
| Marketing Mix Modelling Understanding what works, and what doesn't |
|||
Unpicking the effects of all your different marketing activities – truly understanding your marketing ‘mix’ – usually requires specialist analysis. Marketing mix modelling, using econometrics, is the best-established methodology of doing this. It is also sometimes frighteningly ‘statistical’. As a result, it can appear a ‘black box’ process, somewhat divorced from the real world, with results which may not make intuitive sense. At Billetts we believe part of our role is to demystify the process in giving you clear insight and accurate measurement. Econometrics explained Econometrics literally means 'economic measurement'. It is a combination of economics and statistics. The most important method in econometrics is ‘time-series multiple regression’. Regression analysis is used to ‘model’ relationships between variables and determine the magnitude of those relationships. It is often described as like trying to unbake a cake – to work out the ingredients and the recipe after the cake has been baked. As the name suggests, time-series analysis examines variables over time, such as the effect of media investments on sales. The most important feature of a time-series regression approach is that it allows us to model all the many influences on sales, or indeed any response metric, simultaneously. However, in order to make sure that we will deliver analysis and insight which makes sense, we always start with a complete review of your data. This data ‘landscaping’ stage is an essential first step towards a deeper knowledge of your business and gives you added confidence in the process. Answering your key questions Billetts use advanced econometric techniques to answer the kinds of questions marketers ask:
Keeping it simple Our strategy is to:
|
|||
Contact Andrew Challier: andrew.challier@billetts.com or call +44 (0) 20 7650 9600 to find out more |
|||