Skip to content
Back to all articles

Privacy & compliance: COPPA and GDPR for families

What it means in plain English, what data SafeAnywhere collects (and what it never will), and how that translates into real protection for your kids.

COPPA · GDPR · STRICTER OF THE TWO THE VAULT & WHAT NEVER ENTERS IT Account Policies Routines Alerts Location* MESSAGES KEYSTROKES PHOTOS CHATS BIOMETRICS SOCIAL ACCT INSIDE: WHAT WE HOLD · OUTSIDE: WHAT NEVER ENTERS 4 IN · 6 OUT Minimisation by design The vault is small on purpose
Illustrative data posture. The small vault inside is the entire surface area; the orbit outside is what architecture refuses to touch.

A child-safety product earns its place only if it treats children's data with the highest standards available. "We follow privacy laws" is table stakes; the useful question is what the product deliberately refuses to see, even when the law would allow it. This piece is the honest answer for SafeAnywhere.

We'll cover COPPA and GDPR in plain English, show you the exact four categories we collect and the four we never do, and explain how the architectural choices behind that posture meet the stricter of the two frameworks regardless of where your family lives.

COPPA and GDPR, briefly

Two frameworks, same underlying principle:

  • COPPA (Children's Online Privacy Protection Act, US, 1998, updated through 2024) requires verifiable parental consent before collecting personal information from children under 13, and strict limits on what can be collected and how long it can be kept.
  • GDPR (General Data Protection Regulation, EU, 2018) sets a stricter bar for minors generally, proportionality, minimisation, transparency, data-portability, right to erasure, and goes further on consent, breach disclosure and cross-border data transfer.

We build to the stricter of the two on every design decision. Families in the UK, EU, US and anywhere else get the same posture, by default.

What SafeAnywhere collects, in four categories

Everything we hold falls into one of four buckets. Nothing lives outside this list; nothing is kept "just in case".

  • Policies you set. Rules, routines, time budgets, filter settings, allow/block lists.
  • Safety-relevant events. Filter blocks, category caps hit, high-risk alerts (strangers in DMs, unexpected app installs, location-boundary crossings when you've enabled geofences).
  • Location (optional, opt-in). Only when you turn on geofences or the panic flow. Off by default.
  • Account essentials. The parent's email, the child's age (for defaults), device pairing identifiers so apps can talk to each other.

What SafeAnywhere never collects, by design

These are not features we have and chose to disable. They are capabilities we never built. The device code physically cannot read them, which is the strongest privacy guarantee a product can make:

  • Messages and chats. SMS, Snapchat, WhatsApp, Instagram DMs, iMessage, Discord, in-game chats. None of it. Ever.
  • Keystrokes and passwords. The device never records what your child types.
  • Photos, videos and files. Camera roll, downloads folder, documents, cloud files, all out of scope.
  • Social-media account content. No OAuth into their accounts, no reading of their feeds, no post history.

Biometric data (face, fingerprint, voice) is also never collected or stored, in any form.

Why this posture is the architecturally right choice

A product that reads messages has to defend that choice to regulators, to parents, to children, and to itself every time something goes wrong. A product that never had the capability has nothing to defend. The engineering trade-off is small, the safety-relevant signals we need live at category, time and filter level, not inside private communication, and the trust gain is enormous.

The honest-conversation piece covers the parenting reason this architecture matters. The compliance reason is the same: a child who knows their messages are private tells the truth more often, and a product that cannot read them has nothing to accidentally leak.

Safety and surveillance are different things. Both can be real; we chose both.

How long we keep what we keep

  • Policies you set: kept while your account is active, deletable on request within 30 days.
  • Safety events: kept 90 days by default, configurable down to 7 days.
  • Location (when enabled): most recent position only; no historical trail stored by default.
  • Account essentials: kept while active; erased on account closure within the GDPR-required 30 days.

What to do: once a year, open the Data section of Parental Control and export your family's data. It takes 30 seconds, confirms what's there, and gives you a copy you keep.

Control stays with you

Three concrete rights that live in the Parental Control app, not in a buried policy page:

  • Right to see. Open the Data section to see exactly what SafeAnywhere holds about your family.
  • Right to export. One-tap download of everything, in a portable format.
  • Right to erase. Close the account; everything above is deleted within 30 days.

Location sharing, usage analytics and research-grade telemetry can all be turned off independently. The product keeps working with every optional feature disabled.

How SafeAnywhere's design meets the stricter of COPPA and GDPR

  • Verifiable parental consent. Pairing requires the parent app to approve each device, so no child-side action creates data without the parent in the loop.
  • Data minimisation. The four-category list above is the whole surface area. Nothing else is held "just in case".
  • Purpose limitation. Each category has a stated purpose, documented in the privacy centre.
  • EU data residency. EU customers' data stays in EU regions; the architecture supports regional residency on request for enterprise and schools customers.
  • Breach disclosure. Any security event affecting your account is notified within the GDPR-required 72 hours.

Common questions parents ask about privacy

Does SafeAnywhere sell any data to anyone?

No. We don't sell data, we don't share with ad networks, we don't power third-party profiling. Our business model is subscriptions from parents.

What about anonymised or aggregated data?

Aggregate statistics (how many families use a feature, average weekly block counts) are used internally to improve the product. These numbers are never tied to individual accounts or sold.

Can my child request a copy of their own data?

Yes. The GDPR right of access applies to the child; we provide the export via the parent account because the parent is the legal controller, and we flag it in the dashboard when a child requests it from inside Guard.

What if we leave SafeAnywhere?

Close the account from Parental Control. All data is erased within 30 days; you get a confirmation when the erasure completes.

Both OS tools have broad access to on-device information by nature of being built into the OS; both companies' privacy policies cover what they do with it. SafeAnywhere's architectural choice is narrower by design, we never built the capabilities we don't need, and we're a dedicated child-safety product rather than a platform feature. The browsers-vs-apps piece compares the setups more directly.

Where do I read the formal privacy policy?

Our privacy centre has the full text, data-processing addendum, regional specifics for the EU and UK, and the Data Protection Officer's contact.

The quiet bottom line

Child safety and surveillance are different things. A product that treats them as different builds for both: safety features designed to catch real harm, and a privacy architecture that treats a child's inner life as off-limits. That's the SafeAnywhere posture, and it's why the "never collected" list in our dashboard is as long as the "collected" list is short.

Collect only what helps. Keep it as briefly as possible. Treat the rest as none of our business.

Keep reading

More from the SafeAnywhere blog

Privacy14 min read

Your DPDP rights in Indian education, from Class 1 to college: the parent's playbook for schools, colleges and ed-tech apps

Every Indian parent now has statutory rights over their child's data under the Digital Personal Data Protection Act 2023. Access the data held, correct and erase it, withdraw the consent that permits processing. In 2026 most parents have exercised none of them, mostly because nobody has translated the Act into parent-usable form. Here is the operational companion: sample letter language for a K-12 school, a college and an ed-tech app, the age pivot that changes the parent's role at 18, the indicative response window, the escalation path, and the one-page rights slip. For information only; not legal advice. Confirm with Indian legal counsel before you act on anything specific.

Read article
Parenting11 min read

After the lie: the honest six-week trust-rebuild for Indian families when a child has been dishonest about phone use

A specific Monday in a lot of Indian households this term: a parent discovers their child has been lying about phone use. The usual response is some combination of anger, punishment and a long new rule. The usual outcome, forty-two days later, is a household that has lost more trust than the lie did and a child who has learned specifically not to be caught again. This piece is a different protocol. Six weeks, two tracks (parent and child), both moving together, or the rebuild does not happen. Written honestly for parents of Class 5 to 12 students in India.

Read article
Parenting9 min read

Not your child's dashboard, yours: the 72-hour self-audit every Indian parent should run before the next phone-rule conversation

Every piece of parental-control writing tells you what to do about your child's phone. This piece is about yours. Before the next conversation with your Class 7 or Class 11 child about screen time, run a specific three-day audit on your own phone use. Not to catch yourself out; to see what your child is actually seeing. The findings are usually uncomfortable, specifically useful, and change what the next conversation sounds like.

Read article
COPPA and GDPR explained for families | SafeAnywhere