All of our models are bespoke to each client and built on historic exposure to identify the pure effect of each potential factor on enquiries, sales and revenue. These models are combined and calibrated with a decision support system which can be used to plan optimal investment scenarios.
The technique we use allows the isolation of every factor which may affect performance. We enable the understanding of how much exposure is required to deliver an incremental response (whatever that might be), and also where incremental effects cease beyond a certain level of exposure and investment is effectively wasted.
Pure effect analysis also allows a detailed understanding of how competitor exposure and uncontrollable market or environmental factors affect your performance. Planning around factors which can be predicted but not controlled (competitor spend, interest rate changes or even rainfall) ensures that control is retained on sales generation as well as optimal spend efficiency.
By understanding the strength of a brand it is possible to estimate the sales which are generated from its residual strength. We call this brand momentum and can build simple models to identify baseline sales which are derived from it.
Whilst mediaqube enables a forensic appreciation of the effect of promotions on response and sales, this additional analysis helps to identify two additional effects which are critical in understanding performance overall. Brand momentum analysis identifies the rate at which brands decay in the absence of ongoing promotional exposure. Additionally, estimates can be made on the extent of activity which is required to maintain current ratings.
Brand momentum analysis is based on far less data, usually taking brand tracking at a ‘brand consideration' level, and is consequently far more high level than a mediaqube assignment.
However, brand momentum analysis can provide a more rounded understanding of the balance between short term and longer term effects of exposure and branding. It can also facilitate a clearer understanding of the incremental sales being generated from promotions, which can help significantly in developing business cases and cost justifications for ongoing investment.
The importance of modelling at a postcode level
We consider it critical to model at a very granular geographic level to be able to gain enough variation to accurately incorporate the exposures generated by each type of media - across broadcast, direct and digital channels. As an example, if only certain postcodes are direct mailed in a campaign, then we would only expect to observe a lift in sales from DM in these postcodes. Building models at a state or national level will typically fail to identify the impact of these highly targeted activities.
Similarly, the media footprints illustrated below show that even above the line media channels such as TV, press, online, radio etc each generate very different exposures in individual postcodes depending on the show, publication, website etc, meaning it is essential to understand their reach in order to accurately measure their effect.
Consider the following 1 week TV schedule in Sydney, totalling 250 TARPs.
By using the footprinting methodology to allocate the correct weighting of TARPs to each postcode for each individual spot, then summarising at the postcode level, the TARP level in each postcode can be estimated. While overall the 1 week schedule had 250 TARPs, very few postcodes actually received this figure as demonstrated below.
It is also important to consider the wide range of other non-media factors which can have a significant impact on the level of responses in any individual postcode, and hence on our ability to accurately measure media effects. These include:
This granular geographic approach enables us to generate sufficient data points to accurately determine the impact of each type of media. This ultimately enables us to predict new sales volumes at a postcode level and where appropriate overlay measures of customer value which can also vary dramatically by postcode.
TOP