Bankers internetverksamhet har öppnat nya möjligheter för tredjepartsaktörer inomfintechbranschen som nu kan skapa innovativa ekonomi- och betalningslösningar genom attdirekt interagera med bankkonton. Ett problem för fintechbolagen är att det inte finns någontydlig information om vilka tekniker de kan använda sig av för att integrera med bankernastjänster samt vilken teknik som är mest fördelaktig för deras verksamhet och ändamål.I det här arbetet har de fyra svenska storbankerna Nordea, Handelsbanken, SEB ochSwedbank undersökts, i syfte att identifiera vilka integrationsmöjligheter tredjepartsaktörerhar. I undersökningen fastställdes att fintechbolag kommer kunna eller kan använda sig avnågon av de tre teknikerna SFTP, REST eller webbautomatisering. En testmiljö för de treteknikerna utvecklades för att simulera verkliga användningsfall för fintechbolag. Processor-,RAM- och nätverksanvändning samt total operationstid mättes för de tre teknikerna. För attskicka information var REST den effektivaste tekniken vid mindre datavolymer medan SFTPvar effektivast vid större datavolymer. REST var effektivast för att hämta information,oavsett datavolym. Webbautomatisering var minst effektivt i jämförelse med de andra tvåteknikerna. / Banks internet activity have opened new possibilities for third party actors within the fintechindustry who can now create innovative economy and payment solutions by directlyinteracting with bank accounts. A problem for the fintech companies is that there is noavailable information about what technologies are available for them to use to get access tothe banks services and which technology is the most advantageous for their purpose.In this work, the four biggest swedish banks Nordea, Handelsbanken, SEB and Swedbankhave been examined to find out what options third party actors have to integrate with thebanks services. The examination showed that fintech companies will be able to or are able touse one of the three techniques SFTP, REST or web automation. A test environment for thethree technologies was developed to simulate real use cases for fintech companies. CPU-,RAM- and network usage as well as total operating time was measured for the threetechnologies. For sending information REST was the more effective technology when it cameto lower data volumes while SFTP was the most efficient with bigger data volumes. RESTwas the most efficient for retrieving information, no matter the data volume. Web automationwas the least efficient compared to both the other technologies.

This thesis explores the design and implementation of an API using the Design Science Research (DSR) methodology, aimed at bridging proprietary systems for sustainable lighting solutions. The project addresses the urgent need to replace mercury-based lamps in Sweden with environmentally friendly alternatives due to upcoming EU regulations. The developed API facilitates integration and automation between two distinct systems controlling lighting and power supply. Key research questions include identifying factors critical for a successful API and assessing its impact on client-specific problems and consequences of web automation. The research begins with identifying core problems through stakeholder engagement, followed by iterative design and development of the API. The solution's objectives were derived from practical constraints and stakeholder feedback, ensuring a focused and relevant artifact. The API was partially demonstrated and evaluated through simulations, theorizing its ability to meet the intended goals and providing insights into good practices for web automation. This work contributes to the field of information systems by providing a practical example of API development through DSR, highlighting the challenges and solutions in web automation and system integration. The findings suggest that a well-designed API relies on understanding end-use cases, good communication with minimal assumption and good adaptability. Relying on web automation can also result in post-release complications or pose ethical dilemmas.