Location History

Pick a child, pick a day, and you get a column of cards. Each card is a stop: a name, an arrival time, a leaving time, and the length of the stay written out as “1 hr 40 min” rather than left for you to work out. Between the stops sit thin travel rows that say how long the journey took. A normal school day usually collapses into six or seven cards, which is short enough to take in at a glance.

Where a stop falls inside a place you have saved, the card uses your name for it. Everywhere else it shows the closest street or landmark the map can offer, plus a small note that this is an unnamed stop. If it turns out to be somewhere your family uses often, you can save it as a place from the card itself, and every past and future visit picks up the new name.

Tapping a card moves the map to that stop and shows the accuracy circle for the readings behind it. That is the part most timelines leave out: a stop built from strong outdoor readings and a stop pieced together from weak indoor ones look identical unless the app tells you the difference. Ours does.

• Arrival time, leaving time, and dwell time for every stop.
• Travel time between stops, without a spaghetti trail of dots.
• Saved place names wherever one applies.
• Clearly marked gaps when the phone could not report.

The child app reports on a schedule and on movement, so what actually arrives is a stream of timestamped readings, each with an accuracy figure. On its own that stream is close to useless for a parent. So the readings are clustered: a run of positions that stay inside a small area for longer than a few minutes becomes a stop, and the movement between two clusters becomes travel. Short pauses at traffic lights do not become stops, and a genuine ten minute wait at a bus stop usually does.

Each cluster is then compared against your saved places. If its centre sits inside a place circle, the stop takes that name. Times come from the first and last readings in the cluster, which is why an arrival time can be a minute or two later than the moment your child physically walked through the door. The phone has to take a reading before anything can be recorded.

Gaps are kept honest. If the phone was off, out of data, or had its location permission revoked, the timeline shows a gap of exactly that length rather than joining two stops with a straight line and inventing a journey. When a phone comes back online after a while, queued readings are uploaded with their original timestamps and the day quietly repairs itself.

The most common use is boring and useful: your child said they went to a friend’s after school, you were half listening, and later you want to place it in the day. Open yesterday, scan seven cards, see a stop from 15:40 to 17:10 two streets away, and you are done in ten seconds. No interrogation, no phone call, no atmosphere at dinner.

The second use is spotting a pattern rather than an incident. A walk home that has quietly grown from fifteen minutes to fifty over a fortnight is the kind of thing no single day would tell you, and the kind of thing worth asking about gently. Timelines are much better at this than live maps, because a pattern is made of days.

The third use is the one nobody wants to need. If a child is genuinely missing, a clear list of where the phone has been, with times attached, is far more useful to you and to anyone helping than a memory of glancing at a map earlier. Most families will never use it this way, and it is still the reason many of them keep history on.

History is kept on a rolling window rather than forever. Once a day falls off the end of that window it is removed from your panel and from our storage, and it does not reappear. We keep the window short on purpose: a month of context is genuinely useful to a parent, while years of a teenager’s movements is a liability nobody in the family benefits from.

You can delete sooner. Any single day can be wiped from its own screen, and the whole history for a child can be cleared in one action from their settings. Deletion removes the stored readings and the stops built from them, so a deleted day cannot be reconstructed later by us or by you. Deleting a child from your account removes their history with them.

Children can ask, too. The family app has a request that lands in your alerts, and a great many parents simply agree to it. Handing a fifteen year old the ability to say “please clear last week” costs you almost nothing and buys a surprising amount of goodwill, which is the whole trade this product is built around.

It does not record audio, take photographs, or capture the screen. It does not know who your child was with, and it makes no attempt to guess by looking at other phones nearby. A stop is a place and a duration, and that is all a stop will ever be.

It does not score behaviour. There is no risk rating on a route, no colour coding a neighbourhood, and no machine learning that decides a stop looks suspicious. Judging what a Tuesday afternoon means is a parent’s job, and we would do it badly.

It does not hide, and it is not exportable to anyone else. The child app shows that history is being kept, and there is no public link, no shared feed, and no data sale that puts a child’s past week in front of a stranger.

Used well, this is a low drama tool. Check it when a real question comes up, keep the window short, tell your child what you looked at when you looked, and it stays what it should be: a shared family record rather than a private file on somebody who lives in your house.