It goes without saying (even though that is exactly what I am doing) that an analytic approach to marketing initiatives is absolutely crucial. On that note, the true job of a marketer is to drive sales, period. Again…without saying.
The images above are snapshots of a tool I created, along with my good friend and colleague Dave, to do just that; take an analytic approach to driving sales…and be able to predict the effects. For you nerdy types, the correlation coefficient of the two data sets is 78%. That’s pretty good.
This analysis stemmed from a need for additional product support in key areas. To justify the product families, and ultimately down to the sku level, that would receive the benefit of short and medium term campaigns, I looked at sales data and web analytics for a correlation to see if there was a predictor that could be identified, and therefore the specific justification for a digital push would be self evident.
- Two years sales data were analyzed as a start point.
- 27 months web analytics were overlaid so that I could capture a lead up to the first sales data time period.
- Some data analysis and smoothing was applied.
- And…a correlation coefficient was calculated to determine the accuracy of the model.
- For the data sets utilized, two years, the coefficient was 78%. That means with about 80% certainty, a change in one data set will be reflected in the other.
I like those odds!
*NOTE: ALL DATA HAS BEEN RANDOMIZED AND SANITIZED FOR CLIENT PRIVACY
Wayne Dayberry 