Improving competitive analysis with temporal text visualization

Purpose – The purpose of this paper is to develop a visualization method for texts produced by a company's competitors, partners, or customers. This method can be used for competitive intelligence purposes, and in particular, for spotting changes over time in a company's communication. Design/methodology/approach – Based on the judgments of expert readers, topics considered to be important in the quarterly reports of two telecommunication companies were turned into so‐called collocational networks, using a statistical method originating in linguistics. Findings – The paper shows that when collocational topic networks are produced out of a sequence of quarterly reports, they provide a visualization of how the presentation of these topics changes from one‐quarter to the next. Research limitations/implications – The statistical method used in this paper does not handle low‐frequency topics, that nevertheless might be of great importance, very well. The method could be developed further so that users' judgments of importance are given more weight. Practical implications – The method developed in this paper can be turned into a data visualization tool for intelligence practitioners. Originality/value – Textual data are often overlooked in competitive intelligence, as it is more difficult to visualize and present than quantitative data, such as financial ratios. The method presented here is an easy to grasp “white box” approach to visualizing textual data, as it combines the judgments of subject matter experts with a statistical method.