Kaplan RM helps its clients gain deeper customer insight

Providing a wide range of sophisticated analyses of consumer behaviour

Published on 14-Jun-2012

Validated on 02 Dec 2013

"Thanks to IBM SPSS Modeler, we are now able to offer more sophisticated predictive analyses, and deliver deep customer insights faster than ever before. Combined with our expertise in communications and marketing, this provides a compelling proposition for our clients." - Sofia Bauer, Head of Customer Insight, Kaplan RM

Customer:
Kaplan

Industry:
Professional Services

Deployment country:
Sweden

Solution:
BA - Business Analytics, BA - Predictive Analytics

Overview

Kaplan RM is a specialist consulting agency that focuses on relationship marketing – helping clients understand the way their customers behave and foster customer loyalty. It has around 25 employees and offers services in four business areas: management, creative, technology and customer insight.

Business need:
Kaplan’s clients are increasingly moving to centralised CRM platforms, which enable them to give Kaplan researchers direct access to customer-related data. To take full advantage of these vast data sources effectively, Kaplan needed a sophisticated data mining and modelling solution.

Solution:
Kaplan implemented IBM SPSS® Modeler, which it uses to create a variety of analyses to support clients’ research projects. Recent projects have included customer churn models, cluster and RFM analysis for segmentation, CHAID analysis for product profiling, and descriptive analysis of high-value customers.

Benefits:
Enables Kaplan to offer more sophisticated analytics services and win more business from its clients. As a result, the solution has delivered a full return on investment within one year. Allows researchers to perform large-scale analyses of millions of consumer transactions, which was not possible with spreadsheet-based analysis techniques. Eliminates manual data extraction and formatting tasks, saving up to three weeks per research project for Kaplan and its clients.

Case Study

To read a Swedish version of this case study, please click here.

Kaplan RM is a specialist consulting agency that focuses on relationship marketing – helping clients understand the way their customers behave and foster customer loyalty. It has around 25 employees and offers services in four business areas: management, creative, technology and customer insight.

The customer insight team is composed of four research specialists who focus on analysing data from clients’ CRM systems and combining it with traditional research methods (such as surveys, interviews, focus groups and external information sources) to enable customer segmentation and campaign analysis.

Moving away from spreadsheet-based analysis

“Traditionally, we used to work with our clients’ database administrators to extract the data we needed from their systems,” comments Sofia Bauer, Head of Customer Insight at Kaplan RM. “It was a complicated process because we were often drawing data from several different sources, and we had to do a lot of preparatory work just to find out what information was available.

“We then imported the data into spreadsheets so that we could analyse it. This was far from ideal, because our larger clients often have customer databases that run to several million rows, and our spreadsheets struggled to cope with the scale of the data. Even minor transformations could take a couple of hours to perform.”

Harnessing data from centralised CRM systems

In recent years, Kaplan has seen an increasing number of its clients move from a fragmented systems landscape to a more coherent approach based on central, enterprise-wide CRM platforms. The customer insight team realised that this provided an opportunity to make data gathering a much more painless process for both Kaplan and its clients.

“The integration capabilities of modern CRM systems make it possible for some of our clients to give us direct access to their customer database,” explains Sofia Bauer. “We wanted to find an analysis tool that could help us take full advantage of this by automating the extraction and analysis of data.”

The right tool for the task

Kaplan considered various options, and ultimately selected IBM SPSS Modeler.

“I had used SPSS before in a previous job, and I had found it so intuitive that I was able to teach myself how to use it without any formal training,” explains Sofia Bauer. “This was an important factor in our decision, because we expect to take on a number of projects in the near future that will require statistical modelling, and the customer insight team will need to get up to speed quickly on how the software works.”

The business case for investing in the software was built around the extra capabilities that IBM SPSS Modeler could offer for Kaplan and its clients.

“You can do quite a lot with spreadsheets, but some of the more advanced types of analysis simply aren’t possible – especially when you start looking at predictive modelling,” explains Sofia Bauer. “We forecast the number of new research projects we would be able to win if we were able to offer these capabilities, and estimated that the investment in IBM SPSS software would pay for itself within a year. And looking at the projects we have worked on in the last 12 months, I’d say that we have easily achieved this.”

A consultant from IBM Sweden helped Sofia Bauer install the software, and provided two half-days of consultancy to introduce her to some of the new features that had been released since she last worked with the software. She comments: “The technical set-up only took an hour or two; the value that the consultant brought was in showing me some tips and tricks that really helped with creating models.”

A variety of applications

One of the first projects that Kaplan worked on with IBM SPSS Modeler was the creation of a customer churn model for a Swedish bank.

“We created a model that calculated a risk score for each customer, helping the bank identify those who were at greatest risk of closing their accounts in the next three months. We discovered that customers with risk scores in the top 10 percent were four times more likely to churn than average customers. We’re now testing the model on various target groups, and trying to find out which offers and channels the bank can use to increase customer retention.”

Next, Kaplan used the software to help a retailer understand the purchasing behaviour and attitudes of people who buy beauty products at its stores. The research used a questionnaire to gather information about a set of customers, and then ran cluster analysis on the data to identify the main groups within the respondents.

“We found five significant clusters of behaviours in the data, and we created five profiles to describe the types of people who belonged to each group,” explains Sofia Bauer. “The next step is to connect the clusters from the survey to data about the general population of customers, and find out whether the same five groups appear. If so, we can help our client design marketing strategies that will appeal to each group, which should help to boost sales and increase customer loyalty.”

RFM and CHAID analysis

A third project involved using the IBM SPSS software to assess the recency, frequency and monetary value (RFM) of purchases made by individual customers, and use this information for customer segmentation. First developed in the 1960s, RFM analysis is a proven method for increasing response from marketing campaigns – but it can be laborious to conduct this kind of analysis for large data-sets. The IBM SPSS solution automates the process and delivers results within a few mouse-clicks – enabling Kaplan to spend more time on strategic aspects such as creating effective communication plans for each customer segment.

Kaplan has also used a slightly more sophisticated approach – chi-squared automatic interaction detection, or CHAID – to help a travel company understand how a customer’s general behaviour related to their decision to purchase a particular type of ticket. This has enabled the client to identify other customers whose general behaviour indicates that they might be interested in purchasing the same type of ticket. In the near future, the travel company will be launching a marketing campaign aimed at these customers to raise awareness of the product and attempt to boost sales.

Taking a deep dive into customer behaviour

Finally, the IBM SPSS solution makes descriptive analysis much quicker and easier, enabling Kaplan to offer its clients a closer look at individual customers. For example, when an RFM analysis has revealed a company’s most loyal and profitable customers, descriptive analysis can be used to gain deeper insight into each of them, to find out what products they buy, which loyalty programmes they have joined, where they live, which stores they visit, and so on. This allows clients to build up a picture of what is most important to their core customers on an individual level, and tailor their marketing and sales strategies accordingly.

Sofia Bauer concludes: “Our clients have all this customer-related data at their disposal, but very often they just aren’t able to harness it in a profitable way. The unique value that Kaplan provides is to unlock the value of this data and use it as a foundation for creating new communication strategies that help to build closer, more profitable customer relationships. Thanks to IBM SPSS Modeler, we are now able to offer more sophisticated predictive analyses, and deliver deep customer insights faster than ever before. Combined with our expertise in communications and marketing, this provides a compelling proposition for our clients.”

About IBM Business Analytics

IBM Business Analytics software delivers actionable insights decision-makers need to achieve better business performance. IBM offers a comprehensive, unified portfolio of business intelligence, predictive and advanced analytics, financial performance and strategy management, governance, risk and compliance and analytic applications. With IBM software, companies can spot trends, patterns and anomalies, compare “what if” scenarios, predict potential threats and opportunities, identify and manage key business risks and plan, budget and forecast resources. With these deep analytic capabilities our customers around the world can better understand, anticipate and shape business outcomes.

For more information

For further information please visit ibm.com/business-analytics.

Request a call

To request a call or to ask a question, go to ibm.com/business-analytics/contactus. An IBM representative will respond to your inquiry within two business days.

Products and services used

IBM products and services that were used in this case study.

Software:
SPSS Modeler Server

Legal Information

© Copyright IBM Corporation 2012. IBM Svenska AB, SE-164 92 STOCKHOLM, Sweden. Produced in Sweden. June 2012. IBM, the IBM logo, ibm.com and SPSS are trademarks of International Business Machines Corporation, registered in many jurisdictions worldwide. A current list of other IBM trademarks is available on the Web at “Copyright and trademark information” at www.ibm.com/legal/copytrade.shtml. Other company, product or service names may be trademarks, or service marks of others. References in this publication to IBM products, programs or services do not imply that IBM intends to make these available in all countries in which IBM operates. Any reference to an IBM product, program or service is not intended to imply that only IBM’s product, program or service may be used. Any functionally equivalent product, program or service may be used instead. All customer examples cited represent how some customers have used IBM products and the results they may have achieved. Actual environmental costs and performance characteristics will vary depending on individual customer configurations and conditions. IBM hardware products are manufactured from new parts, or new and used parts. In some cases, the hardware product may not be new and may have been previously installed. Regardless, IBM warranty terms apply. This publication is for general guidance only. Photographs may show design models.