How to choose an AirTag history time range

Pick the narrowest window that contains the disputed event: a day either side of it, not the whole month. A tight range is faster to read, easier for somebody else to check, and harder to argue with. Empty stretches inside it are data rather than a bug, and how far back you can go at all is set by when you connected and how long recording ran after it.

How do you pick a range worth looking at?

Start from the claim, not from the calendar. There is almost always one sentence you are trying to support or knock down: "it left the depot on the 24th", or "the bag never went through Frankfurt". The range is whatever contains that sentence plus a little air on each side.

1. Write down the disputed event and the time somebody else says it happened.
2. Set the start of the range at the last moment both sides agree on: the pickup, the check-in, the handover.
3. Set the end at the next agreed moment: the delivery, the carousel, the arrival scan.
4. Add roughly a day of margin at each end, so the disputed part sits inside context rather than at the edge.
5. Look at the result before you export it. If the window is unreadably busy, tighten it; if it is empty, widen it before concluding anything.

For a move, the useful ranges are usually the loading day, each long stop, and the delivery day: three narrow windows rather than one five-week sprawl.

Why not just export everything?

Because the person you send it to is not curious, they are busy. A month of points polled every 5 minutes is thousands of rows, and thousands of rows is a file somebody scrolls once and puts down.

• A narrow range can be checked in a minute, which means it actually gets checked.
• It shows you understand your own claim, which changes how the reply is written.
• It leaves less surface for an unrelated oddity, such as a stray point or a quiet weekend, to become the subject.
• It is easier to line up against the other side's scan times.

Keep the full export for yourself all the same. If somebody asks what happened either side of your window, producing it immediately is worth more than the narrow file was. Refusing to, or not having it, reads as editing.

What does an empty stretch inside the range mean?

That nobody walked past. An AirTag has no GPS and no SIM: a point exists only when a passing iPhone relays a sighting, so silence is a fact about the surroundings rather than a fault in the recording.

• A sealed warehouse or a locked yard overnight: hours with nothing.
• A container stacked below deck: days, sometimes the whole crossing.
• Rural road at night: long stretches broken by a point when a car passes.
• A city street: points close to every poll.

Gaps are shown as gaps rather than smoothed into a line, and that is deliberate. An interpolated route invents positions the tag never reported, which is exactly the thing that would make the whole record easy to dismiss.

A gap can still carry an argument. If a tag reports from the same warehouse on Monday morning and Thursday evening with nothing between, the claim that it moved on Tuesday has to explain how it came back.

What limits the range you can pick?

You can only select inside what was recorded. Nothing rolls off the back: a point saved on your first day is still there a year later. What bounds the range is the other end — recording starts when you connect and stops when your tracking window closes, so the plan decides how many days of the journey exist to select at all, not how long they survive.

How long does a dispute take to surface?

Longer than the trip, almost always. A shipment arrives, the boxes sit for a week or three, and the damage or the missing crate turns up when somebody finally opens box 41. The argument about what happened in transit starts after that.

That is the mismatch worth planning around: tracking records while live, while claims surface weeks later. Exporting your timeline when a journey ends ensures you have a standalone copy ready to hand over.

• Export the loading day and the delivery day as soon as each happens, rather than intending to later.
• Export any long stop the week it occurs: a depot stall is usually the fact in dispute.
• If a claim window runs for months after delivery, size the tracking window around the claim, not around the journey.

How do you read the range once you have it?

• Read the timestamps first and the map second. The times are what the other side will contest.
• Note which timezone you are reading, and use one throughout.
• Treat clusters of points as a stay and single points as a passing sighting.
• Look for the last point before a gap and the first point after it, because together they bound what could have happened in between.
• Check the accuracy you are implying. A point places a tag near a place, not at a particular door.

The stops-and-dwell view on The Move pass does the clustering for you and flags the long stays, which is the reading most people do by eye anyway.