Parental Controls After Thirteen
My daughter turned thirteen and the first thing she did was remove Google Family Link from her phone. The moment the account flipped over, the option appeared, and she took it. By the time I noticed, the supervision was gone. The screen time limits, the app approval requests, the bedtime cut-off. All of it.
I didn't know Google did this. I'd been using Family Link for years on the assumption that the controls were mine to set and mine to remove. They weren't. At thirteen, Google treats a child as old enough to manage their own account, and the opt-out is built into the system by design. It isn't a workaround or a bug. It's the intended behaviour. If you've been relying on Family Link the way I had been, this is worth knowing before your child's thirteenth birthday, not after.
She had the phone in the first place because she pressured me into it when she started collège. The case she made was the one every parent of a twelve-year-old hears: all her friends had one, the class WhatsApp group was where homework got discussed, she'd be left out of everything social and academic if she didn't have access. Some of that was true. I gave in and bought her a smartphone. A basic Nokia, or one of the kid-specific devices like a Pinwheel or Gabb phone, would have done the job for the WhatsApp groups and the logistics. It would not have had TikTok on it. I didn't know those options existed in any serious way, and I wasn't thinking about the next eighteen months. I was thinking about the next week of school. I rue the day it was a smartphone.
What the app-store market actually offers
The obvious next move is a third-party parental control app. Qustodio. Bark. Net Nanny. Norton Family. I spent a few days working through them and the picture is messier than the marketing suggests.
Bark, which gets recommended constantly in English-language parenting coverage, doesn't operate in France. Or anywhere in continental Europe. It's available in the US, South Africa, and Australia, and if you try to sign up from a French IP address you can't. I only found this out after I'd already been told by several sources that it was the best option for a thirteen-year-old. Net Nanny doesn't support Android at all. Norton Family works but leans heavily on web filtering rather than the screen time and app controls that are the actual battleground at this age.
Qustodio is probably the most feature-complete of the lot for Android. It has uninstall protection, meaning she can't simply remove it the way she did with Family Link. It offers per-app time limits, scheduled access windows, content filtering. The Trustpilot reviews are 3.3 out of 5, which is generous. The recurring complaints are that tech-savvy children bypass it within a day, customer support is effectively non-existent, and cancelling the subscription is harder than it should be. One parent reported their twelve-year-old hacked it within twenty-four hours.
The whole category has the shape of a market built around parental anxiety rather than parental outcomes. Subscription traps, thin feature sets on iOS because Apple restricts what these apps can do, motivated teenagers sharing bypass methods on TikTok itself. The app you install to block TikTok is undermined by content that circulates on TikTok.
There's a structural problem here that app-level controls can't solve. The moment a child is motivated enough, they factory-reset the phone, or use a friend's, or use a VPN, or switch DNS. The app on the device is the weakest layer.
The thing already in your phone contract
What I hadn't fully registered until I started looking at this properly is that European mobile operators already offer parental controls at the network level. Not as a separate product you install. As a service tied to the contract, in the account-holder's name, that a child cannot opt out of by turning thirteen.
In France, this is the legacy of the 2022 loi Studer, which required every ISP and mobile operator to offer free parental controls by default. Orange, SFR, Bouygues, Free all comply in slightly different ways. Orange's SaferPhone plan with the Protection Enfant pack is the one I've ended up looking at most closely because it's the operator we're already with, but the pattern exists across the market.
In the UK, the four main networks (EE, O2, Vodafone, Three) block 18+ content by default through a system called Active Choice, introduced under government pressure over a decade ago. Vodafone's Secure Net now combines mobile and broadband filtering into a single subscription. Germany has a different model again, built around the Jugendmedienschutzstaatsvertrag and state-approved filtering software.
None of this appears in the English-language coverage I'd been reading. The articles are mostly written for US parents, recommend apps that don't work in Europe, and skip past the fact that the operator in your contract probably already has a control layer the child can't strip off. It's a version of the same pattern I keep noticing in my work on ethical technology for small businesses: the EU-native solution already exists, but nobody is pointing to it.
This matters structurally. An app on the phone can be removed, bypassed, factory-reset. Network-level filtering happens between the phone and the internet. It works on both mobile data and home Wi-Fi (depending on the operator's implementation), it's tied to the contract rather than the device, and the child cannot disable it from their end because they don't hold the account.
The three categories of tool work at different levels, cover different ground, and fail in different ways. It's worth seeing them side by side.
| App-store controls Qustodio, Bark, Norton |
Operator controls Orange, SFR, EE, Vodafone |
Router / DNS filtering NextDNS, OpenDNS |
|
|---|---|---|---|
| Where it lives | App installed on the child's device | Service tied to the mobile contract | At the network level, between devices and the internet |
| Who holds the keys | Parent account within the app | Account-holder (the adult on the contract) | Parent account with the DNS provider |
| Can a 13-year-old opt out? | No on Android with uninstall protection; often yes on iOS | No, not tied to the child's age | No, but a tech-literate teenager can switch DNS on their device |
| Works off home Wi-Fi? | Yes | Yes, covers mobile data and Wi-Fi | Only if installed on the device itself; otherwise home only |
| Covers other household devices? | One install per device, per subscription | Typically limited (Orange Protection Enfant covers 3 devices) | Yes, every device on the home network |
| Screen time and app scheduling? | Yes | Yes, on premium operator plans | No, content filtering only |
| Cost | €40 to €100 a year per family | €2 to €10 a month, depending on operator and rebates | Free tier available; €2 a month for unlimited |
| Main weakness | Subscription traps, poor support, bypass guides circulate online | Limited to the operator's own network logic | No screen time or app management; VPN bypasses possible |
| Best used for | iOS households with no better alternative | The primary layer for a child's personal device | A household-wide baseline across all devices |
Once you see it laid out, the case for combining operator-level controls with DNS filtering becomes easier to make. The two categories cover each other's gaps. App-store tools sit uncomfortably in the middle: doing a bit of everything, doing none of it as well as the infrastructure-level alternatives.
What I'm actually doing
After working through the options, the plan I've landed on has two layers.
The first is moving my daughter from the anonymous pay-as-you-go SIM she's been using (a regular Orange Mobicarte I picked up separately) onto the Orange Série Spéciale SaferPhone plan, with the Protection Enfant pack installed on her phone and her tablet. The plan is €9.99 a month, or €5.99 if you're already an Orange internet customer for your home Livebox, which I am. There's an €8 monthly rebate available for under-eighteens with a copy of her ID, which brings the effective cost down to something like two euros. She keeps her existing number through a migration rather than a port.
The Protection Enfant pack covers the things the app-store options were patching together imperfectly. App blocking (including TikTok specifically). Screen time management by day and time slot. The ability to cut internet access on both Wi-Fi and mobile data from my phone. Content filtering. Geolocation. It works across up to three devices, so her phone and tablet are both covered.
The important structural point is that the contract is in my name, the controls are managed from my phone with a PIN she can't override, and turning fourteen or fifteen doesn't change anything. There's no age-based opt-out built in. The tool belongs to the account-holder.
The second layer is NextDNS on the Livebox. This is the household-wide layer rather than the device-specific one. NextDNS is a French-founded DNS filtering service that lets you block categories of content (social media, adult sites, gambling) or specific domains at the network level for everything connected to your home Wi-Fi. It's free for up to 300,000 queries a month, which is enough for a household. You can choose where your logs are stored (I've set mine to Switzerland, for the same jurisdictional reasons I apply to the rest of my stack), it supports encrypted DNS over HTTPS, and because it works at the network level, it covers everything. Her tablet, any laptop, guest devices, the lot.
NextDNS isn't a replacement for the Orange controls. It doesn't do screen time. It doesn't manage apps on a device-by-device basis. A tech-literate teenager can bypass it by switching DNS settings or using a VPN, though on a phone where the account belongs to a parent that's less straightforward than it sounds. What it does well is operate as a steady background filter that covers every device in the house without any software on the devices themselves.
Together, the two layers cover different gaps. Orange handles her specific devices on any network, including when she's at a friend's house or on school Wi-Fi. NextDNS handles the whole household when anyone is at home.
The part that isn't solved by any of this
None of the above addresses what has actually been the hardest part of the last few weeks, which is that my thirteen-year-old does not see any of this as protection. She sees it as control. From her perspective, she's correct. It is control.
The conversations have been difficult. She has said things I would rather not repeat. "I hate my life" came up, once, in the middle of an argument. We've had the full set of teenage arguments: that everyone else has TikTok, that I am terrible, that her friendships will suffer because she can't send the same memes, that I don't understand what it's like. Some of those things are probably true. Being the thirteen-year-old without TikTok in 2026 is not a neutral position socially. The cost of the boundary is real and falls on her, not on me.
What I've tried to hold onto through this is that the strength of the argument isn't coming from me. The people who built these platforms do not let their own children use them. Steve Jobs didn't give his kids iPads. Bill Gates held off on smartphones until fourteen. Evan Spiegel limits his child to ninety minutes of screen time a week. Chamath Palihapitiya, who helped build Facebook, says his children "aren't allowed to use that shit." France is currently considering legislation to ban under-sixteens from social media. Australia has already done it. The people with the best view of how the products work have been voting with their feet for years, and they've been voting the same way.
That doesn't make my daughter feel better when her friends are on TikTok and she isn't. It doesn't make me feel better when she's angry. It does help me hold the line. The discomfort of this moment is not the same as the discomfort of a mistake.
The reason I'm holding it at this particular age is that thirteen is not a neutral age. She's in the middle of puberty, her sleep patterns are changing, her social world is reorganising, her emotional baseline is less stable than it was a year ago and less stable than it will be in three years. This is also the age at which short-form algorithmic video is most effective at what it's designed to do. The dopamine loop of fifteen-second clips, endlessly refreshed, targeting exactly the content her attention catches on, lands on a brain whose reward system is still developing. That's not a metaphor. It's the mechanism. The product works by exploiting something that isn't finished yet.
She is not, by the way, unaware of any of this. We've talked about digital footprints, about what you leave behind, about nothing online being what it appears to be. She is not permitted to have public social media accounts she can post on, and she has respected that. She doesn't post. She knows the rules of the road better than a lot of adults.
None of that is enough though. A thirteen-year-old who has been briefed, who understands the risks, who has sensible instincts about what to share and what not to, still ends up in the same doomscroll as every other thirteen-year-old on the platform. The algorithm doesn't care what you know. It works on the part of the brain that operates below the level where knowledge lives. This is what separates short-form video from almost every previous media anxiety. The content may or may not be uniquely harmful. The delivery mechanism is engineered specifically to defeat the kind of deliberate, considered engagement we tell children to apply to it.
Which is why I've stopped relying on education as the primary intervention. Education is necessary. It isn't sufficient. The infrastructural controls are there to hold the line that her own judgement, however well-developed for her age, cannot reasonably be expected to hold against a billion-dollar attention engine.
What I don't know yet
I haven't walked into the Orange boutique as I write this. The plan is intact but untested. She may find workarounds I haven't anticipated. The Protection Enfant app may turn out to have its own limitations once it's on her actual phone rather than on paper. She may escalate, or she may settle. The peer pressure may get worse before it eases. I don't know.
What I do know is that the default Google offered me (supervision that the supervised child can remove at thirteen) was never going to be the right tool for this period. And that the thing that has made this solvable at all is that we're in France, with a mobile operator whose contract structure assumes the account-holder is the adult, and that this assumption is encoded in French law rather than left to the goodwill of a Californian platform company.
That sovereignty question is the same one underneath most of what I write about here. Who holds the keys. Whose jurisdiction applies. What the default is when things go wrong. For most of the small businesses I work with, those questions are abstract until they aren't. For a parent with a thirteen-year-old and a phone, the abstraction lasted right up until the moment the controls disappeared.
I'll write again about how this is going once the plan has been in place long enough to evaluate it fully.
