Between automation and interpretation: Using data visualization in social media analytics companies

This chapter explores the use of data visualizations in social media analytics companies. Drawing on a dataset of ethnographic field notes and thematic interviews in four Finnish social media analytics companies, we argue that data visualizations are crucially involved in how analytics-based knowledge claims become accepted by companies and their clients. Basing on previous research on visualizations in organizations and as a representational practice, we explore their role in social media analytics. We identify three practices of using visualizations, which we named simple-boxing, flatter-boxing and pretty boxing. We argue that these practices enable analysts to achieve the simultaneous aims of producing credible and valuable analytics in a context marked by high business promises.