad recording-duration range (under 5 seconds, 5–15 seconds or 15–30 seconds), export or share outcome and whether the original answer is included in a reveal video, app version/build, a broad distribution and testing classification (development, simulator, beta/sandbox, App Store or unknown), and bounded technical event/SDK version information, including a coarse reason when distribution verification succeeds, fails or times out. We do not send raw errors or App Store account information. The classification helps keep test activity separate from public-distribution usage; we do not send App Store receipts or transaction identifiers. They include randomly generated installation and session identifiers so we can measure repeat use and completion. These identifiers link events from an installation; the data is pseudonymous, rather than fully anonymous.
Events also include limited technical context: an iPhone model code identifying the device type (or simulator/unknown), the iOS major and minor version, a broad screen-size category (compact, standard or large) and the selected text-size category. We use this information to understand whether recording, playback and other steps work differently across devices and layouts. We do not send your device name, serial number or exact screen dimensions. The text-size category is a display setting; we do not use it to infer health or disability.
We do not attach your name, email, contacts, advertising identifier, precise location or audio contents to these events. Automatic interaction capture and session replay are disabled. We do not record your screen or use this analytics data for advertising, cross-app tracking or sale.
Location enrichment from IP addresses is disabled. PostHog and its infrastructure providers still receive network connection information, including an IP address, when accepting a request. EU hosting does not mean that no service provider can ever process data elsewhere. See PostHog's privacy policy for its service and provider practices.
Retention and choices
Analytics events are retained according to our PostHog plan, currently one year. Removing the app stops new app events but does not automatically erase events already received. Saved audio remains under your device and backup controls.
You choose whether to grant microphone access in iPhone Settings. When you use Share, iOS passes the exported audio or a video containing your audio and waveform to the destination you select; that service's own privacy practices apply. A completed share-sheet action does not prove that another person received or played it.
Contact and updates
For privacy questions or requests, contact hello@signalfern.com. Because there is no app account and we do not attach a name or email to analytics, we may need further information to locate data associated with your installation. We will update this page when the app's data practices change.
This informational page adds no analytics scripts or cookies. The website's hosting provider may process request information and provide security features.