December 20, 2008

Prove It: Matchback Attribution

One of our loyal readers is trying to goad me into ripping a recent article on Multichannel Merchant. I won't link to the specific article, because 95% of the content the author writes about will increase your profits, and I don't want for this to be a criticism of the author, who is simply trying to give advice that will help you.

But I do want for us to consider one of the quotes, the exact quote sent to me by a loyal reader.
  • "Correlation analysis suggests that as much as 90% of the unallocated orders that come to the Web for the multichannel marketer are directly related to catalog mailings".
We need to consider three phrases in the sentence.
  1. Correlation analysis suggests (this means we cannot prove the findings that follow).
  2. As much as (this means the real number is always lower than what we will be quoted).
  3. Directly related (this means we cannot prove causation).
Based on the linkage of three phrases, an entire industry moves forward with an agenda, an agenda based on a linkage of assumptions that may or may not be correct.

Do you realize the leap of faith required to believe in this sentence?

We listen to quotes like this because we want to believe the sentence.

Now it is entirely possible that those who adhere to the tenants of the sentence quoted above are right, and will experience unbridled success. I grant you this point.

But are you willing to grant me my point ... what if one, two, or three components of the sentence are wrong? What does that mean to the success of the businesses we manage?

Matchback attribution. Prove it!!

No comments:

Post a Comment

Note: Only a member of this blog may post a comment.

Have You Ever Looked At Performance This Way?

I bring this up annually. But the analysis is so darn much fun! Take every customer who purchased between Black Friday and Cyber Monday last...