How can you achieve the highest possible impact with a campaign? How should your marketing budget be distributed across different channels over a planned period of time? These are the challenges marketing managers and media planners regularly face in order to achieve their objectives such as brand building, attracting attention to specific (advertising) content and sales.
The underlying question is: how can the effect of marketing be estimated and directly integrated into campaign planning? Media mix modeling or sales modeling offer a solution.
Media mix modeling uses historical data on the desired target indicator and possible influential factors. The model can include data pertaining to your brand, a comparable brand or different competitive brands. First, the time course of the available data is viewed descriptively with the aim of illustrating the relationship between media and the target indicator. A distinction must be made, for example, as to whether the brand in question is a traditional brand or a new brand that is being developed. The relevant influential factors are defined on this basis. It is important that long-term trends and seasonal influences are also taken into account. While media mix modeling shows the connection between media performance such as number of contacts, gross rating point, impressions, budget and brand development, e.g. advertising and brand awareness, sales modeling is used to illustrate the influence of factors such as price, promotion, distribution and media performance on sales.
The influence of the various factors on the target figure is quantified using a statistical estimation method. Descriptive factors can be used to explain the course over the period under consideration (Figure 1). This makes it possible to evaluate how certain media performance affects aspects such as advertising perception of a brand or sales. It is also possible to estimate how strong performance is increasing for a given media. This estimate of media effectiveness provides the basis for budget allocation to different media channels with the highest possible advertising impact or derivation of advertising pressure, i.e. distribution of media performance over the weeks of a campaign. Modeling is an analysis method that not only generates insight into the effectiveness of media but also provides valuable support in campaign planning processes.
At MMT, our data analytics team regularly creates models as a starting point for media planning for various clients. Based on the effective strength of the channels involved, our team first develops different scenarios for budget allocation. The channels then are combined in such a way that the resulting allocation maximizes the effectiveness of the campaign for a given budget (Figure 2).
The next step is to plan advertising pressure across the campaign period. The model is used to forecast the effect of the media performance used per week on the target variable. This makes it possible to identify which allocation over the weeks of the campaign will achieve the desired advertising pressure (Figures 3 & 4).