Average nights and spending per visitor: what TÜİK's figure means for a hotel's price

Average spending per departing visitor is all the money visitors spend in Türkiye on a whole trip, divided by the number of visitors who leave the country. TÜİK puts it at USD 1,008 for 2025, with 10.66 overnights per departing visitor. It is a national trip-level average, not a room rate, and it cannot set your price.

A hotelier reads that visitors to Türkiye spent an average of USD 1,008 each, then looks at a room rate that is nowhere near it. The gap is not a failure of the hotel. It is a failure of reading the number as if it were about one room, one night or one guest. This article explains what the average nights and spending per visitor in Türkiye actually measure, why they are not your average daily rate, and how to build a comparison from your own data that will not mislead you.

What does TÜİK's average spending per visitor actually measure?#

It measures one thing: total spending by departing visitors inside Türkiye, divided by the number of departing visitors. Both the money and the people are counted for the whole trip, not per night and not per room.

The denominator is people leaving the country, not bookings, not room nights and not arrivals at a hotel. A visitor who moves between three cities and four hotels appears once in the average. The numerator is everything the trip costs inside Türkiye: accommodation, food and drink, local transport, shopping and services. KPMG Türkiye, in "Turizm Sektörüne Bakış 2026" (page 6, citing TÜİK), puts food and beverage at 21.1% of the total and international transport at 12.8%, which already shows how little of the total is a room.

TÜİK's 2025 table gives USD 1,008 of spending and 10.66 overnights per departing visitor (TÜİK, "Ziyaretçilerden Elde Edilen Turizm Geliri, Turizm Gideri ve Ortalama Geceleme Sayısı", 2026 (opens in a new tab)). Write the nights exactly as 10.66; they are overnights, not days. The overnights figure is also per departing visitor for the whole trip, so it includes every stop, every city and every hotel in that trip.

Package travel is recorded as a lump: KPMG Türkiye's reading of TÜİK shows 28.2% of 2025 spending arriving as package tours. A hotel inside a package sees a room-night price that the per-visitor average never shows. Because the money and the people are both trip-level, the ratio is a market indicator. It tells you the size and direction of the national pot, not the slice any one property can charge for.

Four different measures that all sound like "average nights"#

Public data. No endorsement implied.

Why is "departing visitors" not the same as your guests?#

Because TÜİK counts a person once, when they leave the country, while a hotel counts that same person again at every stay, every arrival and every night.

TÜİK counted 63,917,057 departing visitors in 2025, up 2.7%, and 17.5% of them, 11,175,394 people, were citizens resident abroad (TÜİK, Tourism Statistics 2025, 30 January 2026). Their spending is inside the average too. The Ministry of Culture and Tourism counted 63,941,261 border arrivals for the same year (Ministry yearly border bulletin 2025). The two totals are close but not identical, and neither publisher explains the gap beyond departures versus arrivals, so never mix them in one calculation.

A ten-night trip across two hotels produces one visitor in TÜİK's count and two arrivals in hotel statistics. Any per-guest figure you build from your own system will therefore look different from the national one, by design. Same-day and excursion visitors sit awkwardly in both systems: they spend, but they do not occupy a bed, so they pull the spending average in one direction and the nights average in another.

The practical rule: match the denominator before you compare anything. If your number is per room night, convert it to per departing guest first, or do not compare at all. If the definitions behind these terms are unclear, the tourism statistics glossary on visitors, arrivals and overnight stays covers them one by one.

Trip spend versus room rate: why USD 1,008 is not your ADR?#

Because the trip total is shared across transport, food, shopping and several hotels, while your average daily rate is what one room earns for one night. They are different objects measured on different units.

Work through the layers. The visitor spends USD 1,008 on the whole trip; part of it is accommodation; the hotel's share splits again into rooms, restaurant and extras; only the room part becomes your average daily rate, and that is per occupied room, not per person. For example, a double room sold at USD 120 with two guests is USD 60 per person per night before anything else is bought. The per-person view and the per-room view move apart as soon as occupancy changes.

Do not derive a nightly figure by dividing one published average by another. TÜİK's table gives USD 1,008 and 10.66 overnights; the published per-night spend for overnight visitors is USD 100 (Bloomberg HT, 30 January 2026, citing TÜİK). The two do not reconcile exactly, which is the point: quote each figure as published and let the publisher own it. For a fuller reading of the spending side, see how to read TÜİK's per-visitor spending figures.

Length of stay changes the per-night economics without changing the per-night price: a two-night stay carries the same cleaning, check-in and breakfast setup as a ten-night stay, spread over fewer nights. The honest use of the national figure is directional. It tells you whether the market's spending per trip is rising or flat, which is a budgeting input, not a pricing formula.

What do averages hide about groups, length of stay and season?#

They hide the mix. An average of 10.66 overnights is a blend of three-night city breaks and three-week resort stays, and no single property sits at the blend.

One large group can move a small property's own average for a whole month. For example, a 40-room group on a ten-night package will dominate a 60-room hotel's September figures even if every other booking is a two-night weekend. Nationality and channel mix change spending more than price does: a guest on a package has already paid for the room abroad, so their on-site spending and their visible room rate tell different stories.

Season matters and the annual average erases it. A July beach booking and a February city booking are averaged into one number that describes neither; the official monthly data is the place to see the shape, as covered in reading Türkiye's tourism season month by month.

Short stays are a service design problem before they are a pricing problem: the fixed cost per stay rises as nights fall, which is why a hotel with a 2.55-night pattern cannot copy the cost base of a hotel with a ten-night pattern. The safe habit is to describe your own distribution, not just your own mean: median nights, share of stays under three nights, share of guests on packages. The design side of this is covered in designing hotel service around short stays.

How can a hotel compare its own numbers with TÜİK's safely?#

Rebuild the same ratio inside your own system, then compare the shape of the two series over time, never the levels against each other.

Step 1, fix the denominator: use departing guests for the period, not arrivals and not room nights. A guest who stays twice in the month is one departing guest only if they leave once. Step 2, widen the numerator: total revenue per departing guest, including restaurant, bar and extras, because the national figure includes everything the visitor spends with you, not only the room.

Step 3, produce both versions: per guest per stay and per guest per night. Keep the definition fixed for at least twelve months so the series is comparable with itself. Step 4, describe your mix next to the number: length of stay, share of package business, nationality mix, share of group business. Without the mix, the number invites the wrong conclusion.

Step 5, compare direction only. If your per-guest revenue is flat while the national per-visitor figure rises, the question is which segment left, not which price to raise. Step 6, never use the public figure as a rate target. It is a market total that includes costs your hotel never sees, and it is published for the country, not for your street.

What are the limits of these figures?#

They are national annual averages from a sample survey, published quarterly, and they cannot be turned into a room rate, a forecast or a growth story for 2026.

2026 is flat. H1 2026 visitors were 25,758,545, down 2.39%, and H1 revenue was USD 25.75 billion, down 0.1% (Ministry of Culture and Tourism, August 2026 bulletin; Hürriyet Daily News, 31 July 2026). Never describe 2026 as growing. TÜİK publishes quarters, not a year-to-date total, so quarters must not be added together to manufacture a 2026 growth figure.

No publisher releases a room-level or hotel-only spending figure. The Ministry's occupancy rate is bed capacity used in certified facilities, which is not room occupancy and must never be called hotel occupancy. The 10.66 figure is per departing visitor for the whole trip and is not the same measure as the 2.55 nights per stay in certified facilities; do not subtract one from the other or present one as a correction of the other.

Press summaries round and relabel: 10.66 overnights becomes "10.7 days", which is both a rounding error and a unit error. Quote the table, not the summary. The survey frame and the border counts answer different questions, so a small gap between them is expected and unexplained; treat each series as its own instrument.

Where should you read the primary numbers?#

Open TÜİK's own table for spending and overnights, and the Ministry's accommodation bulletins for facility-level counts. Read the columns, not the press release.

TÜİK's tourism statistics release carries the table "Ziyaretçilerden Elde Edilen Turizm Geliri, Turizm Gideri ve Ortalama Geceleme Sayısı", with the columns for average spending per visitor and average overnights per visitor. That is the source for USD 1,008 and 10.66 (TÜİK, press release 54158, 2026 (opens in a new tab)). Check the release date and the reference period every time. A 2024 value presented as current is the most common error in secondary coverage.

For facility counts, nights per stay and the occupancy rate, use the Ministry of Culture and Tourism bulletins, and keep them in a separate column from the TÜİK series. For revenue totals and visitor counts, TÜİK's quarterly releases are the reference; for border arrivals, the Ministry's monthly border bulletin. Keep a one-page log of the figures you use, with publisher, title and date in the same line. It makes next year's comparison possible and stops numbers drifting between slides.

What to do next#

Write a one-page internal note: your own spending per departing guest, per stay and per night, with your length-of-stay and channel mix written beside it, and TÜİK's USD 1,008 and 10.66 overnights quoted with their source and date in the same line. Put the national figure at the top as context and your own series below it as the instrument you actually steer with. For the wider market picture around your property, the Türkiye pages collect the primary series in one place. Public data. No endorsement implied.