Which border gate did visitors use? Reading the Ministry's province table without confusing entry with stay

Türkiye border gate visitor statistics count arrivals at border gates and group them by the province each gate belongs to. In 2025, 52,775,055 foreign visitors arrived; gates in İstanbul province took 18,972,699 (35.95%) and gates in Antalya province 16,007,580 (30.33%). The table records where people entered, not where they stayed.

A hotelier on the coast opens the Ministry's border table, sees İstanbul and Antalya at the top of the list, and concludes that demand has moved away from their region. The table never said that. It said where people crossed into the country. After reading this article you will be able to read the province table correctly, keep its numbers apart from the accommodation and revenue series, and run a short exercise on your own province.

What does the Ministry's border statistics table actually count?#

It counts arrivals of foreign visitors at Türkiye's border gates and sorts them by the province each gate belongs to. The unit is an arrival at a gate, not a person, a night or a booking.

The source is the Ministry of Culture and Tourism's 2025 yearly border statistics bulletin (opens in a new tab); the relevant sheet is Table 4, whose title states the grouping: foreign visitors by the province the border gates belong to, and by vehicle type. The population is foreign visitors: 52,775,055 in 2025. The wider border total of 63,941,261 also includes citizens resident abroad (11,166,206), so the two figures answer different questions and should not be swapped.

Because the grouping key is the gate's province, the table is a map of entry points. İstanbul province gates took 18,972,699 foreign entries (35.95%) and Antalya province gates 16,007,580 (30.33%) in 2025.

Source: Ministry of Culture and Tourism, 2025 yearly border statistics bulletin, Table 4. Public data. No endorsement implied.

The vehicle dimension matters: the same province can carry air, land and sea gates, and a reader who sums a province's rows without checking the vehicle column risks double counting or missing a gate. The bulletin counts arrivals, so a traveler who enters the country twice in a year appears twice; nothing in the table links entries to individual people.

This series is not the same as TÜİK's departing visitor count (63,917,057 in 2025, per how Türkiye counts its tourists in the border statistics). Neither publisher explains the gap beyond arrivals versus departures, so do not treat one as a correction of the other.

How should you read the entry shares?#

Read every share against its denominator, which is foreign entries at gates, not nights sold. A 35.95% share means 35.95% of foreign entries happened at gates in that province, nothing more.

Fix the denominator first: 52,775,055 foreign entries in 2025. A share can only be compared with another share built on the same base. Do not set an entry share beside an occupancy rate: the occupancy rate in Ministry-certified (İşletme and Basit) facilities (49.54% in 2025) is not a room occupancy rate, and the bulletin does not state its definition. Do not set an entry share beside a revenue share either. Revenue comes from TÜİK's tourism statistics (USD 65.23 billion in 2025, released 30 January 2026), a survey of departing visitors, and the two series are built from different instruments.

A share of entries is sensitive to airline capacity and route networks: one airport adding a route moves the share without any hotel doing anything. Shares also hide levels. A province can hold a stable share while the national total moves, so always read the absolute number next to the percentage. When you quote a share in an internal report, write the base into the sentence: "35.95% of 52,775,055 foreign entries in 2025", with the publisher and the bulletin year.

Why is the province the gate's province, not the visitor's destination?#

Because the table's grouping key is an administrative fact about the gate, not a statement about the traveler's trip. The province that hosts the gate gets the entry, whoever ends up sleeping there.

İstanbul Airport and Sabiha Gökçen sit in İstanbul province, so every foreign arrival through them is an İstanbul entry, including the ones who connect onward the same day. A visitor who lands in İstanbul and flies to Dalaman for two weeks is counted in İstanbul province and stays in Muğla province. The table cannot show that transfer, and neither can any single table in the bulletin. The same logic applies on land: a gate belongs to the province where it stands, and the traveler's destination may be several provinces away.

An entry gate is not a destination. Treating İstanbul's 35.95% as İstanbul's share of Turkish tourism is the single most common misreading of this table. The practical consequence: this table measures gateway capacity, route geography and border policy, which is genuinely useful for airport and transfer planning, and weak for bed demand. For destination-side questions, change the table rather than the interpretation: the accommodation bulletin counts arrivals and overnight stays by certified facility, which is the closest official measure of where people slept. If you want to know who those people were, see why source markets differ by region in Antalya, Muğla and İstanbul.

Does a high entry share mean hotels in that province are fuller?#

No. Entry share and fullness are measured by different tables with different units, and one does not follow from the other.

Stays are counted in the Ministry's accommodation bulletin: in 2025, certified İşletme and Basit facilities recorded 98,981,217 arrivals and 252,323,483 overnight stays, an average of 2.55 nights per stay. The published occupancy rate for 2025 was 49.54% in Ministry-certified İşletme and Basit facilities, down from 51.91% in 2024. It is the occupancy rate in Ministry-certified (İşletme and Basit) facilities, not a room or hotel occupancy rate; the bulletin does not state its definition.

A province can be a large gateway with a modest bed stock, and a province can hold a large bed stock fed mostly by domestic guests who never appear in the foreign border table. Domestic demand is invisible here. Any conclusion that a province is "full of foreigners" or "empty" needs the accommodation bulletin, not the border table.

Length of stay breaks the link further: two provinces with identical entry counts can produce very different overnight totals if one holds package guests for a week and the other moves people through in a day. If you must combine the two tables, do it explicitly: state the entry figure, state the overnight figure, and say that the difference is transfer and length of stay, which the bulletins do not allocate.

How does seasonality change the picture?#

The yearly bulletin adds twelve months together, so it hides the shape of the year. The monthly border bulletins show that shape, and the province mix moves with it.

A yearly share is an average of a very uneven year. A summer-heavy gate province and a city gateway with steady business travel can post similar annual shares and behave nothing alike month to month. Use the monthly border bulletins for the curve, and note that these files carry a provisional note, so a month can be revised.

Compare like with like: the same month across years, never a month against a year, and never a partial year added up to make an annual growth rate. 2026 is flat. The Ministry's August 2026 bulletin puts first-half 2026 visitors at 25,758,545, down 2.39%, and TÜİK's first-half revenue at USD 25.75 billion, down 0.1%. Plan on a flat year, not on a recovery story. For a hotel, the useful seasonal question is not the national curve but the curve at your own gate, which the monthly file gives you province by province. To know when each series is published, see the Türkiye tourism data calendar.

A mini-exercise: what can your hotel infer from the table?#

You can infer how much gateway traffic your province carries and how it is split by vehicle type. You cannot infer your own demand, your own occupancy or your own guest mix.

Converting entries into room nights needs a length-of-stay assumption that no official table gives you for your province, and any such figure must be labelled as your own estimate. For example, a hotelier in Muğla reads a modest entry share and a large bed stock (Muğla held 243,256 certified beds in August 2026, Ministry figures reported by Anadolu Agency on 24 September 2026) and correctly concludes that many guests arrive through another province's gate or by road.

What are the limits, and where do you find the table?#

The bulletin is a free public spreadsheet published by the Ministry's General Directorate of Investments and Enterprises, and it is only as good as the title you read before the numbers.

Find it on the Ministry's investment and enterprises site (yigm.ktb.gov.tr) under border statistics, as a yearly bulletin and as monthly files, in spreadsheet form. Cite it fully: publisher, bulletin name, table number and year. A bare percentage with no table name is not a citation.

Known limits: entries are not people, the gate's province is not the destination, foreign visitors exclude citizens resident abroad, and no published reconciliation exists with TÜİK's departing visitor series. Do not use the table to say anything about 2026 growth; the year is flat, and the border files are provisional. Where the table is silent, say so in your own reports. A stated limit is more credible than a filled gap. For the exact meanings of visitor, arrival and overnight stay, see the tourism statistics glossary.

What to do next#

Before you quote an entry share again, open Table 4, read its title, and pair it with the accommodation bulletin for the same province and year. The border table tells you where the country was entered; the accommodation bulletin tells you where people slept; the distance between the two is transfer and length of stay, and no official table allocates it for you. Keep the two apart in your own reporting, and treat 2026 as flat.