Navigating Data Ecosystems Using Action Design Research: A Case Study of Designing, Developing and Implementing a Data Warehouse in Clio Online

Hasse Walin Jensen & Mads Skaaning Hansen

Student thesis: Master thesis


The purpose of this thesis is to investigate how Clio Online (CO) can optimize their storage and integration of data, from their multiple internal and external data sources. This optimization is desired to enhance their Business Intelligence (BI) capabilities. The improved capabilities are desired to lead to a better understanding of their products, customers and commercial activities. Furthermore, the thesis serves to realize the needs of business analysts in CO, who wishes to combine data from multiple sources. The thesis utilizes Action Design Research (ADR), as a methodological framework and its empirical grounds are based on multiple workshops, interviews with both practitioners and end-users, documents, API documentation, as well as, interviews with external consultants assigned to CO. In the thesis, a conceptual framework investigates the theoretical domain of BI and the design of data warehouses (DWH). This lead into a complete walkthrough of the ADR approach, in which an IT-ensemble-artifact is created, resulting in the uncovering of nine design principles. The resulting IT-ensemble-artifact consists of: A custom build Extract, Load and Transform (ETL) tool and a Redshift DWH, nested in the Amazon Services cloud. The ETL tool, ClioWarehouse, serves to extract, transform and load data from Marketo into the Redshift DWH. The DWH can then be accessed from BI tools through a JDBC driver and query the data. Using SQL statements makes it possible to manipulate the data in the DWH. From the creation of the IT-ensemble-artifact, nine design principles were derived: 1) No DWH platform fits all, 2) Organizational readiness to share data, 3) Never expect data to be perfect, 4) Include the end users in the ensemble artifact, 5) Consider compatibility with analytical tools, 6) Consider Response Readiness to data ecosystems, 7) Be well informed about data sources, 8) Care about finding the right the database architecture and 9) Bear in mind your value proposition. Both the ensembled artifact and the design principles were deemed valuable for CO, as they can both be used to guide CO in their process of enhancing their analytical potential.

EducationsMSc in Business Administration and Information Systems, (Graduate Programme) Final Thesis
Publication date2018
Number of pages140
SupervisorsRavi Vatrapu