How to Back Up a Travel Record You Spent Years Building

Every app that holds a travel record will eventually stop being maintained. Not most of them — all of them, given enough years. The only real question is whether you find that out before or after the data is gone.

Trail Wallet is the cleanest example in this category. It ran for a decade, passed half a million downloads, and then its developer announced the end on 20 September 2022. The app kept working on phones where it was already installed. What had quietly stopped working underneath it was the backup system, and the announcement is blunt about the consequence: "If something happens to your device, you could lose all of your data." Users were told to open the Summary screen, tap Share, choose All Trip Data, and export to CSV — while the app still opened at all.

Google ran a much larger version of the same manoeuvre on Maps Timeline. Location history moved off Google's servers and onto the phone itself, and every account was given an individual deletion deadline roughly six months after it was first notified. History that was not migrated in time was deleted. Even the accounts that got an automatic hand-off kept only the most recent 90 days; anything older than that went.

Neither of those was a breach or an outage. Both were ordinary product decisions, announced in advance, in writing. In both cases the people who lost years of history were simply the ones who did not act on a notice in time.

A travel record is unusually bad at surviving this kind of event. Most personal data ends up with a second copy by accident — photos sync somewhere, email sits on a server, documents are in two clouds without you deciding it. A travel history is different, because most of it never existed anywhere except inside one app. The date you decided a place counted as visited. The note about why the early ferry was worth it. The rating you gave a museum in 2019. The fact that you passed through Doha twice in the same year. None of that is recoverable from a receipt. If the file goes, you are rebuilding from evidence and memory, which is a genuine project — we have written up the method for doing it, and it takes weeks, not an afternoon.

The fix is boring and takes about five minutes. Get the file out now, while nothing is wrong.

Audit the exit before you commit

Before you put a year of logging into any app, find out what leaving it looks like. There are four questions, and you can answer all of them yourself without contacting anybody.

Is there an export at all, and can you reach it unaided? Look in Settings, Account, Privacy or Data. If the only route to your own history is emailing support and waiting, then that route stops working at precisely the moment support stops answering — which is the moment you will need it.

What comes out, and in what format? An export that produces a PDF summary or a picture of your map is a souvenir, not a backup. You want a file a different program can read.

Is the export behind the paywall? Some apps gate it entirely. That is a legitimate business decision and also a piece of information: it tells you what happens to your record if you ever stop paying.

Does the file actually open? Download it once and look inside. An export you have never opened is a claim, not a backup.

If you are in the EU or the UK there is a legal floor under the first question. Article 20 of the GDPR gives you the right to receive your personal data in a "structured, commonly used and machine-readable format". That wording is deliberate: a screenshot does not satisfy it, and neither does a printout of a summary screen. But a right you have to invoke by writing a formal request is a poor substitute for a button that works today. Use the button where one exists, and treat the legal route as the backstop it is.

CSV and JSON do different jobs

Almost every export you will meet is one of two formats, and they are not interchangeable.

CSV is a single flat table. It opens in Excel, Numbers, Google Sheets or any text editor, and a human being can read it without tools. That readability is the point: a CSV you saved in 2026 will still be legible in 2046 by someone who has never heard of the app that produced it. The cost is structure. Anything nested — a trip containing forty expenses, a place carrying several photographs — either flattens awkwardly or gets dropped entirely.

JSON keeps the shape. Nesting survives, relationships between records survive, and a developer or a future app can reconstruct the original database from it. The cost is that it is unpleasant to read by eye and needs a program to do anything useful with.

If an app offers both, take both. The CSV is what you will actually open in ten years; the JSON is what you would hand to anything trying to reimport your history.

One oddity worth expecting: a well-built CSV export may put a stray tab or apostrophe in front of a field that begins with =, +, - or @. That is not corruption. It stops a spreadsheet from interpreting a place name as a formula, which is a real attack in files that get passed around. Ours does this deliberately — the rule lives in server/routes/subscription.ts.

What ours gives you, exactly

Two exports, both from Settings.

The JSON export is available to every signed-in account on any plan, including Free. It contains your places as complete rows — name, address, city, country, coordinates, status, visit date, rating, notes, and whether the place is a World Heritage site — plus your trips, each with its expense count, its expense total converted to your home currency and its budget, plus a stats block covering places, countries, World Heritage sites and airports, plus the timestamp of the export itself.

The CSV export is a Pro feature. It covers places only, in eight columns: Name, Address, City, Country, Status, Visit Date, Rating, Notes.

Now the part most apps leave you to discover on your own. Three things are not in either file:

• Individual expense lines. A trip carries its expense count and its converted total, not the forty-seven rows behind them. If the line-by-line detail matters to you, keep the per-trip PDF alongside the export — that is a separate feature and it does include the individual items. Our guide to recording what a trip costs covers what is worth writing down in the first place.
• Your airport list. The stats block gives you a number. Which airports produced that number is not in the export.
• Photographs. Place rows carry image references, not the image files themselves.

That is a fair description of a partial export. It is better said here than discovered by you in a year, on the day it matters.

One design decision worth naming plainly: the JSON export is not behind the paywall. The CSV is a Pro feature — Pro is $7.99 a month, $59.99 a year, or $199.99 once — but getting your record out in machine-readable form is not something we charge for, on any plan. An app that bills you for the privilege of leaving has its incentives pointed the wrong way, and the Free plan's five-place lifetime allowance would make such a fee faintly ridiculous anyway.

Then put the file somewhere that is not one device

An export sitting in the Downloads folder of the laptop you are reading this on is not a backup. It is the same single point of failure with an extra step.

Rename it the moment it lands, with the date in the name — travel-record-2026-09.json beats mapmyworld-export (3).json by a distance when you are looking for it in four years. Put one copy somewhere synced and one copy somewhere that is not: an external drive, a second cloud account, a folder on a family member's machine. Two copies in the same account is one copy.

Then open it. Not to read it end to end, just to confirm it is not empty, not truncated and not an error page saved with the wrong extension. Thirty seconds now converts a hope into a fact.

One caution: these files are your movements. Where you live, where you go, what dates you were not at home. That is not a reason to skip the backup, but it is a reason to store it like a document rather than like wallpaper, and to think twice before dropping it in a shared drive.

The half of the problem an export does not solve

Getting data out is largely solved. Getting it into the next app is not — and that includes us. MapMyWorld has no bulk import for personal travel data. The only import machinery in the codebase is administrator-only reference-data loading, in server/routes/admin/media.ts, and it is not reachable from a normal account. If you arrive holding a CSV of three hundred places, you type them in.

Nearly every app in this category is in the same position, which is worth understanding clearly rather than resenting: exports are cheap to build and imports are expensive, because an import has to make sense of somebody else's idea of what a place, a trip or a visit is. Your export is therefore insurance against loss, not a migration path. It guarantees the history survives. It does not guarantee it arrives anywhere convenient.

The practical consequence is to spread your record across fewer apps rather than more, and to keep the authoritative version in the one whose model of a trip actually matches yours. Some of that thinking is the same reasoning behind not handing your movements to background capture in the first place.

A cadence you will keep

Annual is enough, and the trick is attaching it to something you already do rather than to a date you will ignore. When you finish logging your biggest trip of the year, export both files then, while the app is already open and the record is fresh. Two minutes, once a year, and the worst case stops being catastrophic and becomes merely annoying.

Do it today for whatever app currently holds your history — including this one. Then, when the notice arrives, you will already have read the only part of it that mattered.