Kela, the Finnish Social Insurance Institution, is modernising its benefit information system. With the ETL tool supplied by Affecto they expect to see improved work efficiency and easier database maintenance in the long run.

Known as ‘Arkki’ (Arch), the project is expected to last at least ten years because at the same time a large number of modifications due to legal amendments need to be made to the processing systems of which there are dozens.

For the data warehouse and processing tool project included in the Arkki project, Kela has selected Informatica’s ETL (Extract, Transform, Load) tool. The tool supplied by Affecto gathers data from databases and forwards and loads them to the actual data warehouse databases. The new tool is needed for reporting and for the processes that support the compilation of statistics. Additionally, the new tool facilitates and speeds up data processing. Kela has not previously had any actual ETL tool in use.

Greater efficiency and easier maintenance

Jarmo Heikka, Head of Planning of the Technology Team in Kela’s IT Department says that the project to deploy the ETL tool took about nine months. According to him, the project is expected to make it easier to create the ETL code needed for data warehousing and to improve quality.

“We expect to see improved work efficiency and easier maintenance in the long run. We’re just taking the first steps as far as the system project is concerned. Now we have a working ETL tool whose benefits will become evident only later,” says Heikka.

People at KELA have also learnt how to tailor ETL to fit their specific needs. Heikka points out that they are testing the limits of the new tool. Affecto helped Kela’s experts increase their skills in using the data warehousing tool in various situations.

“Deployment went pretty smoothly, actually. As well as Affecto’s experts, we were able to draw upon in-house resources. A slight delay relative to the original timetable was due to the fact that we had to fine-tune the code in order to squeeze the response times to conform the time limits required for the entire ETL process,” explains Heikka.

Adaptation to the needs of the various data volume categories

Of the five data volume categories Kela has defined, the system has now been adapted to the first. In this category, the 24-hour maximum data volume is over 100 gigabytes and over 1,000 million rows. The other categories, in which the data volumes and number of rows are even higher, are also to be included in the project. Heikka expects it to take years before they reach data volume category five.

The ETL tool is used for retrieving and loading the data to the data warehouse while the reports are generated using another tool.

Informatica’s ETL tool was selected after competitive bidding in which the overall economy of the solutions was evaluated. The award criteria were based on the public procurement process where the features and prices were ranked according to the points given. The contract was awarded to Affecto and Informatica.

“We’ve made a decision that will have long-term implications and believe that this tool will be good for years to come. New technology helps us keep abreast with the mainstream of technological advancement,” says Heikka.

Among the long-term objectives established for the system are to improve efficiency in utilising existing data, enhance the transparency and traceability of the data processing chain, make it easier to combine data retrieved from various sources, and improve efficiency in processing large volumes of data.

Kela in brief

Kela looks after basic security for all persons resident in Finland. The terms and conditions of the benefits provided by Kela are defined in legislation.

Kela is also responsible for providing information about its benefits and services, for undertaking research to develop social security further, and for compiling statistics, estimates and projections required for the planning and monitoring of benefit programmes and other operations.


Kela’s ETL system is used for retrieving information from a wide range of databases and forwarding it to specific reporting and statistical systems for further processing.


The ETL system is expected to:

  • improve efficiency in using the databases
  • facilitate maintenance
  • facilitate code creation and enhance quality