Multichannel Sales Attribution and Media Optimization

Sandy D. Jap and Timothy J. Gilbride, 2016, 16-116

Winner of 2017 Top Download Award

Most firms use multiple channel formats or routes-to-market in order to reach their end customers: Internet channels, corporate stores, franchisees, indirect retailers, and more. A multichannel strategy faces two chief challenges: (1) What is the role or contribution of each individual channel to the overall system, e.g., is it synergistic or cannibalizing? (2) What can firms do in the short run to optimize the various roles and outcomes of the channel formats to maximize the system performance?

In this report, Sandy Jap and Timothy Gilbride propose that channel strategies need to move toward a system-wide perspective that accommodates cross-channel complementarities in both sales and media channels. Using a proprietary dataset of 78 weeks of durable product sales, they provide a decision support system to assist firms in optimizing their media allocation spend to account for these effects.

Model and findings

Jap and Gilbride develop a multichannel sales and media model using aggregate market-level data for a durable good product, analogous to classic advertising effectiveness models. However, they use an aggregate logit model that recognizes that in the current time period, channel sales are mutually exclusive, but sales may come from competitors or new customers entering the market. They incorporate these dynamic effects of advertising and channel sales as well as the many interaction effects between them. Their empirical analysis of nine sales markets suggests that accounting for such cross-channel and multimedia effects can lead to increases in profits of 35% on average.

Managerial implications

This research offers a valuable tool to help a firm determine the unique, incremental contribution of each channel member to system sales, including its own and cross-channel carryover or lift effects. Further, given that changing channel strategy is difficult to achieve in the short term, they focus on reallocating media mix spend, offering a decision support tool to help a firm better maximize each channel’s sales and system contribution.

Their approach creates value for firms, enabling them to better compensate and allocate resources to channel members and formats accordingly, and helps wring greater value from each dollar spent on communication efforts. The modeling framework can be implemented with minimal data requirements (sales by channel by week, channel profitability, and media mix spend and costs) and is relevant to firms who have multiple sales channels and purchase advertising in several media. In firms where the actual channel marketing cost data are available, the model’s predictive power and optimization insights would be sharpened.

Sandy Jap is Professor of Marketing at the Goizueta Business School, Emory University. Timothy J. Gilbride is the Notre Dame Associate Professor of Marketing at the Mendoza School of Business, University of Notre Dame. 

The authors are grateful to the participating firm for data access.

Related links

Orchestrating Marketing in a B2B Environment
Stefan Todd Sleep, University of Georgia, and Myoung-Jin Chae, Georgia Institute of Technology (2014) [Conference Report]

Conditions for Owned, Paid, and Earned Media Impact and Synergy
Ceren Demirci, Koen Pauwels, Shuba Srinivasan, and Gokhan Yildirim (2014) [Report]


  • Corporate: FREE
  • Academic: FREE
  • Subscribers: FREE
  • Public: $0.00



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