Privacy policy

In short

Your documents never leave your device. We have no servers, there are no accounts, no file is ever uploaded and nobody but you can read them. Everything the app does with them — reading them, sorting them, searching inside them, recognising text — happens on the phone.

We collect no data about you. There are, however, two Google components, and we would rather say so up here than bury it in the small print:

• Advertising (AdMob), which uses your device’s advertising identifier. In Europe you are asked first, and you can change your mind at any time from Settings.
• Text recognition for scanned PDFs on Android, which sends Google technical diagnostics about its own operation.

Neither receives your documents, their contents or their names. Both are detailed below.

What data is collected

By us: none

We do not collect, transmit or store outside your device:

• Your documents or their contents
• Your name, email address or any identifying data
• Usage data, statistics or crash reports
• Your location

There is no sign-up form because the app has no accounts, and we have no server we could send anything to.

By Google

Advertising is served by Google AdMob, which uses your device’s advertising identifier, technical data and an approximate location derived from your IP address — not GPS. On Android, the text recognition library also sends technical diagnostics about its own use.

What matters: neither AdMob nor any other library receives your documents, their contents, their names or the categories the app assigns them. The ads you see bear no relation to what you keep here.

What the app does with your PDFs

When you add a document, PDF Inbox makes a copy in its own private folder. The original file is never modified or deleted. On that copy, and only inside your device, it:

• Reads the PDF’s text layer so you can search its contents.
• If the PDF is a scanned image, recognises its text with the device’s own engine: Vision on iPhone and iPad, ML Kit with the model bundled into the app on Android. Neither the image nor the recognised text is sent to any server.
• Sorts the document by comparing its text against a dictionary bundled with the app. No external artificial intelligence service is involved.
• Detects data such as the amount, the date, the tax ID or the IBAN using patterns, to show them on the document’s page.

Third parties

The app integrates no crash reporting, no social networks and no analytics of its own. There are no accounts and we build no profile of you.

Google AdMob — advertising

• Your documents, your images and the recognised text are NOT sent anywhere.
• AdMob uses the advertising identifier, technical device data, an approximate location from your IP and data about how you interact with the ads themselves.
• If you are in the European Economic Area, a consent form appears before anything is shown. You may refuse personalised advertising: you will still see ads, but not based on your profile.
• You can change your mind at any time in Settings → Privacy → “Change my preferences”. Withdrawing consent is as easy as giving it.

Google ML Kit — text recognition, Android only

• Recognition happens entirely on your phone, offline, using the model bundled inside the app.
• That library does send Google technical diagnostics about its own use: which functions ran, how long they took, whether they failed and with what code, along with device data. It is a feature of Google’s library, not ours, and it cannot be switched off without giving up text recognition.
• We tell you because we would rather you knew than found out: we spotted it ourselves while reviewing the permissions declared by our own compiled package.
• This does not happen on iPhone or iPad: there the engine is Vision, it ships with the operating system and sends nothing.

Both are governed by Google’s privacy policy.

Permissions

Android — all files access

Optional and requested in the moment, never at install time. It serves one purpose: finding the PDFs you already have on the phone and copying them into the library. No other file types are read and nothing found leaves the device. If you decline, you can still add documents with the system picker.

Android — internet and network state

The app makes no network requests of its own: it uploads nothing, queries no service and works the same in airplane mode. The only data leaving the device is the advertising and diagnostics described above.

iOS — file access

iOS asks for permission only when you pick a file or a folder with the system picker. Access is limited to what you chose.

Children

The app is not directed at children under 13 and does not knowingly collect data from them. Advertising is configured to show general audience content only.

Your rights

Since we store no data about you, there is nothing we could rectify, export or erase on your behalf: everything is on your device and you control it. Uninstalling the app deletes the entire library. As for the advertising identifier, you can reset or disable it from your operating system’s settings.

Contact

Any questions about this policy: info@pdfinbox.app.