Integrating Social Media Metrics

Wendy W. Moe and David A. Schweidel, 2015, 15-113

User-generated content on social can be a valuable source of insights for marketing researchers, but one of the biggest challenges is gauging social media metrics against appropriate benchmarks. Social media generates a plethora of new metrics which marketers struggle to interpret. Because each brand has a unique personality and target customer, comparing metrics across brands, even in the same industry, may not provide the appropriate benchmarks.

In this report, Wendy Moe and David Schweidel develop an integrated modeling framework for multiple social media metrics. Specifically, they develop a joint model for a variety of commonly monitored metrics. Drawing on the intuition that underlies perceptual maps, their model assumes that an underlying latent space characterizes each brand’s social media presence (relative to its competitors) and generates each metric.

They apply the model to Twitter data associated with the top 20 quick service restaurants (QSR) over a six-month time period. They jointly model (1) the number of Twitter comments that mention the target brand, (2) the sentiment expressed in each comment mentioning the target brand (i.e., positive, neutral, or negative), and (3) the number of comments that mention the target brand and simultaneously mention another top 20 QSR brand as a function of the brand’s position in a latent perceptual space. The model fit the data well both in-sample and out-of-sample.

Moe and Schweidel illustrate how the proposed model can be used to establish appropriate benchmark metrics, given each brand’s overall social media presence relative to others in the industry. Based on the brand’s position in a latent perceptual space (as inferred from the voice of the consumer provided by social media), a marketer can derive expected social media metrics and monitor for deviations from these expectations. These expected metrics provide reasonable benchmarks for brands monitoring social media presence.

Further, they examine how a brand’s position in the social media-inferred latent space relative to other brands in the industry can be linked to financial performance. They propose a social-media-based differentiation metric that characterizes how each brand’s social media presence compares to that of others, and show that resulting latent space is a significant predictor of sales revenues.

Wendy W. Moe is a Professor of Marketing at the Robert H. Smith School of Business, University of Maryland. David A. Schweidel is Associate Professor of Marketing at the Goizueta Business School, Emory University.

Related links

Social Media Intelligence: Measuring Brand Sentiment from Online Conversations
David A. Schweidel, Wendy W. Moe, and Chris Boudreaux (2012) [Report]


  • Corporate: FREE
  • Academic: FREE
  • Subscribers: FREE
  • Public: $18.00



Employees of MSI Member Companies enjoy the benefits of complete online access to content, member conferences and networking with the MSI community.



Qualified academics benefit from a relationship with MSI through access to, conferences and research opportunities.



The public is invited to enjoy partial access to content, a free e-newsletter, selected reports and more.




Learn more about becoming a member of the institute

Read More

Stay Informed

The MSI Mailing List

Subscribe to our email list to stay informed about upcoming events, news, etc.

Social Media