Skip to main content

Data & Intelligence

Lift Analysis and IBM SPSS

Defining what lift is

“Lift” is the measure used to determine how well your targeting model does at prophesying cases as having a greater response with respect to the population as a whole.

Your model may be doing its job if the response (within the target) is better than the average response of the population as a whole. Lift is simply the ratio of these values (target response divided by average response).

A working Example

IBM SPSS Statistics “Control Package Test” demonstrates a technique used to compare marketing campaigns to see if there is a significant difference in the effectiveness of different marketing packages.

Let’s walk through this example to see if we can understand the idea of lift.

Once you open the provided sample data file, from SPSS Viewer, you can select Direct Marketing and then Choose Technique.

SPSS displays the “Direct Marketing” dialog:

 

 

 

 

 

 

 

 

 

Next, you can select “Compare Effectiveness of Campaigns”. At this point SPSS gives us the “Control Package Test” screen.

Here is an impressive display of the power of SPSS. On your left is the list of the fields in your data file. (SPSS offers the ability for you to sort or reorder these fields for clarity).

Data Intelligence - The Future of Big Data
The Future of Big Data

With some guidance, you can craft a data platform that is right for your organization’s needs and gets the most return from your data capital.

Get the Guide

Below the field list are a series of icons that let you automatically select relevant fields in the file. For example, with a single click, SPSS identifies and selects:

  • All your fields
  • Only the nominal fields
  • Only the ordinal fields
  • Only the continuous fields

On the right side of the screen, you have the ability to designate a “Campaign Field” and an “Effectiveness Response Field”.

In this simple example, we are select “Control Package” for Campaign Field,

And for our Effectiveness Response Field, we’ll select “Responded to test offer”.

Next, we select “Reply” for “Effectiveness is defined as”.

SPSS is making it easy on us as it assumes that when we selected “Compare Effectiveness of Campaigns”  we would base our results on either a reply/response or on the amount purchased (as a result of our campaign) – these are the most obvious effects of an organizations marketing campaign or promotion. So back to the example,

for a “Positive reply value”, we select “Yes” from the provided drop-down list. (1 is displayed in the text field because “Yes” is actually a value label associated with a recorded value of 1 in the data file (you can read how to do this in one of my prior blog posts).

In the bottom right of the screen, you will see a “framed area” named “Name and Label for Recorded Effectiveness Response Field”. Here, based upon what we’ve selected so far, SPSS automatically performs a “recode” of our selected response field to create a new field to perform the analysis on. It gives the new field a name and label. You can change the default name and label and provide your own if you wish.

Finally, click “RUN”.

The Results Are In!

The output supplied by SPSS Statistics shows a table that displays counts and percentages of positive and negative responses for each case defined by our campaign field and a table that indicates if the case response differs significantly from each other:

 

 

 

 

The positive response rate for the total data file (Control) is 3.8% (this is our average response rate), while the positive response rate for the test case (Test) is 6.2% (this would be out targeted response rate), so – our lift seems to be 6.2 divided by 3.8 or

“A 1.6% lift might be assumed by using the new marketing package over the existing one”.

Realistically it’s not that straight forward for there may be other aspects that you need to consider, such as any additional costs associated with the new promotion or campaign.

Cumulative Gains and Lift Charts

This is a very simple example and the resulting table provided by SPPS is straight forward. As part of an actual lift analysis, you might want to create a Cumulative Gain and Lift Chart to better visualize the outcomes generated. A Cumulative Gain chart and Lift chart graphically represent of the advantage of using a predictive model (in our example to determine the effectiveness of a marketing campaign).  Both charts consist of a lift curve and a baseline, generally the greater the lift, the better.

IBM SPSS provides the ability to easily generate both a Cumulative Gains and List chart as part of your analysis output.

Next Time

For next time, I will narrate the results of my thorough exploration of the process of generating these charts.

Until next time…

General Barnicke: Are you telling me that you men finished your training on your own?
John Winger: That’s the fact, Jack.

 

 

 

 

 

Leave a Reply

Your email address will not be published. Required fields are marked *

This site uses Akismet to reduce spam. Learn how your comment data is processed.

Jim Miller

Mr. Miller is an IBM certified and accomplished Senior Project Leader and Application/System Architect-Developer with over 30 years of extensive applications and system design and development experience. His current role is National FPM Practice Leader. His experience includes BI, Web architecture & design, systems analysis, GUI design and testing, Database modeling and systems analysis, design, and development of Client/Server, Web and Mainframe applications and systems utilizing: Applix TM1 (including TM1 rules, TI, TM1Web and Planning Manager), dynaSight - ArcPlan, ASP, DHTML, XML, IIS, MS Visual Basic and VBA, Visual Studio, PERL, Websuite, MS SQL Server, ORACLE, SYBASE SQL Server, etc. His Responsibilities have included all aspects of Windows and SQL solution development and design including: analysis; GUI (and Web site) design; data modeling; table, screen/form and script development; SQL (and remote stored procedures and triggers) development and testing; test preparation and management and training of programming staff. Other experience includes development of ETL infrastructure such as data transfer automation between mainframe (DB2, Lawson, Great Plains, etc.) systems and client/server SQL server and Web based applications and integration of enterprise applications and data sources. In addition, Mr. Miller has acted as Internet Applications Development Manager responsible for the design, development, QA and delivery of multiple Web Sites including online trading applications, warehouse process control and scheduling systems and administrative and control applications. Mr. Miller also was responsible for the design, development and administration of a Web based financial reporting system for a 450 million dollar organization, reporting directly to the CFO and his executive team. Mr. Miller has also been responsible for managing and directing multiple resources in various management roles including project and team leader, lead developer and applications development director. Specialties Include: Cognos/TM1 Design and Development, Cognos Planning, IBM SPSS and Modeler, OLAP, Visual Basic, SQL Server, Forecasting and Planning; International Application Development, Business Intelligence, Project Development. IBM Certified Developer - Cognos TM1 (perfect score 100% on exam) IBM Certified Business Analyst - Cognos TM1

More from this Author

Follow Us