Charting the Path to Purchase Using Topic Models

Abstract

In gathering information for an intended purchase decision, consumers submit search phrases to online search engines. These search phrases directly express the consumers’ needs in their own words and thus provide valuable information to marketing managers. Interpreting consumers’ search phrases renders a better understanding of their purchase intentions, which is critical for marketing success. In this article, the authors develop an integrated model to connect the latent topics embedded in consumers’ search phrases to their website visits and purchase decisions. Using a unique data set containing more than 8,000 search phrases submitted by consumers, the model identifies latent topics underlying the searches that led consumers to the firm’s website. Compared with a model lacking any textual information from consumers’ search phrases, a model using textual data in a heuristic approach, and a model based on the latent Dirichlet allocation, the proposed model provides a better evaluation of a consumer’s position on the path to purchase and achieves much better predictive accuracy, which could in turn substantially increase the firm’s revenue. The authors also extend the discussion to aggregators, affiliated websites, and segments of consumers who are exposed to the firm’s outbound ads. Marketing managers can use this method to extract structured information from consumers’ search phrases to facilitate their inference of consumers’ latent purchase states and thereby improve marketing efficiency.

Converting free users to paid subscribers in the SaaS context: The impact of marketing touchpoints, message content, and usage

“…In search engine marketing, researchers have explored the characteristics of search keywords, including brand names, retailer names, geo‐locations, the popularity of keywords, and the specificity of keywords (Ghose & Yang, 2009; Li et al., 2016; Rutz & Bucklin, 2011). Recent applications of machine learning techniques extract latent semantic meanings from search phrases and evaluate the keyword ambiguity (Gong et al., 2018), consumers’ content preference (Liu & Toubia, 2018), and the impact of latent themes on purchase propensity (Li & Ma, 2020).…”

Wisdom from words: The psychology of consumer language

“…Tirunillai and Tellis (2014), for example, used topic modeling to explore dimensions linked to quality, how they shift over time, and how that relates to brand positioning. Berger and Packard et al (2018) used the approach to identify the main themes in song lyrics, and Li and Ma (2020) illustrate how marketers can use topic modeling and consumer search data to identify where consumers are in the decision‐making journey. Topic modeling can also be used as control variables, identifying different topics that might arise in things like customer service conversations (Packard & Berger, 2021) or news articles (Berger, Moe, & Schweidel, 2022).…”

“Inside” versus “outside” trends in consumer research

“…It is noteworthy that in addition to identifying topics studied in the academic literature, researchers have applied the methodology of topic models to a diverse range of data sources for consumer research, including online product search terms (Li & Ma, 2020) and social media content (Zhong & Schweidel, 2020). With the digitization of information making a wealth of consumer textual data available for consumer research, such as online customer reviews, open‐ended responses, and customer communications (Berger et al, 2020), topic models are relevant to a wide range of consumer research interests and applications.…”