Benchmark Determination...

#1
Hello, folks, first post.

Suppose I have a set of 1000 elements (customers), each with a corresponding value (net profit gained from that customer), ranked in descending order. My goal is to establish a benchmark that allows me to conclude, 'I should be focusing my marketing efforts primarily toward the top n percent of customers.' How would I go about establishing such a benchmark?

Thank you for any assistance that you can provide!

bubbis
 

JohnM

TS Contributor
#2
Try applying the "Pareto Principle" or "80/20 rule" where the top 20% of your customers will comprise 80% of your profits (or thereabouts...)

Create a second column that represents the cumulative profits for a particular customer and the customers above him/her.

Here's an example showing a Pareto bar chart for individual items, combined with a line chart that shows the cumulative percentage:

http://www.isixsigma.com/library/content/c010527a.asp
 
#3
Thanks for the response, John! :)

Typically I would conduct the exact same analysis that you suggested, but was just wondering if there was some other way of doing it that I may have forgotten about. Analyses containing <50 data elements I'm comfortable with using Pareto along with (relative) cumulative frequencies, etc; but with 1000 data elements I was hoping for some sort of tool that would tell me, 'hey, buddy, for this set of data, you need to focus on the top 46% of your customers.'

I'm degreed in mathematics, but I was just curious if I was forgetting some sort of special tool that would assist me with benchmarking.

Thanks again, John!

Regards,

bubbis