Two Phones, One Deal: How the Parent Side of HabitApp Works

Most parental-control apps are a schedule and a block button. You allot the time; the app enforces it; you get a feed to worry at. HabitApp's family mode is built on a different arrangement, and this post is a straight description of it — what each phone does, what the parent sees, and where the edges are.

The shape of it

Two phones, one deal.

Your phone holds the board. You post tasks with a reward on them — screen-time minutes on the apps you've limited, money toward something they want, or both. You verify what's done. And every two weeks, the app reads their fortnight back to you and asks what should change.

Their phone holds the earning. Tasks arrive as offers, not orders — they accept the ones they want, do them, submit. When you verify, the reward lands: minutes into their jar, dollars onto a tally you settle however your family settles things. The app never moves real money; it keeps the score.

The line between the two is the part worth understanding. Which apps are limited is chosen on their phone, with you there. Nothing pays until you verify. And nothing changes on their phone silently — every change you make is announced to them, on their screen, in plain words.

Setting it up

The first run on your phone walks through it, and all of it happens before any payment screen:

Who's who. One screen: what you do, what they do, the line.
The link. You type their name and the app mints an invite link. Send it any way you like — the link is the invite.
Their phone. They install the app, choose I was sent a family link, and paste it — or just tap the link. Then the one step that needs you physically holding their phone: turning on parent-managed mode. Apple asks for your credentials, and Apple's own app picker opens — on their phone, because Apple only shows a phone its own apps. You choose what's limited. Then you choose how much of it you hold — four levels, from observed (you see how it's going; they run their own settings) to managed (you hold the lot, for a child too young to hold any of it). In between, agreed means the app list and the doors out of the deal need your yes; guided adds the rates. Your yes is a six-digit code your phone works out from a code on theirs, or a PIN you set for when you're not there. You can move the level up or down later from your own phone, and every move is announced on theirs.
The first task. Name it, price it, post it.
What you'll see. A sample of the dashboard, so you know what you're paying for before the plan screen.

The child's side is deliberately short, with copy written for a child — short sentences, nothing about willpower or brain chemistry, and your name wherever the app knows it. It ends on: When your parent posts a task, it shows up here. Say yes to the ones you want, do them, and you earn time on your apps. Your parent checks it's done, then it pays.

One subscription covers every child you add. There's no family tier and no per-child price; Apple Family Sharing carries it.

What the dashboard shows — and the honest part

This is the question parents actually ask: can I see how much they're on TikTok?

The answer has two halves, because the app uses two instruments, and it never blends them into one number.

Instrument one: exact, but narrow

Apple's Screen Time framework doesn't hand apps a usage figure. It lets an app set thresholds — tell me when the limited apps have been used for 10 minutes today, 20, 30… — and fires an event when each one is crossed. HabitApp arms a ladder of those on the child's phone and files each crossing with the minute it happened.

What that gives the parent is a floor: at least 90 minutes today on the limited apps. It's exact, it's real, and it's honestly labelled as a floor — the app will never tell you "90 minutes" when what it knows is "more than 90." It covers only the apps you put behind the shield, as a total. It can't tell TikTok from YouTube, because the framework never tells the app which apps those tokens are.

On the dashboard this is the bar chart: fourteen days, one bar each, heights in floors. Two details matter:

A hollow bar is a day their phone never reported in. A small tick is a day it did, and nothing was used. "We weren't looking" and "they used nothing" are different facts, and a dashboard that draws them the same way is lying.
Weekend days carry a faint underline, because the rhythm is usually the story.

Beside it: the kept rate over the fortnight (how often their own check-ins came back yes — blank if they didn't answer, not zero), minutes earned, and days the jar ran dry.

Instrument two: broad, but approximate — and optional

The network view is the per-app answer, and it works differently. It's an on-device filter that sees which hosts the phone talks to — tiktok.com, roblox.com, their content networks — and counts the minutes that had traffic to each. It runs on their phone, on Wi-Fi and cellular alike, and it blocks nothing. The verdict on every flow is hard-coded to allow; it's a telescope, not a wall.

It's off by default. It can only be turned on from the child's phone, by a parent, with Apple's own consent sheet in the way — the child sees exactly what's being switched on. You can turn it off from your phone at any time. You cannot turn it on from your phone. That asymmetry is deliberate, and it's the sentence we'd repeat to anyone reviewing the feature.

Because it's inferring apps from domains, every figure it produces wears a tilde: YouTube ~180 min. It overcounts background refresh and undercounts long-lived connections; it degrades when traffic is hidden behind private relays. It also does two things the exact instrument can't: flag something new (Discord started this fortnight), and note activity after 11pm.

When the network view is off, that block of the dashboard says not watching — and the copy says unknown, not zero. We'd rather show you a gap than a fabricated zero.

So: can you see how much they're on a specific app?

With the network view on: approximately, yes, by app, with the tilde earned. With it off: you see exact floors for the limited apps as a group. The dashboard shows you which of those you're looking at at all times. No competitor in this category separates the two, which is why their numbers have decimal points and ours don't.

(If you've read our earlier post on why the app can't see your screen time — all of that still holds. Threshold events aren't figures, and the network view is a separate, opt-in instrument the child's phone has to switch on.)

The fortnightly session

A feed asks you to watch forever and gives back anxiety. HabitApp gives parents a session instead: every two weeks — sooner if the board has something urgent to say — a card appears on your home screen: Two weeks on Maya's board. Read it back?

It takes about ninety seconds. It shows the fortnight (the same dashboard, so the session and the screen never disagree), then a short deck of questions generated from the record — never more than six, and only the ones the record earned:

3 offers unanswered, oldest 9 days. Is it the pay, or the task? Raise the pay, take it off the board, or leave it.
Hoover the stairs: accepted 8 days ago, not submitted. Retire it, or leave it standing?
Maya is owed $13. Settle up.
Limits paused 9 days ago. Still a truce, or an oversight?
Snapchat started this fortnight. I know it — or post a task about it.
Kept 4 of 19. Not a verdict, a question: pause the limits for a breather, post an easier task, or leave it.

Every answer lands on a proposal screen. Nothing is written until you tap Done. Anything that takes something away — retiring a task, resuming paused limits, unpairing — needs a second tap to confirm. And every applied change is announced on the child's phone: "Hoover the stairs" was taken off your board. Your limits are back on.

The proposal the app is required to make

When a child's record holds — a high kept rate over a month or more — the session proposes handing the app picker back: they choose their own limited apps again; the limits stay on, but the authorship returns to them. The generator is required to reach for this when the record earns it. It's the difference between coaching and surveillance, and it's the one thing in the category nobody else ships, because their business model can't.

What it is not

It isn't message scanning, location, or content classification. It doesn't read what they type or what they see. The core sale of most apps in this category is we read your kid's texts and tell you if something's wrong — a different product with different liabilities, and one this app's own voice couldn't carry.

The pitch is narrower and, we think, stronger: an earned economy instead of an allowance; a fortnightly hour of thinking about your kid's phone, done for you, that ends with decisions instead of dread; two instruments, honestly labelled; and a built-in path to its own removal.

Mechanics referenced: Apple Family Controls .child authorization and FamilyActivityPicker (selection authored on the child's device); DeviceActivityMonitor threshold events (the census floor); NEFilterDataProvider (the network view, allow-only). One subscription across Apple Family Sharing.