Eleven hotels in twenty nights

There is a moment, midway through a long trip, when you repack your suitcase without knowing what day it is.

It's not a problem. At the time it's almost funny. But a few weeks after you get home, something odd happens. Someone asks how the trip was, you say it was excellent, and that's true. Then they ask what you did on day eight, and there's a pause.

Not a bad memory. No memory at all.

The symptom nobody has a word for

Two or three days out of twenty-one that leave nothing behind. You describe them vaguely — "a bit flat", "a driving day", "we went through some villages". They didn't spoil the trip. They simply added nothing to it.

And because the rest was good, you don't make a thing of it. You put it down to tiredness, or chance.

That's where it gets interesting: those days aren't chance. They have a structural cause, the cause is the same across every operator, and it can be measured.

Why we started counting

August 2026, back from three weeks in the Baltic states. First verdict: almost perfect, except for two or three slightly flat stages that could have been merged.

"Flat" isn't a diagnosis. It's an admission that you can't name the problem.

So we dug out the older documents — Australia, Réunion, Croatia, Iceland, the American West — and did the dumbest possible thing: counted the nights. How many nights, how many hotels, how many times you repack.

Itineraries from different operators, 2017 to 2026. Not to indict operators, who did good work. To find out whether the impression had a shape.

What the count shows

Different operators. Four continents. Nine years between the oldest and the most recent. None of them knows the others.

And in every case: a different bed every day and a half.

A number like that on one trip means nothing. On six trips designed by people who have never spoken to each other, it means something else. This isn't individual carelessness. It's an industry default.

It isn't laziness, it's an incentive

The explanation is simple, and it isn't dishonest.

Every stage of a tour is a billable line. And an itinerary that ticks twelve places looks more generous than one that ticks six. On the quotation, the first one feels like better value.

The unsettling part is that it isn't even wrong. With twelve stages you genuinely see more. You just retain less. And nobody — neither the seller nor the buyer — has a word for that difference at the moment of signing.

Two times the number lied to me

A raw count gets things wrong, and that needs saying.

Six campgrounds, two parks. On the camper trip through the American West: six different pitches in six nights. The number screams fragmentation. In reality those six nights covered only two national parks — we weren't moving, we were changing pitch because that was what remained bookable. Three nights in Moab to explore two enormous parks has nothing in common with three nights in three different towns. Counting beds overstates the problem here.

Iceland had no choice. The worst figure in the set — twelve lodgings for fourteen nights. But the Icelandic ring road has a fixed circumference: to close it in two weeks you have to drive a certain number of days, and every driving day forces a bed. The geometry of the country forces about eight. The itinerary uses twelve.

So: roughly half imposed, half chosen — and only the second half is open to criticism.

What it actually costs

This is where the flat days come from.

The third castle nobody gets out of the car for. On the weakest day of the Baltic trip: three medieval towns back to back. The operator's own documents describe all three — and the three paragraphs are interchangeable. Same skeleton, almost sentence for sentence: founded in the thirteenth century, castle of the Order, Hanseatic prosperity, Reformation, wars, decline, independence in 1991. Three times.

When the description is interchangeable, so is the stop. By the third one you stay in the car. That isn't ingratitude, it's diminishing returns.

A driving day dressed as a sightseeing day. Three hours behind the wheel plus three cultural stops is a transfer day in costume. On the programme it appears as a full day. It isn't one.

And the trip you didn't order. The Baltic itinerary was commissioned with a clear bias towards nature — call it seventy per cent nature, thirty per cent culture. Counted in overnight stays, it delivers forty to forty-five per cent culture. The gap isn't dramatic, but it runs in the same direction on every operator-built itinerary in the set. Cities are easier to sell than trails, and a site's fame is a simpler criterion to work with than a rhythm.

Telling detail: the one itinerary built without an operator lands exactly on the requested profile.

Three questions to ask before you sign

No complicated method. Three questions about the quotation, before you pay.

1. How many nights per stage — and why so few?
If there are many one-night stages, ask whether the shape of the route forces it. A loop across a large territory sometimes does. A straight line between two cities rarely does.

To find the number that suits you, see How many hotels for a road trip.

2. On driving days, which stops — and do they resemble each other?
Three stops of the same kind on the same day is one too many. Read the descriptions: if you can swap them without anything changing, so can the operator.

3. Count city nights and nature nights separately.
Does the result match what you asked for? It's the most common gap, and the easiest one to fix before departure.

If your itinerary comes from an AI assistant, the same questions apply, and a few more: seven checks before you go.

What we don't know

That independent operators converge on the same reflex is solid — a coincidence of taste can't produce that. Everything else — how many nights make a stage comfortable, how much rest is enough, the right balance of nature and culture — is a parameter of the engine, set per traveller, not a general truth.

And there's a more awkward limit. A travel memory detects excess, never absence. A day too many is felt. A day that should have existed stays invisible, because you don't know what you didn't do. Any analysis built on trip memories therefore leans towards the over-cautious. Including this one.

Manual count from original tour documents and quotations, check-in and check-out dates per property. The source documents contain personal and pricing data and cannot be published; the count is reproducible on any comparable itinerary.

Think a figure is wrong, or do your own documents say otherwise? Write to us — corrections are published.