The Cheapest Cruise Is Not the Cheapest Holiday

Cruise lines advertise a nightly fare. Nobody advertises the other half of the bill — the part you spend standing on a dock in a town you have four hours to see. So we measured it.

Add up the whole week

Take a seven-night cruise for two adults. Add the lead-in fare to what our index says those port days actually cost, and the regions sort themselves in an order no brochure would print. A Caribbean week is the cheapest to buy and the one where the ship is the smaller line item.

In the Caribbean, 55% of what a week costs is spent after you walk down the gangway. The Mediterranean is the mirror image: it has the dearest fare in the table, $386 per night against the Caribbean’s $129, and the smallest ashore share at 28%. Europe sits close by at 31%, where dear fares and dear days ashore compound instead of cancelling.

The honest caveat, up front. That fare is the lead-in price — the cheapest cabin on the ship. Book a balcony and the fare side grows while the ashore side does not, so 55% is a ceiling, not an average. It is still the cheapest week in the table, and still the one where a day ashore outruns the cabin.

Five things cost money on a port day. One of them is nearly all of it.

We price the same basket everywhere. Across 132 fully-priced ports, the split is lopsided.

The excursion is 60% of the bill, which is unsurprising, and the reason the whole index lives or dies on pricing excursions honestly rather than quoting a ship’s brochure.

The genuine surprise is underneath it. Transport is $0.00 at the median port — at 72 of 132 ports, the town is close enough to the berth that walking is the honest answer. The received wisdom that you will be shaken down for a shuttle is, mostly, wrong. It is spectacularly right in a handful of places, and those places are worth knowing: from London’s Thames berths, getting into the city you came to see and back costs two adults about $135 — our estimate, since no single published fare covers the trip.

Levies are the smallest line and the fastest-moving one — 90 of 132 ports charge a passenger nothing at all to step ashore. That is changing, and it is the section after next.

The twenty most expensive port days on earth

Two adults, one day, in dollars. Hover or tap any bar for the breakdown; every port name links to its guide.

Ketchikan is a town of about 8,000 people that will take $654 off a couple between breakfast and all-aboard — more than any city in Europe. It is not a scandal so much as a season: Alaskan operators earn a year of costs in twenty-two weeks, and there is no budget substitute for a floatplane.

Alaska takes the top two, but the company it keeps is the interesting part. Tortola and Nassau are Caribbean ports most people file under “cheap”, sitting third and fourth. Shimizu is there on the strength of a Mount Fuji excursion market and almost nothing else — its food basket is $38, the cheapest lunch in the top twenty.

And Reykjavík earns its place the hard way: it is the only port in the top ten that is expensive in four components at once — excursions, food, a real transport cost, and a per-passenger infrastructure fee.

Why the order might not match other rankings. We rank on dollars. Our internal composite score — which weights the five components by percentile rather than by cash — puts Reykjavík first and Tortola seventeenth. Both orderings are defensible; only one is a number a reader can check against their own credit-card statement, so dollars is what we publish.

And the ten cheapest

A day ashore in Ha Long Bay (Hanoi) costs two adults $143: an $82 excursion, $21 food basket and $25 transfer. It is a gateway port, so the day is priced in Hanoi and the transfer to it is counted. Kuşadası follows at $144 — less than a quarter of Ketchikan, for a port that puts Ephesus within an hour — then Cádiz at $173. Ha Long Bay is cheap for the opposite reason — a low-cost market that absorbs even a paid transfer — but behind it the pattern holds: cheap port days are old cities with dense centres, walkable berths and a local restaurant economy that serves locals rather than a gangway.

Two ports we refuse to rank. Private islands — Perfect Day at CocoCay, Half Moon Cay, Celebration Key — would top the “cheapest” list by a mile, because there is no independent economy to spend money in. The ship owns the beach and bills your cabin. That is not a cheap port day; it is a closed one, and putting it in a ranking of places would be dishonest.

The fastest-moving number in cruising

Port levies are the smallest component in our basket and the only one changing by the month. They are also, routinely, reported wrong — because two entirely different charges get called the same thing.

That last column is the whole story. A Greek levy hits you again at every island; a Mexican one hits you once, at booking, inside the “taxes, fees and port expenses” line. Add them the same way — as most coverage does — and you double-count an itinerary tax at every stop. We keep them in separate columns, which is why Mexico’s headline-grabbing doubling moves our port-day index by exactly nothing.

Two more that belong in a policy timeline rather than a price: Norway’s cruise fee is expected from 2027, and Hawaii’s 2026 “green fee” was enjoined before it ever commenced.

How this was built, and where it is still thin

Every figure carries a source and a capture date. That is the entire point of the exercise, so we have put them in the open rather than in a footnote.

Four weaknesses, stated before anyone else states them:

• Alaska’s regional median rests on two ports. Ketchikan and Juneau are the only Alaskan ports fully priced; Skagway, Sitka, Seward and Icy Strait Point are blocked on tariffs their operators do not publish. We report Alaska as a region with that caveat attached, never without.
• The fare axis is confounded by line mix. An expedition-only port inherits an expedition fare. We never rank on fare — it appears only in the seven-night comparison, labelled.
• Excursion medians need a real market. A port with fewer than 20 resolved products and a wide price spread is not ranked, because the middle of that distribution is not a place. That rule removed seven ports — one of which had been sitting third.
• Some transport values are our judgement, not a published fare. They are marked as such in the underlying data and sit disproportionately in the expensive tail.

Corrections we made to our own data while doing this, because they affected the numbers above: four port coordinates (Amsterdam, Rotterdam, A Coruña, Liverpool) resolved to cargo terminals rather than passenger berths; Belfast was in Maine and San Juan in Washington State; and one port arrived under four different spellings.