Location is the most revealing data an ordinary person generates. Where you live, where you work, when the house is empty, which building you visit every Thursday evening — a month of position history answers all of it without needing a single other fact about you.
Which makes it strange that the normal way to measure a walk is to hand that stream to a company, in exchange for a number you could have computed yourself.
The claim
Track Me records to the phone and nothing is transmitted. Concretely:
- No account, no email, no sign-in
- No session created on the server, no point uploaded
- No analytics on the recording page
- Recordings live in the browser's IndexedDB, on the device
The loop is watchPosition → IndexedDB → the map on screen. There is no step in
the middle where a server could be involved, because there is no server call to
make.
Do not take that on faith
Any product can print a paragraph like the one above. Here are three ways to check it in about a minute — and they work on any tracker, not just this one.
Flight mode. Load the page, then turn on flight mode, then start recording and walk. If it keeps working, nothing is being sent. This is the whole test, and it is conclusive: a tracker that uploads cannot function with the radio off.
The network panel. Open the browser's developer tools on the network tab and record for a minute. You should see the page load, the map tiles, and then nothing. Map tiles are the one genuine external request — they come from a tile server as you pan, which reveals roughly which area you are looking at and not your track.
Storage. In developer tools, Application → IndexedDB. The recordings are sitting there, in your browser, readable by you.
The flight mode test is the one to remember. It requires no tooling and no trust.
Where the data actually is, and what that costs you
Browser-local storage is a real trade, not a free win.
Clearing site data deletes your recordings. Not "archives" — deletes. "Clear browsing data" with cookies and site data ticked will take them, and so will some privacy extensions and cleaner apps.
They are per-browser and per-device. A recording made in Safari is not in Chrome, and one made on your phone is not on your laptop. There is no sync because there is no account to sync through.
iOS evicts unused site data after about seven days of not visiting a site, on some configurations. A recording you have not looked at in a fortnight may not be there.
So: export the GPX for anything you want to keep. That file is yours, portable, and outlives the browser's opinion about storage.
Adding the page to your home screen also helps — an installed page's storage is treated as more durable by both platforms than a page you visited once.
What the server does know
Being precise about this is more useful than a blanket claim.
When you load the page, the server serves it. That is an ordinary web request, so it sees your IP address and a user agent, the same as any site you visit. That is where it ends: it does not learn that you started a recording, because starting a recording does not contact it.
The marketing pages carry analytics. The recording page and the dashboard do not — on the dashboard the URL carries a session id, and a session id in Google's logs is exactly the thing that should never be there.
The other half of the product does upload
Worth being clear, because the site does two different things.
The telematics dashboard — where a device pushes points and a browser somewhere else watches them — necessarily transmits. That is what "live" means: the point has to reach the person watching.
Even there, points are relayed through memory and never written to server storage, and the ring holds a thousand of them before the oldest falls out. But it is a network service, and it is a different privacy story to the one on this page. Track Me and Meterless do not use it.
Sharing a walk without giving it away
- GPX to a friend — a file you send, not a link you publish
- A screenshot of the summary — distance and route, no coordinate stream
- Trim the ends. A track that starts at your front door says where you live. Most editors will cut the first and last few hundred metres, and this is worth the ten seconds before posting a route publicly
That last one applies to any platform. A public activity feed with home-to-home routes is a map of where you sleep, and it is the most common way people accidentally publish their address.
Why this design, and not the usual one
The uncomfortable honest answer is that the usual design exists because location history is worth money, and free trackers are not free.
But there is a plainer engineering reason too: for a phone recording itself, the server round trip adds nothing except failure modes. No offline queue, because there is no network to lose. No rate limit, because there is no shared resource. No retention policy to write, because there is nothing retained. The private version is also the simpler one, which is why it is not a premium feature.
Where to go next
- Track your run or ride with just your phone
- Export your run to GPX — the backup this design makes necessary