About the dashboard

The dashboard shows key figures for Icelandic tourism along with forecasts for the next 24 months. It is produced by Augur, an independent Icelandic analytics firm. Forecasts are updated as soon as new data arrive.

How to read the charts

  • The solid line shows measured data — figures that have already been published.
  • The dashed line shows the model's forecast ahead.
  • The shaded area is the forecast's uncertainty band. The wider it is, the greater the uncertainty. The actual figure usually lands inside the band, but not always.
  • The vertical dotted line marks the handover from measured data to the forecast.
  • A dot with a whisker is an early estimate from road traffic (see below), for months that official figures do not yet cover. The whisker shows the likely range.

At the bottom of each chart is a slider that sets the period shown. By default the chart shows the last two years along with the forecast; drag the slider to the left to see the longer history back to 2018.

Forecasting method

The forecasts are built on machine learning models, and the model is tailored to each forecast.

Each forecast is based on the series' own history and seasonal pattern. The models therefore do not know about airlines' decisions — new routes, fewer flights or an airline leaving — until they show up in the numbers. Nearly all foreign visitors arrive by air, so seat capacity limits their number months ahead, but no public monthly data on seat capacity exist.

Easter falls in either March or April, and some travel moves between the two months with it. The models take into account where Easter falls in the series where doing so proves to improve the forecast.

Forecast accuracy

Every forecast is tested on a period the model did not see: forecasts are made as things stood two years and two and a half years before the latest figure and compared with what was later measured. The error is the average deviation as a share of the measured value, by how far ahead the forecast reaches. A "Forecast accuracy" table accompanies every forecast series. On the National, Markets and Regions pages, the "Earlier forecasts" chart also shows the forecast the same models would have made six months before the latest figure, from the data of that time only — the choice of model included — and what was then measured. Isavia's forecast of foreign visitors is shown for comparison.

Traffic and early estimates

The Icelandic Road Administration (Vegagerðin) counts traffic at around 170 sites across the country and publishes the counts the following day. On some roads, for example in South Iceland east of Vík and at Gullfoss, foreign tourists make up by far the largest share of the traffic. The 41 counting sites where the tourist share is highest are grouped together as tourist routes.

Traffic on the tourist routes changes from year to year almost in step with overnight stays. The dashboard uses this in two ways:

  • Early estimate. For months Statistics Iceland has not yet published, an estimate is calculated from the traffic, for overnight stays by region and for foreign visitor arrivals. The estimate is shown only for series where it has proved more accurate than extending the trend of recent months.
  • Traffic. The front page and the Traffic page show traffic on the tourist routes over recent days compared with the same time last year, updated daily. The Traffic page also has a traffic forecast for the main tourist sites over the next 90 days.

Vehicles are counted, not people, and a coach counts as one vehicle. Days on which a counter did not count correctly are estimated from nearby counters.

Derived figures

Derived figures such as average card turnover and the average number of overnight stays are calculated from the relevant data. Their forecasts carry no uncertainty band.

The same applies to market groups (e.g. the Nordic countries) and to countries a user combines into a group of their own: the countries' forecasts are added up, but the uncertainty band of a sum is not simply the sum of the bands, so no band is shown.

Market groups

When a group is selected, the figures of the countries in it are added up. The countries in each group are also listed under the chart's heading.

The breakdown by nationality is not the same in the arrival figures and the overnight stays — the overnight-stay statistics are much finer — so the groups differ between the pages.

Visitor arrivals (Markets):

GroupCountries
Nordic countriesDenmark, Norway, Sweden, Finland
British IslesUnited Kingdom, Ireland
North AmericaUnited States, Canada
Central EuropeGermany, France, Netherlands, Belgium, Switzerland, Austria, Poland
Southern EuropeSpain, Italy
Eastern EuropeEstonia / Latvia / Lithuania, Russia
AsiaChina, Japan, South Korea, India, Hong Kong, Singapore, Taiwan, Israel

The arrival figures break out only 30 countries. All other countries — including Portugal, Greece, Czechia and Brazil — are counted under Other nationalities, which is about 10% of all foreign visitors. They cannot be viewed on their own here. Australia and New Zealand come as a single series and so do not form a group of their own.

Overnight stays (Regions):

GroupCountries
Nordic countriesDenmark, Norway, Sweden, Finland, Faroe Islands, Greenland
British IslesUnited Kingdom, Ireland
North AmericaUnited States, Canada
Central EuropeGermany, France, Netherlands, Belgium, Switzerland (incl. Liechtenstein), Austria, Poland, Czechia, Hungary, Slovakia, Luxembourg
Southern EuropeSpain, Italy, Portugal, Greece, Malta, Croatia, Slovenia, Cyprus
Eastern EuropeRussia, Ukraine, Romania, Bulgaria, Estonia, Latvia, Lithuania
AsiaChina, Japan, South Korea, India, Israel, Turkey, Other Asian countries
OceaniaAustralia, New Zealand, Other Oceania and other territories
AfricaSouth Africa, Other African countries
Central and South AmericaBrazil, Other Central and South America

These groups cover nearly all overnight stays by foreign guests. Outside them stand only Other European countries (about 0.2%) and Iceland, since Icelanders are not a foreign market — they are included in the total "All guests".

Countries that fall into no group can always be selected on their own or combined into a group of your own.

Seasons

The season overview divides the months into four periods: winter (Dec–Feb), spring (Mar–May), summer (Jun–Aug) and autumn (Sep–Nov). Winter is named after both years, e.g. "Winter 2025–26". Darker bars show measured seasons and lighter bars seasons that are partly or wholly forecast.

Revenue and prices

Turnover is VAT-registered turnover excluding VAT by tourism industry, in two-month periods, from companies' VAT returns. The forecast reaches twelve periods ahead.

Revenue per room-night is accommodation turnover divided by overnight stays at all registered accommodation, times guests per occupied hotel room. It estimates what a let room-night brings in on average; it is not the average rate hotels report, which is not published.

The price level for visitors from each market is the price of tourism in Iceland — accommodation, restaurants, transport, recreation and the rest of what visitors buy, weighted by their spending according to Statistics Iceland's tourism satellite account — converted into the market's currency and compared with prices there, so that 100 is the 2019 average. The price level has no forecast.

Data

SubjectSource
Visitor numbers, overnight stays, guest arrivals, hotel room occupancyStatistics Iceland
Foreign card turnoverCentral Bank of Iceland
Rental cars in useVehicle registration figures
Road trafficIcelandic Road Administration (traffic counts)
Cruise ship callsThe DOKK port booking system
Turnover in tourism industriesStatistics Iceland (VAT returns)
Prices and exchange ratesEurostat, BIS (Bank for International Settlements)

Overnight stays cover all registered accommodation: hotels, guesthouses, hostels, campsites, holiday homes and more. Unregistered stays, such as most Airbnb stays, are not included. Hotel nights are shown separately. They are about half of all overnight stays nationally but only a quarter to a third in East, West and North Iceland and the Westfjords. Room occupancy applies to hotels only. Totals for all accommodation are published monthly, their breakdown by region and nationality once a year.

Data for 2020–2022 are distorted by the pandemic. The forecasting models apply a correction for that period so the pandemic does not skew forecasts of the future.

Updates

New data are fetched and the forecasts re-estimated monthly. The date at the top right ("Updated") shows when the current forecast was calculated. Traffic figures are updated daily.

About Augur

Augur is the name under which Viðar Ingason publishes forecasts and analyses. Viðar runs the site as an individual and produces the forecasts with statistical and econometric models on public data.

Companies and institutions: get in touch about subscriptions, custom analyses or API access to the data at augur@augursignals.is.

Updated 25 October 2026, 17:41. The dashboard is free during an open trial period.