Why retrospectives are no standard in agile companies

Proceedings of ‏The 2nd International Conference on Applied Research in Business, Management and Economics

Year: 2020

DOI: https://www.doi.org/10.33422/2nd.bmeconf.2020.09.228

[Fulltex PDF]

Why retrospectives are no standard in agile companies

Philipp Simanek

 

ABSTRACT: 

Companies are increasingly using agile structures and practices. Agile frameworks, such as scrum, contain mandatory reflection meetings, also called retrospectives, in order to improve regularly. This exploratory study examined the practice of 14 agile-structured companies in Germany (90 – 2,600 employees) with qualitative interviews and content analyzes. The study showed the surprising result that nine of the 14 companies had not established the retrospective or a similar practice as a standard. They were only established when either the entire organization was strictly aligned to scrum or when top-management prescribed the practice to all teams. Even in companies that had not established retrospectives as a standard, all interviewees rated retrospectives as effective and desirable. But the top-management did not want to make them mandatory because the value of self-determination was rated higher. The self-organized teams should be free to choose their methods. The analysis revealed three factors with a positive influence on the use of retrospectives at team level: 1. The team decides to follow the scrum framework. 2. The use of retrospectives is suggested by team members with a high reputation and prior experience with the method. 3. A colleague in a method advisory role, e.g. agile coach, suggests it. However, in many teams these factors did not apply or have not been effective. As a result, many formally agile teams remain exclusively committed to value creation and direct customer benefit without regularly using time for reflection and improvement – although this is a central agile principle.

Keywords: improvement; institutionalization; scrum; self-organized; a qualitative study.