Bayer Schering Pharma

Global pharmaceuticals maker analyzes data with predictive analytics

Published on 05-Oct-2010

Validated on 04 Sep 2012

"Carrying out your own data analysis leads to deeper insights than if we were to rely solely on external market research institutes." - Dr Thomas Hein

Bayer Schering Pharma

Life Sciences

Deployment country:

Business-to-Business, BA - Business Analytics, BA - Business Intelligence

Smarter Planet:
Smarter Healthcare


Bayer Schering Pharma is one of the ten largest specialty pharmaceutical companies in the world. It markets products in more than 100 countries, and in 2008 generated sales of over €10.7 billion.

Business need:
Bayer Schering Pharma is looking for effective ways to identify illnesses that currently have no treatment methods, as well as medicines for preventing the side effects caused by taking other medicines. A further challenge lies in collecting detailed knowledge about patients in terms of their satisfaction levels, the structure and forms of treatment, as well as information about their product knowledge.

Bayer Schering Pharma uses IBM SPSS Predictive Analytics for target group segmentation and analysing survey and trial data.

• Identification of very precisely defined target groups using a priori segmentation • Segment-specific targeting of doctors using mailings, e-detailing measures or company representatives • Significant savings of resources and higher levels of customer satisfaction (doctors) • In-house analysis of survey and trial data leads to deeper insights and competitive advantages over and above that provided by external market research institutes

Case Study

Click here for a German version.

The deeper insights provided by predictive analytics create welcome competitive advantages for this specialty pharmaceuticals company. Questions its in-house market researchers ask include: Which illnesses do not yet have any form of treatment? And which medicines produce severe side effects that could be avoided by using a new medicine? How satisfied are patients with their treatment? How many patients could be helped by a certain medicine? What do patients know about a particular medicine? Are the blister packs or the tablet packaging structured logically enough so that patients understand them?

This broad range of questions forms the core of the work carried out by the 42 employees at the Bayer Schering Pharma Global Market Research departments in Berlin, Montville (New Jersey, USA), Mexico City and Singapore. The team of market researchers led Dr. Thomas Hein answers these questions in a very precise way on behalf of the pharmaceutical company’s Marketing and Sales divisions. They know, for example, that the birth control pill Yasmin is mostly prescribed by female doctors between the ages of 30 and 40 years old who practice in major cities.

Do-it-yourself market research with IBM SPSS Statistics and IBM SPSS Modeler
One crucial factor that distinguishes Global Market Research at Bayer Schering Pharma from other market research departments in the sector is the fact that the statistical analysis of the data collected to answer these questions is not completely left to external market research institutions. Dr. Hein’s department also analyzes survey and trial data in-house and is one of the most intensive users of the suite of IBM SPSS predictive analytics solutions.

Although Bayer Schering Pharma does employ external companies to carry out market research, there are three reasons why it ensures that not only the results of the analysis but also the raw data are handed over in the form of predictive analytics files. The first is quality control; the second is the ability to perform ad hoc analyses; and the third is the ability to carry out metadata analyses. The evaluation of “ordered” market analysis results often throws up new questions that the Global Market Research department can then quickly answer by performing ad hoc analyses of the data using predictive analytics software. Bayer Schering Pharma can use these files many years later for further metadata analyses from differing perspectives.

A priori segmentation creates tangible target groups
But that’s not the whole story. Bayer Schering Pharma generally segments the customer target groups – essentially doctors – using predictive analytics software and in line with its own methodology. Instead of the usual clustering methods, Dr. Hein and his employees use a priori segmentation to identify target groups which are actually achievable in reality. If the more commonly used cluster analysis of the segmentation reveals that a particular oral contraceptive or pill is being prescribed – especially by the group of young innovators – then the Marketing and Sales divisions will ask themselves the question: how can we target this group with specific promotions?

Therefore, during a priori segmentation and before the data analysis stage, the market research department develops “meaningful” and “addressable” variables – such as demographic data about doctors, surgery locations, hospital sizes, etc. – which it can compare with the surveys’ behavioral and setting variables. Using the factor and the CHAID analyses included in the IBM SPSS Modeler it is then determined which independent variables are meaningful, which behavioral factors and settings create the most distinct segmentations and which have the best predictive performance in terms of the behavioral trait being examined.

The results provide Bayer Schering Pharma with very precisely defined target groups – for instance, female, between 30 and 40 years old, practising in a major city – which provide indispensable guidance for the Marketing and Sales divisions. This leads to valuable resource savings for the company and increased customer satisfaction. This is because doctors will then only be targeted by Bayer Schering Pharma with segment-specific mailings, e-detailing measures or representatives when there is a sufficiently high likelihood that they will be interested in a particular medicine.

Feedback from the Marketing and Sales divisions supports the approach taken by Dr. Hein’s team. They can see that using target groups defined using a priori segmentation makes their work more successful than ever before. Although other approaches delivered customer groups with impressive sounding names, these were, however, not achievable in reality.

Competitive advantages thanks to in-house use of predictive analytics software
According to Dr. Hein, the advantage of using these statistical tools in-house – in contrast to competitors – is that it creates actual competitive advantages. As he states: “Carrying out your own data analysis leads to deeper insights than if we were to rely solely on external market research institutes.”

It allows a completely flexible approach and the software can even be used directly in a workshop with internal customers. Global Market Research has selected IBM SPSS Statistics as its standard statistics software because it sets the standard in the market research sector. Industry research institutes use it and so do almost all of the leading market research firms. Researchers know how to work with it because typically they have already used it in college or university courses. This facilitates cooperation between internal and external
market researchers, and eliminates any additional expense for training employees to use the software.

Only the industry standard statistical data analysis software is good enough
Yet there are also other reasons why Dr. Hein, his team and the whole market research operation at Bayer Schering Pharma use IBM SPSS applications to generate knowledge from data – although he regularly checks the market for viable alternatives. One is its user-friendliness, reflected in the fact that users can work with the IBM SPSS suite without any prior programming knowledge. A working knowledge of statistics and software is sufficient to run analyses using the IBM SPSS Windows-based interface. Advanced users with programming skills appreciate the software’s ability to save the syntax of a project and reuse it at a later date, a feature which can lead to significant time savings.

Another key point in favor of the IBM SPSS software: the training and consulting services offered by the software supplier. This was particularly helpful during the introduction of a priori segmentation when Global Market Research profited from one-to-one coaching offered, which helped the employees responsible quickly implement segmentation techniques using IBM SPSS Statistics.

The future with predictive analytics: data warehouse combines primary and secondary data sources
Bayer Schering Pharma will continue to rely on IBM SPSS software in the future. The next large project planned is for a data warehouse in which the company’s key performance indicators regarding its products, sales and revenue will be combined with those of its competitors, as well as with medical information on diagnoses and the subsequent medication. The analysis front end of the data warehouse will be an IBM SPSS solution. Dr. Hein is himself a veteran user of IBM SPSS software. He has been working with the product for 25 years and has followed its development as an active user from the rather cumbersome mainframe version to today’s user-friendly Windows version.

About Bayer Schering Pharma: Science for a better life – medical advancements for the benefit of the patient
Bayer Schering Pharma is one of the ten largest specialty pharmaceutical companies in the world. It markets products in more than 100 countries, and in 2008 generated sales of more than €10.7 billion ($14 billion).

Products and services used

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

SPSS Modeler, SPSS Statistics Base, SPSS Data Collection Data Entry, SPSS AnswerTree

Legal Information

© Copyright IBM Corporation 2010 IBM Corporation Route 100 Somers, NY 10589 US Government Users Restricted Rights - Use, duplication of disclosure restricted by GSA ADP Schedule Contract with IBM Corp. Produced in the United States of America May 2010 All Rights Reserved IBM, the IBM logo,, WebSphere, InfoSphere and Cognos are trademarks or registered trademarks of International Business Machines Corporation in the United States, other countries, or both. If these and other IBM trademarked terms are marked on their first occurrence in this information with a trademark symbol (® or TM), these symbols indicate U.S. registered or common law trademarks owned by IBM at the time this information was published. Such trademarks may also be registered or common law trademarks in other countries. A current list of IBM trademarks is available on the Web at “Copyright and trademark information” at SPSS is a trademark of SPSS, Inc., an IBM Company, registered in many jurisdictions worldwide. Other company, product or service names may be trademarks or service marks of others.