VPN App Privacy Policy: What to Declare in Each Store
A VPN app privacy policy has to match three other things: the App Privacy details on Apple’s product page, the Data safety form on Google Play, and what your servers and SDKs actually collect. VPN apps also carry stricter rules than most. Apple’s Guideline 5.4 bars a VPN app from selling, using or disclosing data to third parties for any purpose, and requires the privacy policy to say so. Google’s VpnService policy requires a prominent in app disclosure and consent when the app collects personal or sensitive data, plus encryption between the device and the tunnel endpoint.
This guide shows what each store asks, maps typical VPN data to the questions, and gives an audit method so your “no logs” line survives contact with your own backend.
This guide is general information, not legal advice. Have qualified counsel review your privacy policy and any no logs claim before you publish them.
What Does Apple Require From a VPN App?
Guideline 5.4 holds the VPN specific rules. Use NEVPNManager and enroll as an organization. Before any purchase or use of the service, show an app screen that says what user data is collected and how it is used. Do not sell, use or disclose data to third parties for any purpose, and commit to that in your privacy policy. Where a territory requires a VPN license, put your license information in the App Review Notes. Apps that do not comply are removed.
Three general rules sit on top. Guideline 5.1.1(i) requires a privacy policy link in App Store Connect and inside the app, covering what you collect, how, every use, retention and deletion, and that third parties receiving data give equal protection. Guideline 5.1.1(v) requires in app account deletion when the app supports account creation. Guideline 2.3 says your metadata, including privacy information, must accurately reflect the app.
Then comes the label. Apple’s App Privacy details require a public privacy policy URL on your product page, and your answers must include what your partners’ SDKs collect. Two definitions matter for a VPN. Apple says data is “collected” when it leaves the device in a way that lets you or your partners access it for longer than needed to service the request in real time. And data sent for a request and discarded straight after need not be disclosed: Apple’s example is an IP address sent on a server call and not retained. You can update your answers at any time without shipping a build.
Our reading on SDKs: 5.4 says any data and any purpose, and we found no carve out in the guideline text for analytics or crash vendors. Treat every SDK that sends user data off the device as something to justify to App Review, or remove it. The wider review checklist is in our VPN app store approval guide.
What Does Google Require From a VPN App?
Google’s VpnService policy allows VpnService only where VPN is the app’s core functionality, or a remote server is needed for it, with a short list of permitted categories. You submit a declaration in Play Console that includes a video of the app and a video of your disclosure flow. You document the VpnService use in your listing and encrypt data from the device to the tunnel endpoint. You may not use VpnService to redirect or manipulate other apps’ traffic for monetization. Undeclared or deceptive use can lead to suspension or account termination.
Where the app accesses or collects personal or sensitive data, the policy requires a prominent disclosure with consent. It must sit in the app, not only in the store description or on a website. It must appear during normal use without menu digging, describe the data and how it is used or shared, require an affirmative action such as tapping accept, and stand apart from other disclosures. A privacy policy alone does not count.
The User Data policy and Data safety help add the paperwork. You need a privacy policy link in Play Console and in the app, on a public URL that is not a PDF. You complete the Data safety form even when you collect nothing, and keep it consistent with the policy. “Collect” includes data sent by SDKs. If Google finds a mismatch, the app can face blocked updates or removal. Apps with accounts need deletion in the app and through a web link.
Google also has ephemeral processing: data kept in memory no longer than the real time request needs, and never used for profiles or advertising. It stays out of your listing’s Data safety section. The help page words whether it still goes into the form differently in two places, so follow the form’s own prompts.
Which Questions Does Each Store Ask About Typical VPN Data?
| Data | Apple asks | Google asks | Watch for |
|---|---|---|---|
| Account email or ID | Contact Info: Email Address. Identifiers: User ID | Personal info: Email address, User IDs | Marketing use adds an advertising or marketing purpose |
| Purchase tokens and receipts | Purchases: Purchase History | Financial info: Purchase history | Apple says payment details entered outside your app, which you never access, are not collected |
| Device and install identifiers | Identifiers: Device ID | Device or other IDs | Google’s examples include Firebase installation IDs |
| IP addresses | No named type. Declare by use: location, Device ID or diagnostics | No named type. IP derived location is Approximate location | Access logs, rate limit tables and abuse lists are retention |
| Connection timestamps | Usage Data: Other Usage Data (closest fit) | App activity: Other actions (closest fit) | Tied to an account or device, they are linked data |
| Bandwidth counters | Usage Data: Other Usage Data, if per user | App activity: Other actions, or Diagnostics if only performance | Per user counters are linked data. Apple treats data as unlinked only when identifiers are removed or the link is broken before collection |
| Crash and diagnostic logs | Diagnostics: Crash Data, Performance Data, Other Diagnostic Data | App info and performance: Crash logs, Diagnostics, Other app performance data | Check what the tool attaches: IPs, device IDs, server hostnames |
| Support tickets | User Content: Customer Support | No support type. Closest fit: Messages | Attachments can carry logs and configs |
| Analytics or ads SDK events | Usage Data: Product Interaction, Advertising Data. Device ID. Tracking questions | App activity: App interactions. Device or other IDs. Analytics or advertising purposes | The riskiest row against Apple 5.4 and Google’s VpnService policy |
Rows marked “closest fit” are our reading, because neither store names that data type. Check them against the live forms. If advertising is part of your revenue plan, read our guide to monetizing a VPN app with these rules beside it.
How Do You Audit Declarations Against What You Actually Collect?
Run this before every store submission.
- Inventory every source. One sheet, one row per field: app storage, SDKs, the backend database, application logs, reverse proxy access logs, VPN node tooling output, and third party services such as purchase validation, push, email and the support desk.
- List every SDK, including transitive ones. Pull the dependency list from your Gradle and Xcode builds, then read each vendor’s data collection guidance. Google points to provider guidance and the Google Play SDK Index. Remove what you cannot justify.
- Capture the traffic. Run a release build on a clean test device, before and after each consent screen, and record every host and payload field. Capture at the network level, such as a router or hotspot, because your own app will be using the device’s VPN slot.
- Read the server side. Open every table schema and log format. WireGuard’s own tooling can print each peer’s endpoint, latest handshake time and transfer counters (see the wg(8) man page), so decide on purpose whether anything stores them.
- Map each retained item to an Apple type, a Google type, a purpose, linked or not, a retention period and a recipient. Every difference between that sheet and your forms or policy is a bug.
- Repeat on every release and SDK update. Both stores make you responsible for keeping answers current. The Play side of the release calendar is in our guide to the Google Play target API level requirement for VPN apps.
Why Must a “No Logs” Statement Match the Real Data Flow?
Both stores treat your listing as a promise. Apple’s Guideline 2.3 requires privacy information in your metadata to reflect the app, and Google says a mismatch between declaration and behavior can end in blocked updates or removal.
Regulators can read the same claim as advertising. The FTC’s advertising guide says advertisers need a reasonable basis before an ad runs, for express and implied claims. In the UK, CAP Code rule 3.7 requires documentary evidence for claims consumers are likely to regard as objective. A “no logs” claim is likely to be read as objective and checkable, so make it only if it is true and you hold records that show it. Your audit sheet is a starting point for that file, not a guarantee. Rules differ by country, and we checked only these two sources.
The phrase fails in practice because neither store defines it. Split it into four parts: traffic content and destinations, source IP linked to an account, connection timestamps, and per user bandwidth. Then say plainly what you do keep, such as account email, payment or purchase records, device count and any connection metadata, and for how long. A policy that promises no logs while a database holds a last connected time per user breaks its own claim.
Both stores also carve out short lived data. Apple’s real time wording and Google’s in memory wording describe data that lives only as long as the request. If it is truly discarded, it can stay out of your public label. These are store labelling rules, not a statement of what privacy law requires in your policy.
What Does a Minimal Data Design Look Like?
These are general patterns, not a description of any particular system.
- Separate payer from connection. Let tunnel nodes validate a token without storing the account’s identity.
- Keep session state in memory. Discard it when the connection ends, and avoid per connection records.
- Store entitlements, not receipts. Verify the purchase with the store, then keep the plan and expiry unless you need more.
- Count at the right level. Use server load totals with no user key, and per user counters only where you enforce quotas, with the shortest retention that works.
- Scrub or skip third party events. Strip IPs and device IDs from crash and analytics events where the tool allows, or leave analytics and ads out of the VPN app.
- Publish a retention table and build the deletion path both stores expect for accounts.
Abuse control, device limits and support all want some data. Minimal does not mean none. It means each item has a reason, a retention period and a row in your declarations.
FAQ
Does a VPN app need a privacy policy on both stores?
Yes. Apple requires a privacy policy URL on the product page and a link inside the app. Google requires a link in Play Console and in the app, and the Data safety form needs it even if you collect nothing.
Can a VPN app use analytics or crash reporting SDKs?
Possibly, but each one is a risk. Apple’s 5.4 bars disclosing data to third parties for any purpose, and we found no exception for analytics vendors. Google makes you declare what SDKs collect. Ask App Review, or remove them.
Do I have to declare IP addresses?
Neither store has an IP address data type. Apple says to declare by use and not to disclose an address sent for a request and not retained. Google says approximate location inferred from IP addresses must be disclosed.
Can I say “no logs” in my store listing?
Only if it is true for every system that touches user data and you can document it. Neither store defines it, both treat listing accuracy as policy, and the FTC guidance and UK CAP Code cited above expect evidence for advertising claims. Say what you keep and for how long, and have counsel review the wording.
Do I need Google’s prominent disclosure if I collect nothing?
The policy ties the consent requirement to accessing or collecting personal or sensitive data. The declaration form still asks for a video of an in app disclosure explaining why the app uses VpnService and how, so plan that screen either way. You also need the declaration, a privacy policy and the Data safety form.
Write the VPN App Privacy Policy Last
Your policy only holds up if the order is inventory, minimize, declare, then write. Each step makes the next one shorter, and the audit sheet becomes the evidence behind any no logs line you publish.
Preparing a VPN app for store review? Tell us about your app and backend on the contact page, and we can talk through the engineering side of your store disclosures.
Last checked against Apple’s App Review Guidelines and App Privacy details, Google Play’s VpnService policy, User Data policy and Data safety help, the FTC advertising guide, the CAP Code and the wg(8) man page on 30 October 2026.