A Travel App That Doesn't Follow You: The Case for Typing It In

In December 2024, a lot of people discovered that the travel history they thought they owned had been deleted.

Google had announced a year earlier that Maps Timeline would move from its servers onto your phone. The migration came with a default: keep the most recent ninety days, erase everything before it. If you saw the notification and acted, your history survived. If you were busy that week, or it landed in a tab you never opened, years of movement were gone — quietly, on schedule, exactly as documented.

Nobody was hacked. Nothing broke. The system did what it said it would do. That is the part worth sitting with, because it says something about what "automatic" actually buys you.

The bargain nobody reads

Background location capture is a genuinely good trade on the surface. You install something, grant a permission once, and a record assembles itself while you get on with your life. No typing, no discipline, no gaps on the week you were too jet-lagged to log anything.

The bill arrives later, in two parts.

You didn't do the work, so you don't control the result. A history you never touched is a history you have no relationship with. You can't correct it, you rarely check it, and when the terms change — a default flips, a product sunsets, a company is acquired — you find out what you actually had at the moment you lose it. The people who kept their Timeline in 2024 were, almost without exception, the ones who had already exported it by hand.

The data has to live somewhere, and somewhere is not neutral. In January 2025 the location data broker Gravy Analytics was breached. Reporting put it at around 17TB of precise coordinates, timestamped, harvested from ordinary phone apps — dating apps, games, weather widgets — and sold onward to advertisers and government agencies. Most of the people in that dataset had never heard of the company. They had simply, at some point, tapped Allow.

That is not an argument that every app doing background capture is careless. It is an argument that a continuous log of one person's physical movements is a uniquely dangerous object to create, and that the safest version of it is the one that was never collected.

So we make you type it in

MapMyWorld has no background location capture. Nothing runs in the background. Nothing asks for an always-on permission. If you were in Hanoi and you don't enter Hanoi, the app does not know you were in Hanoi.

We are aware how that reads next to a competitor's feature list. It is the single most common reason someone bounces off us in the first minute, and we are not going to soften it, because the alternative is worse and we would rather lose that person than build the thing that follows them.

What you get in exchange is an object with a very unusual property for software in 2026: it contains only what you decided to put in it, and it does not exist anywhere you did not put it.

What deliberate entry actually buys

The surprise is that manual entry produces a better history, not merely a safer one.

Automatic capture knows where your phone was. It does not know whether you went inside. It cannot tell the difference between the museum you spent four hours in and the museum you gave up on because the queue went round the block. It records the restaurant you walked past on the way to the one you ate in. It knows you were at the airport; it does not know you were at the airport for eleven hours because of a missed connection, or what the whole thing cost.

Everything that makes a travel history worth having a decade later is a judgement, and judgement is the one thing a sensor cannot supply.

So what accumulates instead:

• A world map coloured by countries visited, against a reference set of 199
• Places you chose to log, with photos, dates and coordinates
• Airports, against a database of more than 10,000 by ICAO and IATA code
• UNESCO World Heritage Sites, against all 1,250 inscribed
• Expenses per trip that roll up across every trip, in 27 categories, split card versus cash and converted to your home currency
• 33 achievements, friend connections, and a public profile if you want a page to point at

Trips hold dates, places and a PDF export, and advance from planned to ongoing to completed by date. But the trip is a container inside the history rather than the top-level object, which is what makes cross-trip questions answerable at all: not "what did this trip cost" but "what does Japan cost me, across four visits, against Portugal".

There is a longer version of that argument in our comparison of travel history apps, and a specific one about visual journals in our Polarsteps comparison.

The ingredients, printed on the front

The obvious objection to all of this is that it sounds like a lot of work invented to justify a missing feature. Fair. So here is the cost, unhedged.

It is about ten minutes per trip. Not zero. If you will not spend ten minutes, you will not end up with a record, and no amount of design on our part changes that. This product rewards a particular kind of person and is genuinely mediocre for everyone else.

There is no CSV import yet. This is the worst thing about MapMyWorld right now and it hurts exactly the people we most want — someone with fifteen years of history in a spreadsheet currently has to type it in. It is the single largest gap in the product and we are not going to dress it up as a feature.

The free tier is small and one limit is unusually harsh. Free accounts get 5 places, 10 countries, 1 trip, 10 airports and 1 friend connection. The place allowance is a lifetime one, not a live count: deleting a place does not give the slot back. Every place added costs us money at a paid mapping API, and an add-delete-re-add loop was billing us repeatedly for the same indecision, so we closed it. The side effect is that Free is a demonstration rather than somewhere to live, and it is better to say that here than let you find out at place six.

Paid is $7.99 a month, $59.99 a year, or $199.99 once. The lifetime option is not a gimmick. A record you intend to keep for twenty years sits badly with a subscription you might cancel in three, and we would rather take the money once than watch a decade of someone's history go dark over an expired card.

Bulk export is Pro-only, as CSV and JSON, alongside a per-trip PDF. That one matters more than it looks: the whole argument of this article collapses if you cannot get your history back out. If we ever become the wrong company to hold it, you should be able to leave with everything, and you can.

The test

There is a question worth asking about any app that offers to remember your life for you: if this company changed its mind tomorrow, what would I still have?

For a service built on automatic capture, the honest answer is usually "whatever I exported, which is nothing". December 2024 was not a scandal or a failure. It was a default, applied on time, to people who had been told. It still cost them years.

Type it in. It takes ten minutes, it produces a better history than a sensor can, and it is yours in the only sense of the word that survives a change of terms.