How to Record Every UNESCO World Heritage Site You've Visited

Most people who have travelled seriously for a decade cannot answer a simple question: how many UNESCO World Heritage Sites have you been to? They can guess. The guess is almost always wrong in both directions — too low, because a third of the sites they walked through were entire historic city centres they never thought of as "a UNESCO site", and too high, because the famous ruin they remember queueing for turns out never to have been inscribed at all.

This is a solvable problem, and it does not require a better memory. It requires a method: decide what counts, rebuild the list from evidence rather than recall, match it against the official list as it stands today, and then keep it current with about ten minutes of work a year. Here is how to do it, and an honest account of where our own tool helps and where it makes you do the typing.

Decide what counts before you start counting

The World Heritage List is not a list of buildings. It is a list of properties, and a property can be almost any shape. Some are a single monument. Some are a whole city. Some are scattered across a continent.

The Struve Geodetic Arc is the clearest example: one entry on the list, made up of 34 surviving survey station points spread across ten countries, from Norway to Moldova, marking a meridian measured between 1816 and 1855. Four of those points are in Norway, six in Finland, five in Belarus. Stand at one obelisk in Finland and you have, by any reasonable reading, visited the Struve Geodetic Arc — but you have seen roughly three per cent of it.

So set your rule first and apply it to every entry:

• One inscribed property is one entry. Reaching any component of a serial property counts as reaching the property.
• Entering counts. Passing counts only if you decide it does. Driving through a protected landscape on a motorway is a judgement call — make it once, not case by case.
• Airside transits do not count. If you never left the terminal, that belongs in your airport log, not here.

None of these rules is objectively right. What matters is that you use the same one for 1998 and for last month, because a number produced by inconsistent rules is not a record — it is a mood.

Work from evidence, not memory

The method that works for a general retroactive travel log works here too, with one addition: you are matching against a fixed external list, so you can work backwards from the list instead of forwards from your life.

Start with your own evidence. Photo libraries are the strongest source, because most phone photos carry a date and many carry coordinates. Then old boarding passes, hotel confirmations in your email archive, bank statements showing an entrance fee, and passport stamps for the year boundaries. Build a rough itinerary per year — country, city, month — and stop there.

Now flip direction. For each city and region on that rough itinerary, look up which properties are inscribed there. This is where the count corrects itself. Vienna's historic centre is inscribed; so are the Historic Centre of Prague, Budapest's Danube riverbanks, and dozens of other places people visit as cities rather than as heritage properties. Meanwhile the famous ruin you assumed was inscribed may sit on a country's Tentative List, which is the waiting room, not the list itself.

Expect this exercise to take an evening for twenty years of travel, and expect the number it produces to be higher than your guess.

Match against the list as it stands today

The list moves every year, which means any count you quote has a date attached to it whether you say so or not.

After the 48th session of the World Heritage Committee, held in Busan, Republic of Korea in July 2026, the list comprises 1,273 properties across 173 countries. That session inscribed 25 new properties — 19 cultural, five natural, one mixed — and produced first-ever inscriptions for Comoros, São Tomé and Príncipe, and South Sudan. It also added six properties to the List of World Heritage in Danger and removed the Historic Centre of Vienna from it.

Two practical consequences. First, a number you calculated in 2024 is not wrong today, but it is answering a slightly different question — the denominator has grown. Second, if you want a percentage rather than a count, quote the denominator with it: "84 of 1,273" survives the next committee session; "6.6 per cent" quietly does not.

What the record looks like in MapMyWorld

Everything below is in the product today, and each claim names the file it lives in.

Browse the full reference list at /unesco, filtered by country, region or category, or searched by name (client/src/pages/unesco-sites.tsx). Adding a site asks one question — have you been, or do you want to go — and stores it as visited or as a bucket-list entry. Every entry takes a visit date you can backdate to any year (shared/schema.ts), which is the whole point when you are reconstructing 1998.

Once entries exist, three things fall out of them automatically:

• A progress breakdown. Your visited count split into Cultural, Natural and Mixed (server/routes/airports.ts, /api/unesco/progress).
• A share of the whole list. Your statistics page carries your visited count as a percentage of every inscribed property in the reference list (server/routes/stats.ts).
• Suggestions you can act on. The app looks within 200 km of places you have already logged and returns up to ten inscribed properties you have not marked — the "I was forty minutes away and had no idea" list (server/routes/airports.ts, /api/unesco/nearby-recommendations).

There are also public country pages at /guides/unesco, one per country, which show each property with an In Danger badge where it applies (server/services/seoPageService.ts). And if you connect a friend, the comparison view puts your two heritage counts side by side (server/routes/friends.ts).

Where this falls short

Four limitations, stated plainly, because you will hit all of them.

Our reference list runs behind UNESCO's. It currently holds 1,250 properties against the official 1,273 — the July 2026 Busan inscriptions are not in it yet. The list is refreshed by an administrator running a seed job (server/scripts/seed-unesco.ts), not by an automatic sync, so there is a lag after every committee session. You can check the current number yourself: the site list endpoint returns its total.

Each heritage site occupies a place slot. A UNESCO entry is stored as a place with a heritage flag (client/src/pages/unesco-sites.tsx), which means it counts against the Free plan's allowance of 5 places. That allowance is a lifetime one — the counter never decreases when you delete something (server/routes/places.ts) — so a completionist working through twenty years of travel will exhaust it in the first ten minutes. That is an honest mismatch between this job and the free plan. Pro removes the cap at $7.99 a month, $59.99 a year, or $199.99 once for Lifetime.

There is no bulk import. Every entry is added by hand, one at a time. If you are rebuilding eighty sites, that is eighty small interactions. We would rather say so than let you discover it on site forty.

There is no component-level detail and no verification. One inscribed property is one entry, so your four Struve station points collapse into a single record, and nothing checks that you were ever there. This is an honesty system, the same way a record you type yourself is an honesty system. Nothing follows you around to prove it.

A routine that keeps it accurate

Once the backlog is rebuilt, maintenance is small:

• Within a week of returning, add the properties you actually entered, with real dates. A week is roughly how long the details survive.
• Once a year, in August, after the committee session concludes, re-check your matched list against the new inscriptions. A property inscribed this year may be somewhere you went in 2011 — your count can grow without you moving.
• Keep a planned layer. Mark the sites you intend to reach as bucket-list entries and let the nearby suggestions do the work of finding the ones near a trip you have already booked.
• Sweep sideways. The same evidence pile that rebuilds heritage visits also rebuilds the airports you passed through, which is a faster job with a bigger undercount.

The point is not the number. The number is a side effect. The point is that in five years, when you are trying to remember whether you actually walked into the cathedral or only photographed it from the square, there is a dated entry that answers it — instead of a guess that gets rounder every time you repeat it.

Start your record — free.