Article Preview
Top1. Introduction
Organizations that frequently use analytics (derive insights from collected data) to gain competitive advantages are often top performers in their business (Davenport & Harris, 2017; McAfee & Brynjolfsson, 2012). This is in contrast to organizations that mostly make decisions based on gut feeling or rarely use computerized decision support systems; these organizations are rarely classified as top performers in their business. Hence, many organizations try to increase their usage of analytics to become top performers in their business (e.g., transportation, manufacturing, higher education, and health care).
However, not all organizations manage to increase their usage of analytics and become data-driven. It is well-known in the literature that most of the common pitfalls for introducing and using analytics are non-technical, e.g., lack of support from management, lack of skills, poor data quality, or resistance among employees (Berndtsson, Lennerholt, Svahn, & Larsson, 2020; Davenport & Bean, 2018; Halper & Stodder, 2017; LaValle, Lesser, Shockley, Hopkins, & Kruschwitz, 2011).
Establishing a data-driven culture where a group of people frequently and openly discuss insights from collected data is problematic. For example, the share of Fortune 1000 organizations that claimed they had managed to establish a data-driven culture has steadily declined from 28,3% in 2019 to 20,6% in 2023, and the major underpinning barriers are people, culture, and process (NewVantagePartners, 2023). This means that roughly 80% of Fortune 1000 organizations are struggling with establishing a data-driven culture in their teams, despite having advanced analytics (e.g., data mining, artificial neural networks, rule systems) in place.
The literature assessing a group’s maturity in data-driven culture is fragmented and rare. A large number of maturity models in business intelligence & analytics have been proposed in the research literature and by practitioners, e.g. (Eckerson, 2009; Elsa & Xiaomeng, 2022; Halper & Stodder, 2014; Lahrmann, Marx, Winter, & Wortmann, 2011; Lismont, Vanthienen, Baesens, & Lemahieu, 2017). The limitation of existing maturity models in business intelligence & analytics is that they mainly target the organizational level and rarely assess the group level. In Davenport (2022), one of the interviewed Chief Data Officers said that they assessed the shift to a data-driven culture in groups by observing whether people asked analytics-related questions in meetings, e.g., “What does the data tell us? Do you have data to support that hypothesis?”.
A group’s success also depends on how well members collaborate. Previous work has investigated relationships between group development and maturity in a specific domain. Gren, Torkar, and Feldt (2017) used the group development model by Wheelan (2016) to investigate the performance of agile teams. Similarly, Guttenberg (2020) used the model by Tuckman and Jensen (1977) to investigate the performance of Lean Six Sigma projects. In addition, Edmondson (2018) argues that psychological safety is crucial for groups to be successful.
A matrix for assessing maturity in analytics and group development was recently proposed by Berndtsson and Svahn (2022). The matrix is based on progression in analytics (Watson, 2013) and progression in group development (Wheelan, 2016). Similar to any maturity model, the matrix provides a snapshot of the current maturity state.
The objective of this paper is to investigate how the matrix suggested by Berndtsson and Svahn (2022) can be used for assessing maturity in data-driven culture from a historical perspective, and to reflect on how external influence effected the groups maturity, in this case the lack of physical meetings due to the Covid-19 pandemic. The historical assessment can then be used as a starting point for making suitable improvements.