Posted by: zyxo | April 20, 2015

Datamining for marketing: a real simple explanation of LIFT


In using dataminig models for marketing campaigns, a definition of lift could be:

lift is a measure of the performance of a targeting model (association rule) at predicting or classifying cases as having an enhanced response (with respect to the population as a whole), measured against a random choice targeting model.” (wikipedia)

Yeah, wright!

Now in plain language for a normal human being:

“lift” means:   “how much times is my model better than a random selection”.

Example:

Divide you customer database at random in 3 equal parts: part I , part II ,  and part III

– when you use no model at all:  take a random selection A of 10,000 persons out of part I. Send them an email in an effort to sell something.  Let’s assume that 2% of them bought you product.  Result (no model)=2%

– use your first datamining model (model M1) to select 10,000 persons out of part II. Send them the same email.  Let’s assume that 6% of them bought your product. Result (M1)=6%

– use another datamining model (model M2) to select 10,000 persons out of part III. Send them the same email.  Let’s assume that 10% of them bought your product. Result (M2)=10%

It is straightforward to see that model M1 does 3 times better than without a model and model M2 does 5 times better.

Formula:                      lift=  ( result of Model X) divided by (result of no model at all)

Hence:

Lift of first model (M1)        =>      6% / 2% = 3
Lift of second model (M2)  =>    10% / 2% = 5

_________________________________________________________________________________________________

Associated posts:

Datamining: what is a good lift?

Put a crowbar under your marketing campaign!

Datamining for marketing campaigns: interpretation of lift.


Leave a Reply

Fill in your details below or click an icon to log in:

WordPress.com Logo

You are commenting using your WordPress.com account. Log Out / Change )

Twitter picture

You are commenting using your Twitter account. Log Out / Change )

Facebook photo

You are commenting using your Facebook account. Log Out / Change )

Google+ photo

You are commenting using your Google+ account. Log Out / Change )

Connecting to %s

Categories

%d bloggers like this: