Quantifying the causal effects of the marketing mix is the marketing 'holy grail'. Marketing mix modellers amazingly claim to be able to do this with multivariate statistics, usually from a single set of data. This magic trick would be great if it worked, but there are fundamental reasons why the model will most probably be very wrong, and you can never tell how wrong it is - until you try to use it to predict the future impact of changes. Read More...

In order to understand the DVR effect we need a realistic picture of how much avoidance occurs already in a non-DVR environment. It could very well be that DVR facilitated ad skipping merely replaces other forms of passive and active ad avoidance - in which case DVR impact will be negligible. Read More...