ICT usage in enterprises, 2025

Significant increase in AI usage among enterprises in 2025

Statistical news from Statistics Sweden 2025-11-06 8.00

In 2025, the use of artificial intelligence (AI technology) increased compared to 2024, according to new statistics from the survey on ICT usage in enterprises. The share of enterprises performing data analytics also rose, while the proportion of enterprises engaged in e-commerce remained unchanged.

The use of some form of AI technology among enterprises with at least 10 employees in the industry sectors NACE 10–82 (excluding 64–66) and 95.1 increased from 25.2 percent in 2024 to 35.0 percent in 2025.

“Statistics show that AI technology is being used by an increasing number of enterprises, but a significant share lacks clearly defined purposes for its use. This is particularly evident in the transport and storage sector, whereas enterprises in the information and communication industry are more likely to have defined areas of application for their AI usage,” says Fredric Nyström, statistician at Statistics Sweden (SCB).

Among the enterprises that reported using AI in 2025, 71.9 percent stated that they used AI for one or more specific purposes. This represents an increase of 2.7 percentage points compared to 2024. However, this also means that just over 28 percent of enterprises did not specify any purpose for their AI usage. The most common purpose was marketing and sales, cited by 41.7 percent of enterprises using AI technologies. This was followed by organization of administrative processes and business management, as well as production or service processes, with 35.0 percent and 26.2 percent respectively. The least common purpose was logistics, mentioned by 5.7 percent of the companies that had used AI.

Among enterprises that used AI, the most common method of acquisition was purchasing ready-to-use software. This method increased by 7.5 percentage points compared to 2024, reaching 62.1 percent. Commercial software or systems modified by in-house staff also increased by 4.0 percentage points to 37.1 percent, while other methods of acquiring AI software or systems either declined or remained nearly unchanged.

Commercial software/systems ready for use continue to be the most common method of acquisition.

Methods of acquiring AI technology used by enterprises, 2024 and 2025

Among the enterprises that stated they did not use any type of AI technology, 11.1 percent were considering adopting AI. This was a decrease of 3.4 percentage points compared to 2024. Among the enterprises considering AI adoption, the most common barrier was a lack of relevant in-house expertise, cited by 74.7 percent in 2025 compared to 71.9 percent in 2024. This was followed by concerns about data protection and privacy violations, with 49.1 percent in 2025 versus 41.8 percent in 2024. The least common reason given was that AI is not useful for the enterprise, reported by 16.9 percent in 2025 compared to 20.2 percent in 2024.

Lack of relevant expertise remains the most common reason for not adopting AI technology.

Reasons why enterprises chose not to adopt AI technology, 2024 and 2025

Increased use of data analytics among enterprises

The proportion of enterprises conducting data analytics amounted to 38.8 percent in 2025, an increase of 3.7 percentage points compared to the most recent measurement in 2023. Data analytics can be performed either by the enterprise’s own employees or by external providers, and both options have become more common since 2023. In total, the proportion of enterprises whose own employees carried out data analytics was 34.5 percent, while 15.0 percent reported hiring external providers—an increase of 3.2 and 3.6 percentage points respectively compared to 2023. The proportion of enterprises that combined internal and external resources amounted to 10.7 percent; the corresponding figure for 2023 was 7.6 percent.

Among enterprises whose own employees performed data analytics, the most common data source used was customer data, 78.8 percent, compared to 68.8 percent in 2023. With the exception of transaction data and satellite data, which decreased between 2023 and 2025, all data sources have become more commonly used for data analytics among enterprises.

Customer data surpasses transaction data as the most common data source for data analytics.

Data sources used by enterprise employees for data analytics, 2023 and 2025

E-commerce sales (via Web and EDI) among enterprises in 2024

In 2024, 36.7 percent of all enterprises with at least 10 employees in the sectors NACE 10–82 (excluding 64–66) and 95.1 conducted some kind of e-commerce sales. Of these, 31.2 percent sold goods or services via the web—such as through their own websites, apps, or trading platforms—while 11.2 percent used EDI (Electronic Data Interchange). 5.7 percent of enterprises used both sales channels. The total share of enterprises engaged in e-commerce remained at the same level as in 2023. Use of the web as a sales channel increased from 29.4 percent the previous year, while EDI usage decreased from 13.7 percent. The share of enterprises using both methods declined from 6.4 percent.

E-commerce activity varied by enterprise size. Among small enterprises, the share was 34.0 percent, while medium-sized enterprises reached 43.8 percent. For large enterprises, the share was significantly higher— 67.4 percent of enterprises in this size category conducted e-commerce in 2024.

Total turnover from sales via web and EDI amounted to 2 718 billion SEK in 2024, corresponding to 26.3 percent of the combined turnover of enterprises with at least 10 employees in the sectors NACE 10–82 (excluding 64–66) and 95.1. Among small and medium-sized enterprises, e-commerce accounted for 12.5 and 18.8 percent of the group’s total turnover, respectively, while the corresponding share for large enterprises was 32.1 percent. However, when looking only at enterprises that conducted e-commerce in 2024, the difference between size categories becomes smaller. In small e-commerce enterprises, e-commerce accounted for 32.2 percent of turnover, while the corresponding figure for medium-sized and large e-commerce enterprises was 37.3 percent and 38.5 percent, respectively.

E-commerce sales also varied across industries. The most e-commerce-intensive industries were Travel agency and tour operator activities (NACE 79), Repair of computers and communication equipment (NACE 95.1), and Manufacture of food products, beverages, and tobacco products (NACE 10–12). However, the highest turnover over the internet was found in Wholesale and commission trade (NACE 46), the combined ICT sector consisting of both manufacturing and service industries, and the Automotive industry (NACE 29–30).

E-commerce sales (web sales and EDI) by industry, 2024

SEK millions and shares of each industry group's total turnover

Definitions and explanations

The statistical news only reports on enterprises with 10 or more employees. More information on the statistics, as well as results for other study domains, can be obtained from the Statistical Database on Statistics Sweden’s website.

NACE – Classification of economic acitvities

The survey reports the following operations:

Manufacturing industry – NACE 10–33
Energy and waste-disposal – NACE 35–39
Construction industry  – NACE 41-43
Trade; repair establishments for motor vehicles– NACE 45–47
Transportation and storage companies – NACE 49–53
Hotels and restaurants– NACE 55–56
Information and communication companies – NACE 58–63
Real estate companies– NACE 68
Other service companies – NACE 69–74, 77–82, 95.1
ICT-sector – NACE 26.1–26.4, 26.8, 46.5, 58.2, 61–62, 63.1, 95.1

For more information on economic classification, see NACE and the corresponding Swedish standard for industrial classification, SNI.

Production and quality of the statistics

The questionnaire on which the data is based, as well as more information on the statistics, can be found in the documentation on the Production of the Statistics (2025). If you have any questions on the quality of the statistics, please refer to the Quality Declaration of the product (2025). These documents are provided here: ICT-usage in enterprises

Next publishing will be

2026-11-05

Publication on Artificial Intelligence in Sweden

A publication that build on these statistics will be published December 11th 2025 on Statistics Sweden's website

The publication will also be made available on Statistics Sweden's new webpage for AI in Sweden.

Statistical Database

More information is available in the Statistical Database