Operational processes were originally designed before of large-scale information technology revolution in industries. Based on this fact, it has been argued that by redesigning business processes with help of information technology, efficiencies in operational processes may be achieved. Process integration and automation should save valuable time and resources at the same time. However, in small and medium sized businesses limited resources reduce possibilities to implement both information technology and changes in business processes. The research in this area concentrates whether on large corporations (in depth or on large samples) or on small and medium sized companies based on statistical information. Longitudinal and deep analysis on business process change in small and medium sized companies is scarce. In depth analysis on small and medium sized companies in the area is scarce. 

This research investigates the savings in labour and capital that are achievable through automating and integrating routine purchase process within a supply chain. This study investigated the area of purchasing and especially routine purchasing of direct production materials. Through the use of action research, quantitative and qualitative analysis, a variety of methods are used in order to give a broader view to the data and results. 

From the practical point of view, the purchase process of production materials is often repetitive work based on information from diverse information systems and sources. This study investigated the possibility of increasing efficiency through automated routine processes. As an indicator of increased efficiency, change in inventory turns was chosen. Three research questions were set: will purchase order automation have an effect on inventory turns, what are the main contributors to any observed changes in inventory turns and are the observed changes relevant in a larger context of small and medium sized companies? 

For the two first questions action research methodology was used with participant observation. Two projects of purchase order automation were conducted and effects on inventory turns were measured. The automation projects were conducted in a medium sized electronics manufacturing company and with two of its suppliers. A controlling sample was collected from the company item transaction database. These results were analysed with quantitative analysis methods. Thedata includes three samples of data between years 2000 and 2004. Results showed that automation does not clearly improve inventory turns but the results are mixed. Observed changes result from improved parameter setting and change in business conditions. 

The third question was treated separately through semi-structured interviews within purchase professionals in medium sized manufacturing companies. Four interviews were conducted in 2014-2015 to verify results from purchase automation projects. The answers were analysed with qualitative methods. It was found that the process of automation and its results are relevant to medium sized manufacturing companies. Special care, however, should be taken while implementing the system and setting its basic parameters. 

From the theoretical perspective, the research emphasizes the assumption that the benefits of business process changes are not always clearly visible but some positive results may be achieved. In regard of human decision-making biases it is clear that they exist and the effects of these biases may be reduced through process automation. The use of several different methods in data collection and analysis opens a larger picture for a deep analysis of the processes and their results. Through action research insights to the actual phenomenon are gained and a longitudinal analysis gives more than a snapshot to it. 

From the managerial perspective, process automation is an option to consider but at the same time it should be part of a larger process change initiative to see that it does not result into sub optimisation of processes at the expense of general process optimisation of a company. It is crucial to understand the biases made by individual decision makers in business processes and see, how the negative effects of these biases may be avoided through careful consideration of process automation.