Implementing Robo-advisor in a Danish Bank: The role of the middle manager

Implementing Robo-advisor in a Danish Bank: The role of the middle manager A critical perspective

Anne-Christine Rosfeldt Lorentzen, Aarhus University, acrosfeldt@mgmt.au.dk

Introduction

Intelligent machines are continuously being implemented in various types of organizations providing better performance (Davenport & Ronanki, 2018), generating new occupations and making existing jobs evolve (Manyika et al., 2018). The machines are also filling more roles in management (Fuchs, Silverstone, and Thomas 2016). Fuchs et al. (2016) categorize three such roles by their degree of autonomy and proactivity: Assistant, advisor, and actor (table 1). This paper-in-progress focuses on an advisor, which bank managers use when talking to customers.

Assistant	Advisor	Actor
Creating scorecards Maintaining reports  Monitoring the environment	Answering questions Building scenarios Generating options	Evaluating options Making decisions Budgeting and planning

Passive                                                                                                                                                                                                                                        Active  Table 1: Accenture Strategy analysis, 2016: 2  We do not know much about the middle managers’ role in implementing systems based on artificial intelligence (Marr, 2018; Saphir, 2018; Felten et al. 2018). However, middle managers play an important role when implementing change in general. The literature traditionally describe middle managers as change agents responsible for aligning subunits with the organization (Hrebiniak & Joyce, 1984; Nonaka & MacMillan, 1986; Nutt, 1987; Raes, Heijltjes, Glunk & Roe, 2011). Disruptions are often perceived as active resistance (Bower, 1970; Ketokivi & Castañer, 2004) driven by individual middle managers (Courpasson, Dany, & Clegg, 2012; Zoller & Fairhurst, 2007) based on their rational self-interest (e.g. Guth & MacMillan, 1986; Meyer, 2006).    Taking a less linear and sequential perspective (e.g. Carter, Clegg & Kornberger, 2008; Chia & MacKay, 2007; Chia, 1999; Tsoukas & Chia, 2002) disruptions are less deliberate and caused by  e.g. social embeddedness (Regnér 2003); difference in sensemaking between top and middle (Balogun and Johnson 2005; Mantere, 2013; Mantere & Vaara, 2008); middle managers’ emotional stress (Thomas & Linstead, 2002); and in general the difficulties of being a middle manager (e.g. Harding et al., 2014).  Both perspectives provide a basis for understanding disruptions related to the role of branch managers and department managers when implementing robo-advisor. However, they fall short in explaining the role of the IT Director: Preliminary observations depict both the IT Director and the employees as wanting robo-advisor to solve specific time-consuming tasks. Still, the employees are very frustrated interpreting the quality of robo-advisor negatively, even though it is state of the art. Taking a point of departure in both the ethnographic data and brief literature review, this paper seeks to answer the following: What is the role of the middle manager when employees become critical of the quality of a well-functioning robo-advisor that they do want solving time- consuming tasks?   

Approach 

Ethnography (see Spradley 1980) proposed itself as a helpful method, because studying people in their everyday activities help the researcher to understand the complexity, intricacy and mundanity of organizational life (see Ybema et al. 2009). Scholars as Whyte, Goffman and Beckman (Chicago School) stated that fieldwork, participant observation, and native interpretations are the essence when investigating the empirical world (Becker, Hughes & Strauss, 1961; Blumer, 1964).  The empirical setting of this ethnographic study is a middle-sized Danish bank. A ‘typical instance’ (Cunliffe & Karunanayake, 2013) because banks are one the many types of organizations in which functions are continuously being automated, and an emblematic case (Silverman 2014: 73) in which involvement, dialogue, motivation, continuously educating managers, and the well-being of employees are of high priority, resulting in the organization often being nominated as one of the best places to work in Denmark.  Building on Cunliffe’s (2015) recommendations, data collecting methods include observing, informal conversation (see, for example Fenton and Langley 2011) questions at meetings with top management, middle management, and employees. Interviews, participation at events, participation in strategy seminar for the top management, and document access. Furthermore, data consists of posters made by middle managers and employees. Notes were made during the activities or as soon as possible. The ethnographic data consists of approximately 292,5 hours (table 2): 

Participant observation Approximately 211,5 hours	Non-participant observations Approximately 56 hours	Informal conversations As part of the observations:	Workshop (big focus group)  Approximately 8 hours	Interviews    Approximately 17 hours
Meetings with top upper middle management/shadow	5 x 8 hour meetings with all middle managers (35)	E.g.:    During coffee breaks from	Workshop with employees representing all departments in	Branch managers: 3 hours (1h40m, 20m, 1h)

   

advisory board: 75 hours    Meetings with top management: 25 hours    Lunch with different ‘colleagues’: 45 hours    Strategy kick off event: 8 hours    New years party: 9 hours    Meetings and workshops with external consultants and top management/upper middle managers: 21 hours    Every day work in the organization: + 38,5

5 x 1 hour morning meetings in branches    1 x 8 hour internal course on communication    1 x 3 Meeting with union people

meetings in the top management, and with the middle management group.    During transportation with upper middle management when meeting consultants, and with employees to parties/activities

all 16 branches discussing strategy and in that regard also Robo-advisor.

Union people: 2 hours    HR director: 3 hours (6 x 30m)    Communications director: 2,5 hours (5 x 30m)    Head of business development: 1 hours (2 x 30m)    IT Director: 4  hours (2 x 1h + 4 x 30m)    IT  Implementing consultant: 45 minutes (1x)

As is it often the case with rich ethnographic descriptions the paper-in-progress is encountering the problem of presenting it briefly (Van Maanen, 2010). Vignettes presented itself as a way to solve this issue, as it is a well-established way of communicating the context and ‘feel’ of ethnographic data (Barter & Renold, 2000). Therefore, the paper will present data through vignettes, and findings will be discussed in relation to the literature on change and middle management. 

Litterature 

Accenture Strategy analysis, 2016: https://www.accenture.com/t00010101t000000z w /br- pt/_acnmedia/pdf-13/accenture-strategy-wotf-machine-csuite.pdf    Balogun, J., & Johnson, G. (2004). Organizational restructuring and middle manager sensemaking. Academy of Management Journal, 47, 523–549.    Barter, C., & Renold, E. (2000). ‘I wanna tell you a story’: The use of vignettes in qualitative research. International Journal of Social Science Methodology, 3, 307–323.    Becker Becker, H. S., Hughes, E. C., Geer, B., & Strauss, A. L. (1961). Boys in White: Student Culture in Medical School. Toronto: University of Chicago Press.Blumer (1954): ‘What is wrong with social theory?’, American Sociology Review, 19/1; 3-10    Bower, J. L. (1970). Managing the resource allocation process. Boston, MA: Harvard Business School.  Carter, C., Clegg, S. R., & Kornberger, M. (2008). Strategy as practice. Strategic Organization, 6, 83–99.    Chia, R. (1999). A ‘rhizomic’ model of organizational change and transformation: Perspective from a metaphysics  of change. British Journal of Management, 10, 209–227.    Chia, R., & MacKay, B. (2007). Post-processual challenges for the emerging strategy-as-practice perspective:  Discovering strategy in the logic of practice. Human Relations, 60, 217–242.    Courpasson, D., Dany, F., & Clegg, S. (2012). Resisters at work: Generating productive resistance in the workplace. Organization Science, 23, 801–819.    Cunliffe, A. L. (2015): Retelling Tales of the Field – In Search of Organizational Ethnography 20 years on. Organizational Research Methods. Vol. 13 (2.)    Cunliffe, A. L., and Karunanayake, G. (2013): Working within hyphen-spaces in thnographic research: implications for research identities and practice’, Organizational Research Methods, 16/3: 364-92    Davenport, T. H. & Ronanki, R. (2018): Artificial Intelligence for the Real World. Harvard Business review    Felten, E. W., Raj, M., & Seamans, R. (2018). A Method to Link Advances in Artificial Intelligence to Occupational Abilities. AEA Papers and Proceedings, 108, 54–57  Fenton, C., & Langley, A. 2011. Strategy as practice and the narrative turn. Organization Studies, 32(9): 1171–1196.    Fuchs, R., Silverstone, Y., & Thomas, R. (2016). A machine in the C-suite, Accenture Strategy. Guth, W. D., & MacMillan, I. C. (1986). Strategy  Harding, N., Lee, H., & Ford, J. (2014). Who is ‘the middle manager’? Human Relations, 67, 1213–1237.    Hrebiniak, L. G., & Joyce, W. F. (1984). Implementing strategy. New York: Macmillan.    Ketokivi, M., & Castañer, X. (2004). Strategic planning as an integrative device. Administrative Science Quarterly, 49, 337–365.    Mantere, S. (2013). What is organizational strategy? A language-based view. Journal of Management Studies,50, 1408–1426.    Mantere, S., & Vaara, E. (2008). On the problem of participation in strategy: A critical discursive perspective. Organization Science, 19, 341–358.    Manyika, J., Bughin, J., Chui, M., Silberg, J., & Gumbel, P. (2018). AI problems and promises. McKinsey. McKinsey Global Institute.    Marr, B. (2018). The AI Skills Crisis And How To Close The Gap. Forbes. Retrieved February 27, 2019, from https://www.forbes.com/sites/bernardmarr/2018/06/25/the-ai-skills-crisis-and- how-to-close-the-gap/Mokyr et al., 2015    Meyer, C. B. (2006). Destructive dynamics of middle management intervention in postmerger processes. Journal of Applied Behavioral Science, 42, 397–419.    Nonaka, W. D., & MacMillan, I. C. (1986). Strategy implementation versus middle management self-interest. Strategic Management Journal, 7, 313–327.    Nutt, P. C. (1987). Identifying and appraising how managers install strategy. Strategic Management Journal, 8, 1–14.    Raes, A. M., Heijltjes, M. G., Glunk, U., & Roe, R. A. (2011). The interface of the top management team and middle managers: A process model. Academy of Management Review, 36, 102–126.    Regnér, P. (2003). Strategy creation in the periphery: Inductive versus deductive strategy making. Journal of Management Studies, 40, 57–82.  Saphir, A. (2018). As companies embrace AI, it’s a job-seeker’s market. Reuters. Retrieved from https://www.reuters.com/article/us-usaeconomy-artificialintelligence- idUSKCN1MP10DScarcello, 2019    Silverman, D. (2014). Interpreting Qualitative Data. Methods for Analyzing Talk, Text and Interaction. London: SAGE Publications.    Spradley, J. P. (1980). Participant observation. New York, Holt Rinehart and Winston. Thomas, R., & Linstead, A. (2002). Losing the plot? Middle managers and identity.  Organization, 9, 71–93.    Tsoukas, H., & Chia, R. (2002). On organizational becoming: Rethinking organizational change.  Organization Science, 13, 567–582.    Van Maanen, J (2010): Organizational Research Methods.Volume: 13 issue: 2, page(s): 240-255    Ybema, S., Yanow, D., Wels, H., & Kamsteeg, F. H. (Eds.) (2009). Organizational ethnography: Studying the complexity of everyday life.  Zoller, H. M., & Fairhurst, G. T. (2007). Resistance leadership: The overlooked potential in critical organization and leadership studies. Human Relations, 60, 1331–1360.