Refreshing the Concept of Share of Voice in Media Effectiveness


At Middlegame, we usually work with direct shopper response levers like assortment, pricing, and merchandising. It is probably at our insistence that buyers shop by SKU which makes it our go-to level of analysis. However, we sometimes get asked to provide a detailed assessment of incrementality versus transferred demand for the media investment.

When describing our approach on the topic of share of voice, our modeling intrigued one of our new clients.  The analytic firms he had encountered focused on actual impressions or rating points that generated wear-out curves. Since the Middlegame approach is simulation based, I explained that we could create those curves with our scenario planning engine. However, the curves didn’t need to influence the model inputs. In addition, a share of voice approach adds a distinct layer on top of the traditional assessment.

We start with acceptance that the impact of advertising persists well beyond the period it is activated. The concept of advertising persistence is an old one, but has gained a lot of momentum in the last decade due to multichannel attribution. Tools like Vector Auto-Regressive (VAR) are regularly used as a way to capture these impacts for marketing mix modeling. Unfortunately, VAR does not provide parameter estimates that can be interpreted directly in a dynamic resource allocation framework such as marketing mix modeling. (But that is another topic left for our friend Dr. Peter Cain at Market Science Consulting”. Regardless, measuring advertising persistence remains of key relevance to ROMI assessment.

Simon Broadbent is the godfather of longer-term advertising assessment. Before he passed away, he argued that the persistence of advertising must be assessed in the context of competitive interaction … the “CI” in the Middlegame CIA® platform. For Broadbent, the contribution of medium-term effects leveraging his ad-stock methodology, expressed as the impact of our brand less the impact of competitors, was the true longer-term persistence. Substituting the “less” in “impact of our brand less the impact of competitors” with “relative” and you are led to the discussion on share of voice.

Gascoingne and Ward (2001) further develop the concept persistence through the use of a Cumulative Long Term Share of Voice. They stated that brands and their advertising have a form of inertia by looking at the cumulative change in the share of voice across all brands in a competing category. Although Middlegame focuses on SKUs, we do so for all SKUs in the category and hence we provide the opportunity to address all brands in the category and create a longer-term assessment of advertising.

I mentioned our friend Dr. Peter Cain from Market Science Consulting. One of the reasons we tend to shy away from media assessment is because we like to refer clients to Pete for advanced econometric modelling. He is by far one of the most noted experts in applying time series analysis to marketing. Please check out his company at

Reference: Gascoingne and Ward (2001) “The Long and Short of it … Advertising Effectiveness”, presented at the 47th ARF Annual Convention and Research Infoplex.

Middlegame is the only ROMI consultancy of its kind that offers a holistic view of the implications of resource allocation and investment in the marketplace. Our approach to scenario-planning differs from other marketing analytics providers by addressing the anticipated outcome for every SKU (your portfolio and your competitors’) in every channel. Similar to the pieces in chess, each stakeholder can now evaluate the trade-offs of potential choices and collectively apply them to create win-win results.

Contact us at info@