He started by putting the whole aspect of measuring in to the total marketing/strategy picture by discussing the importance of an good and integrated marketing foundation:
1) How important it is to make sure that your marketing is connected to your business strategy – and how many marketing departments that misses this point!
2) Having clear guidelines for what you are trying to achieve trough your marketing. He had examples of companies that presented the objectives of successful marketing campaigns, but the objectives presented was not the same as they stated before the campaign, but other objectives the campaign achieved.
3) Making sure that you has the guidelines for measuring ready before you start the campaign.
His point being that having clear and precise objectives defined before the campaign is launched give a better learning experience. They should be:
- Time specific
- Target group specific
And the requirements for cause/effect relationships is:
- Significant changes
- Cause before effect
- Excluding alternative cause variables
And this of course was the perfect intro to my presentation of econometrics measuring. Econometric measuring is the only way of excluding the other variables and give you the best picture of how you’re marketing effect can be measured.
And one of the most important elements of a good econometric model is to have clear and specific objectives about what you are trying to obtain.
As a simple example lets use the old sales funnel. Trying to explain what variables affects leads vs. prospects will give you different variables that will be tested in a econometric model. Leads (the first stage in your funnel) will have more variables based on your general marketing activity in the market, while doing a model on prospects will have more direct variables, since this is a group that’s already in your CRM system.
The two biggest obstacles, from the customer’s view, when it comes to econometric modelling is:
- They don’t believe in the modelling and they don’t understand how it works. Even though this is a straightforward process with a regression formula, many people do not understand it. The focus is not on whether they understand regression, but their believe that a regression analysis can predict which variables effects their marketing. And there is a group of scientists, The Austrian School of Economics, which agrees with this and says that doing econometric analysis on humans is useless since the human mind is so complex. At the moment this is a small group compared to those scientists that believe that econometric modelling can help you analyse human behavior.
- The other reason is the unwillingness to use media and marketing money on analytics. Which is strange because you do this to get more insight on what works and therefor you will be able to do more with less when you incorporate the learning you get from the modeling. From my own experience we had a factor of 3 to 1 (spending 1 and getting 3 back) by using the model.
You can read my previous postings on econometric modelling here: