Telia CrowdInsights

Telia CrowdInsights data is derived from raw antenna signals captured from mobile devices on the Telia network. These signals are processed to identify staypoints — locations where devices remain stationary for a predefined duration. Contextual data such as population density and land use are incorporated as prior probabilities to refine signal allocation. The resulting statistics are aggregated to administrative boundaries (county, municipality, district, grunnkrets) or 250m grid cells. Data quality and privacy are ensured through K5 filtering, meaning insights are only shown when at least five devices contribute to a given spatial and temporal unit.

The data covers Norway (Oslo and 21 municipalities belonging to Akershus county), Sweden (Stockholm, Hiddinge, Järfalla, Solna, Sollentuna, Botkyrka, Haninge, Tyresö, Sundbyberg, Nacka, Danderyd), and Finland (Helsinki, Espoo, Vantaa, Kauniainen), and includes three product categories: Visits data, Trips data (Origin-Destination Matrices, ODM), and Traffic data.

Visits data

Visits data reports the estimated number of people dwelling in a specific location during a given time period. It comprises three metrics: Foot Traffic (visit counts per spatial and temporal unit), Visit Length (average duration of visits), and Trade Areas (distributions showing where visitors to a location originate from). Foot Traffic is available at both hourly and daily temporal resolution, while Visit Length and Trade Areas are available daily only.

Trips data (Origin-Destination Matrices, ODM)

The ODM data provides information on where people travel from (origin), where they go (destination), and how they move between the two. A trip is defined as each subsequent pair of visits recorded for a device. The basic mode of transport is inferred from device speed: slow-moving devices are classified as walking, faster devices as road/rail, and trips with sufficiently high speed passing via two distinct airports are classified as airplane. The ODM is available as pre-defined spatial levels from county to 250m grid, at hourly and daily resolution.

Traffic

Traffic data shows the estimated number of people traversing specific segments (ways) in the transport network during a given time period. Path inference is used to determine the most likely route taken by a device, using the origin and destination locations together with cell tower pings recorded during the trip. The transport network is derived from OpenStreetMap and includes roads down to tertiary level, as well as rail segments (train, subway, tram) and ferry routes. Traffic estimates are available at hourly, daily, and weekly resolution, and include directional breakdowns as well as aggregate totals.