I am a CEO

In my previous post about macro-economics I promised to write how to analyse buying behaviour and weather of your customers. Today I am going to use a simple mechanism called Pearson Coefficient which is the measure of linear dependence between two variables X and Y. The correlation values are between +1 to -1 both inclusive. I am going to explain how temperature affect your product sales. I have considered two variables one is temperature another one is the product sales.

The tool we can use here are excel, correlation coefficient, graphs etc.

Please download the  data dump from Google analytics and temperature information from metgov or any other source to an excel. I usually monitor and log temperature in the database.   Go to data tab and click on the data menu on the top right of excel. Click on the image to see it in full.

If you are doing offline business gather the data which has product sales and corresponding temperature information.

Now select the correlation from the data analysis window.

The next step is to select the data range and output range.

Now you will get the correlation of different variables as the output.

Now the main part is to understand the value and infer from the correlation coefficient. The correlation coefficient ranges from −1 to 1. A value of 1 implies that a linear equation describes the relationship between X and Y perfectly, with all data points lying on a line for which Y increases as X increases. A value of −1 implies that all data points lie on a line for which Y decreases as X increases. A value of 0 implies that there is no linear correlation between the variables.

Now what do you think about our data?

There is a slight negative correlation between temperature and Product A and B sales; however, the relation between temperature and Product C is positive. This means that as temperature increases the product sales decreases for Product A & B but increases for Product C. The linear relation is however weak in both cases.

There is a strong positive linear dependency between sale of Product A and Product B; however, this inference should be investigate further to understand the other factors that affects the buying.

Most of the managers are quite comfortable seeing the output in a graph to understand different relationships. The best way to do this is to plot scatter diagram in excel.

Change the scatter diagram with layout slope and function that represent the linear relationship line.

Once you select the layout with line and equation of line you will get your graph something similar to below.

I hope you understood the post, please share it with your friends via twitter and Facebook. I will be writing about advanced segmentation analysis in Google analytics in my next post. Until then keep tuning to iamaceo and share your valuable feedback to me.

Tags: analysis, behaviour analysis, correlation, customer buying, data analysis, excel data analysis, Pearson Coefficient, weather and customer buying