Learn how to get insights from your customer data, understand your customers deeply and target the right customers with the right products!
The SPSS program offers a comprehensive customer analytics tool – the Direct Marketing module. With this tool you can conduct powerful analyses without being an expert in statistics and data analysis.
The everyday interactions with your customer generates a high amount of valuable data. The customer marketing analysis is the best solution to transform these data into real knowledge. The goal of this analysis is to get you a precise view of your customers, identify the most profitable groups of customers and send them the most appropriate marketing messages.
The Direct Marketing toolkit in SPSS includes six practical analysis procedures. Each of these procedures has its own section in this course.
- The RFM analysis allows you to classify your customers according to the recency, frequency, and monetary value of their purchases. You can pinpoint your most valuable customers (those who buy often and spend much money), as well as adapt your strategy for each RFM customers (e.g. encourage new customers to buy more, reward good customers with discounts and prizes, re-gain old customers that stopped buying from you etc.)
- The cluster analysis procedure helps you segment your customers or prospects using their most relevant demographic, economic or behavioral characteristics. In each cluster you will find customers that are similar with eah other and different to the others. You can combine this procedure with other analyses, to identify the segments with the highest RFM values, for example, or to estimate the buying probability in each segment.
- The customer profiling technique helps you detect the customer groups with the highest response rate, based on the results of previous campaign. This way you can know in advance which customers are more likely to respond to your future offers. In consequence, you can significantly improve the targeting of your future campaigns, reduce campaign costs and increase sales and ROI.
- Another procedure allows you to identify the responses to your campaign by postal codes. This is extremely useful for direct mailing campaigns, because you can find out the geographical areas where most of your customers live. You can compare the response rate of each geographical zone to your target rate and decide where to send your future mailing packages so you can maximize your profits.
- The Direct Marketing module in SPSS also helps you estimate the probability of purchase for each contact in your list, using an advanced prediction analysis method (binomial regression). You can send your future messages only to the prospects who are most likely to buy from you and remove the inactive prospects from your list. Moreover, you can predict the probability of purchasing for new customers, those freshly added to your list.
- The Control Package Test method allows you to compare the effectiveness of two or more marketing campaigns. This is useful especially when you intend to test existing campaigns against new campaigns. The differences between the campaigns response rates are evaluated using the binomial test.
Most of the procedures above use sophisticated statistical analysis techniques to process your data. However, you don’t have to be a statistician in order to use them. You can get the results you need with a few clicks only, in a few seconds. This is what you will learn in this course.
Every procedure is explained live in SPSS, and the output is interpreted in detail. At the end of each section you can find a couple of practical exercises to strengthen your knowledge.
Join this course today and you will be able to analyze your customer data using state-of-the-art predictive techniques and make informed decisions!
Segmenting Customers
-
2Introduction to RFM
Just in case you're not familiar with the RFM concept in marketing, I explain here this powerful segmentation technique.
-
3Independent RFM
What is the independent RFM analysis method - where the recency, frequency and monetary scores are independent of each other.
-
4Nested RFM
What is the nested RFM method - where the recency, frequency and monetary scores depend on each other. The advantages and disadvantages of this method.
-
5Our Example Files
In this lecture I describe in detail the data sets we are going to use for our practical examples of RFM analysis.
-
6Executing the Independent RFM Analysis When Data Are Customers
How to perform the independent RFM analysis procedure in SPSS when our data represent unique customers.
-
7Interpreting the Independent RFM Analysis When Data Are Customers
Here I explain thoroughly how to interpret the output of an independent RFM analysis when our data represent unique customers.
-
8Performing the Nested RFM Analysis When Data Are Customers
How to execute the nested RFM analysis when our data represent unique customers and how to interpret the output.
-
9Executing the Independent RFM Analysis When Data Are Transactions
How to execute the independent RFM analysis when our data represent unique transactions.
-
10Interpreting the Independent RFM Analysis When Data Are Transactions
How to interpret the results of an independent RFM analysis when our data represent unique transactions.
-
11Performing the Nested RFM Analysis When Data Are Transactions
How to run the nested RFM analysis when our data represent unique transactions, and how to interpret the output.
-
12Practical Exercises
Here you can find the practical exercises for the RFM analysis.
Generating Customer Profiles
-
13Two-Step Cluster Technique - Introduction
In this lecture we present the basics of the two-step cluster method, that is used by the Direct Marketing module to segment your customers or contacts into homogeneous classes.
-
14Performing the Cluster Analysis (1)
How to execute the two-step cluster method in the Direct Marketing module and how to interpret the main output.
-
15Performing the Cluster Analysis (2)
More details about the output of the two-step cluster analysis.
-
16Practical Exercises
Here you can find the practical exercises for the cluster analysis
Identifying the Top Responding Postal Codes
-
17Executing the Procedure
How to run the procedure that generates prospect profiles (based on the most important prospect characteristics) and computes the response rate for each profile.
-
18Interpreting the Results
How to identify your best profiles based on the output provided by the SPSS program.
-
19Practical Exercises
Here you can find the practical exercises for the profile generating technique
Estimating Buying Probabilities
-
20Running the Procedure
How to execute the procedure that helps identifying the postal codes with the highest response rate for your next direct mailing campaign.
-
21Interpreting the Results
The SPSS program computes a number of indicators for each postal code. We'll learn how to interpret them and how to select the postla codes with the best customers.
-
22Setting a Maximum Number of Contacts
How to run a postal code analysis when we have to restrain the number of contacts we send the message to (it can happen frequently, for budget reasons for instance).
-
23Practical Exercises
Here you can find the practical exercises for the postal codes analysis
Comparing Campaign Effectiveness
-
24Running the Procedure
How to execute the procedure that estimates the buying (or responding) probability for each contact in our list - and how to save our model for later use.
-
25Interpreting the Output
How to select the best contacts based on their buying probability (or buying propensity).
-
26Validating Our Model
In this lecture we divide our sample of contacts in two sets: training set and test set. Next, we create the model in the training set and validate it in the test set (in other words, use it to estimate the buying probabilities of the contacts in the test set). This way we can know how our prediction model performs in an independent data set.
-
27Interpreting Validation
In this lecture we interpret the output of the validation procedure, comparing the success rate (percentage of buyers) in the training set and test set.
-
28Predict the Propensity to Purchase For New Contacts
How to use our model predict the buying probabilities for new contacts that were just added to our list. This is all-important, because that's why we have created the model in the first place.
-
29Practical Exercises
Here you can find the practical exercises for the propensity to purchase models
Download Links
-
30Categorical Response Field
How to compare the effectiveness of two or more campaigns when the response field is categorical (e.g. yes/no, bought/did not buy etc.)
-
31Numeric Response Field
How to compare campaign effectiveness when the response field is numeric (usually, the purchase amount).
-
32Practical Exercises
Here you can find the practical exercises for the control package test procedure