Card spending on accommodation: reading the central bank's weekly sector figures

The Central Bank of the Republic of Türkiye (TCMB) publishes weekly bank and credit card sectoral spending statistics every Thursday at 14:30, with an accommodation category covering hotels, motels, resorts and timeshares. The figures are nominal thousand lira, unadjusted and provisional, and count card payments at Turkish POS terminals, not guests.

The Thursday release lands, the accommodation line moves, and someone in the meeting asks whether demand has turned. The number cannot answer that question. It can answer a narrower and still useful one, if you know what it counts. This guide explains what the Türkiye card spending statistics measure according to the central bank's own metadata, why a rising lira line is not rising demand, how bayram shifts a weekly series, and how to read the number next to your own booking pace without treating it as a forecast.

What is the central bank's weekly card spending release?#

It is a weekly publication of card spending amounts and transaction counts by sector, issued by the Central Bank of the Republic of Türkiye under the name Banka ve Kredi Kartı Sektörel Harcama İstatistikleri, bank and credit card sectoral spending statistics. It appears every Thursday at 14:30 on the central bank's Electronic Data Delivery System (EVDS), the bank's public data portal, and the series start in March 2014, according to TCMB's sector categories and metadata (accessed 4 November 2026) (opens in a new tab).

The reference period is the previous week: the release describes a week that has already closed, so it is a description, not a forecast. In weeks that contain public holidays, the release follows the calendar of the weekly money and banking statistics rather than the usual Thursday slot, so a missing Thursday is a calendar effect, not a data problem.

Sectors follow the ISO 18245 merchant category codes (MCC), the international scheme that classifies a merchant by what it sells, as updated by the Interbank Card Center (BKM) and shared with banks. One of the published categories is 06 Konaklama, accommodation, which is the line a hotel reader wants. Each sector is published as an amount and as a transaction count, which lets you compute an average payment per transaction as a rough check, though that average is not a room rate.

What does the accommodation line actually count?#

It counts spending made in Türkiye with domestic and foreign bank and credit cards at the POS terminals of banks operating in Türkiye, including virtual POS, and it reports that spending in nominal thousand lira. A card issued by a bank in Türkiye or abroad counts when the spending happens in Türkiye, so the line mixes resident and visitor spending and cannot separate them. This is the coverage described in TCMB's metadata for the series (last updated March 2022) (opens in a new tab).

Cash advances and cash withdrawals are left out, so the series measures purchases, not cash taken out at an ATM. Internet purchases and mail or telephone orders are shared for information only and are not added to total spending, which matters for any hotel that sells online. The metadata also listed customs duty payments as an information-only row; the central bank's implementation-changes note of 20 August 2026 says that row was removed from the publication from the 14 August 2026 reporting period.

Classification follows the merchant's category code, not the hotel's own chart of accounts: a restaurant inside a hotel may sit in a food category, and a spa may sit elsewhere. Category 06 covers hotels, hotels, motels and resorts not classified elsewhere, and timeshares, so the line is broader than hotels and should never be described as hotel revenue.

Why can the accommodation line rise while guest numbers stay flat?#

Because the series is nominal lira with no base period and no deflator: when room prices rise, the same number of guests produces a larger line. There is no base period and no index, so the published figure is an amount, not a volume; nothing in the release adjusts for price.

To get a volume view, divide by a price measure that covers the same activity, such as the accommodation services index inside the official inflation data, and treat the result as an approximation. If you want to understand that index before using it, read what the accommodation index in Türkiye's inflation data measures. The division is your work, not the central bank's, and the result inherits the weaknesses of both inputs.

Transaction counts are payments, not guests: one stay can produce several payments, and a prepaid booking charged through an agency may never appear at the hotel's own terminal. The data carries no nationality, no length of stay, no room nights and no occupancy, so it cannot confirm a guest count in either direction. The same arithmetic applies to the average payment per transaction: it moves with prices, with the mix of merchants and with how guests split their payments.

How do bayram and calendar shifts move a weekly series?#

A week is seven days, and a public holiday moves spending between weeks, changes how many of those days are trading days, and shifts the date from year to year, so week-on-week and year-on-year comparisons both mislead around bayram. Religious holidays move through the Gregorian calendar, so the same week number in two different years can contain a holiday in one and not in the other.

There is no seasonal adjustment in the series, so the reader does the calendar work. A single week is noisy; compare like-for-like windows of four weeks or more, and say which weeks you used. Holiday weeks can also move the publication day, so a late release is not a sign of a data problem.

Booking behavior shifts earlier than the holiday: reservations and prepayments may land in the card data in the weeks before the bayram, while the stay itself lands later. If you plan rates around these dates, the planning work belongs further ahead, as described in bayram and public holidays: planning rates and restrictions early.

How does a hotel read the weekly number next to its own pace?#

As a cross-check on direction, not as a forecast: it tells you whether a soft week was national or yours, and it cannot tell you what your own guests will do next. The series is national and sector-wide; your hotel is one merchant inside one category, so the line cannot confirm your occupancy, your rate or your segment.

Put it beside your own pickup report, the day-by-day record of rooms taken against the same date last year, and compare direction rather than level. The mechanics of that report are covered in how to read booking pace day by day in a pickup report. Remember the lag and the revisions: you are reading last week, and last week's number may change.

Use the series to separate an internal problem from a market-wide one: if the national line and your pace both fall in the same weeks, the question changes from "what are we doing wrong" to "what moved the market". Do not treat it as an early signal of the season; a season is a pattern of months, and this is a flow of payments in one week. If a weekly move would not change any decision you are about to make, do not build a decision on it.

How do you cite the weekly card figures correctly?#

Name the publisher, the series, the reference week and the unit, state that the figure is provisional, and add the house line that public data implies no endorsement. Publisher and series: the Central Bank of the Republic of Türkiye, bank and credit card sectoral spending statistics, with the Turkish name on first use. Week and unit: the reference week the release covers and the unit, thousand lira, because a figure without its unit is not a citation.

Status matters as much as source: weekly figures are provisional and subject to revision, and the release carries no seasonal adjustment, so say both. Cite the central bank, not a bank research page, a brokerage note or a news story that repeated the number. Never convert the line into a per-guest or per-visitor figure; the series contains no guest count to divide by.

To place the release in the wider rhythm of official figures, see which official Türkiye tourism figures come out when. The weekly card series arrives before the Ministry's monthly visitor bulletins and TÜİK's quarterly revenue release, which is exactly why it tempts readers into conclusions the monthly sources would correct.

What to do next#

Download the last thirteen weekly releases from the central bank's data portal, mark the public holidays on the same chart, and set the line beside your own pickup report for the same weeks. Read the direction, deflate before you say growth, and cite the central bank with the week and the unit. The weekly series is a fast, narrow instrument: it tells you what card spending did in Türkiye last week, and it leaves the guest count, the rate and the season to the sources built for them.