I want to combine multiple data sets of customer spend data together. I want to calculate how the bounds of the data decrease as I add in more data sets (so I can do things like estimate the ROI error for a whole year, as opposed to one week's spend).
I believe there is a Bayesian method of doing this but I cannot find a simple explanation of the method.

1) Can someone point me to some articles please?
2) Do you know if this works over multiple data sets if I just keep adding more and more on top?