Published on 24-Aug-2012
"For CEWE COLOR, IBM SPSS is definitely a quantum leap in analytics. In the future we will also integrate online customer data on surfing habits and click-rates into the solution, and this will continue to sharpen our insight into customer needs." - Eugen Neigel, Head of Sales and BI, CEWE COLOR AG & Co. OHG
CEWE COLOR AG & Co. OHG
BA - Business Analytics, BA - Predictive Analytics, Smarter Planet
Smarter Solutions for Retail
As the technological and market leader in photographic printing and production, CEWE COLOR AG & Co. OHG is the leading service provider for the European photography market. In 2011, CEWE COLOR supplied around 2.5 thousand million photos and over 5.1 million CEWE PHOTOBOOKS and photo gift items to over 40,000 retail partners in 24 European countries. Its retail customers include drugstores, photo specialist retailers, online retailers, specialist electronics stores and department stores.
CEWE COLOR needed a deeper insight into customer data and market demand, so it could increase customer profitability and improve its position in the highly competitive world of photo service providers.
With its data mining solution — IBM® SPSS® Modeler — the company can compile shopping cart analyses and detailed forecasting models that report on the preferences of customers and their likely behaviour.
Forecasting models deliver the basic facts required to tailor marketing campaigns precisely to customers’ needs. By issuing newsletters, mailings and targeted product offers at the right time, the company is in a position to build up a loyal customer base and increase retention.
To read a German version of this case study, please click here.
Instrumented: The solution integrates data from different primary systems and processes it for customer analysis and forecasting.
Interconnected: The solution connects company systems and specialist divisions, providing a central analysis system that allows users in marketing and sales to make intelligent analyses.
Intelligent: The solution identifies patterns in customer-specific data using intelligent statistical algorithms, generating detailed shopping cart analyses and intelligent forecasts of customer behaviour. Specialist users in marketing and sales obtain deep insights into customer data, and use them to offer customers the right products at the right time. This increases the company’s customer loyalty and profitability rates while preventing potential defections.
As the technological and market leader in photographic printing and production, CEWE COLOR AG & Co. OHG is the leading service provider for the European photography market. In 2011, CEWE COLOR supplied around 2.5 thousand million photos and over 5.1 million CEWE PHOTOBOOKS and photo gift items to over 40,000 retail partners in 24 European countries. Its retail customers include drugstores, photo specialist retailers, online retailers, specialist electronics stores and department stores. With 3,100 employees and 13 production sites, CEWE COLOR earned €469 million in revenue in the 2011 financial year.
The aim: to increase customer profitability
The photo service market is on an upward trend, and CEWE COLOR is no exception – with a broad product portfolio, the company has worked its way up to the number one spot among its European competitors. The company also meets the increasing demand for digital photo products and services with a number of innovations, including digitally produced photobooks, digital photos in different formats and printed gift items. CEWE COLOR has recorded a significant increase in sales of these items in recent years.
But the competition is not lying idle, and CEWE COLOR still has the potential to expand – especially in terms of customer profitability. “From around 25 million registered CEWE COLOR customers, only about half are active – which means they order regularly, not just as a one-off,” explains Eugen Neigel, Head of Sales and BI at CEWE COLOR AG & Co. OHG. “There is still considerable untapped potential here.”
This leaves CEWE COLOR asking itself: how can we make our customer relationships even more profitable? How can the company improve its customer loyalty and prevent people defecting to the competition?
Wanted: a view of the future
Customer data holds the answers to these questions: if you know your customers, you can tailor your product range precisely to customer needs and improve your position, relative to the competition. However, this requires an ability to reliably forecast your customers’ desires and future behaviour – a condition that the company’s existing information infrastructure was rarely able to fulfil.
“CEWE COLOR has been relying on IBM data warehousing and business intelligence technology for years now, and it delivers an exceptional insight into past business performance – sales, costs and much more,” explains Eugen Neigel. “However, we needed to adopt new technologies to enable detailed modelling and forecasting of customer development.”
CEWE COLOR needed a data mining solution that could detect patterns in historical customer data and predict likely future behaviour based on individual customer attributes. For example, answering questions like: which products is a customer likely to order next and when? Which other products might they be interested in? These insights should provide the sales and marketing department with the knowledge they need to understand the product preferences of different customer groups, identify customers who are at risk of changing to a different provider, and make relevant, targeted product offers.
Clear guidelines, clear goals
These requirements were echoed in a project called “Next Best Activity”, which was launched in mid-2010. The impetus for the project came from the executive marketing and sales committees.
“The guidelines from these specialist divisions were clear,” says Eugen Neigel. “Managers needed deeper insight into customer data so they could arrange promotional campaigns more efficiently and select the best customer interaction, every time – for example, sending out a specific newsletter at a specific time.”
From concept to customer data
Which solution would be best to achieve these goals? This question was answered by Eugen Neigel’s project team at the end of 2011. After several test runs and an IBM feasibility study, it was clear: IBM SPSS Modeler offered the necessary functionalities. With this data mining solution, customer and transaction data can be quickly and intuitively copied into forecast models, without a huge amount of effort or cost spent on programming. “Using intelligent statistical algorithms, the solution identifies patterns and trends that indicate likely customer behaviour,” explains Neigel. “Users in the specialist divisions were particularly impressed with how easy the solution was to use.”
The implementation completed in the spring of 2012 was just as easy, as Eugen Neigel explains: “Within a few weeks the solution was implemented. IBM SPSS was easily linked up to both the existing data warehouse and to the operational CRM system – a further advantage of the solution. The new solution extracts data from all the primary systems via the data warehouse. This includes financial data from the SAP ERP system, transaction data from our retail partners, data from newsletter marketing drives and the master data of registered customers.
Sound forecast models sketch a customer profile
This process draws on a vast amount of customer-specific data, which it uses as a basis for detailed forecasting models. SPSS processes customer data such as surname, address, gender and age, as well as registration method and order history. The company can even learn how newsletters are used: which newsletters did a customer open, which links did they click? The new solution takes these attributes and compiles scoring models which divide customers into different segments – best customers, average customers, liable-to-defect customers, and so on. The system also calculates product preferences, potential optional extras and likely order times for each customer.
“Even with several million customers and the large datasets that entails, IBM SPSS delivers forecast results within a few moments – the performance is outstanding,” says Neigel.
Predictive marketing increases sales
The marketing department stands to profit most from the forecasting models.
“Users can select any product name or customer attribute and obtain detailed, visualised shopping cart analyses practically at the push of a button,” explains Neigel. “Before now, this would have been either impossible, or only achievable at substantial cost,” explains Neigel. “Moreover, the users are completely independent of system administrators – they can create the analyses for themselves via the SPSS interface.”
The detailed forecasting models supply accurate, product-specific action suggestions for each customer. For example, on the basis of order history, master data and newsletter clicks the system can predict that “customer A” is very likely to order a glossy, hardcover CEWE photobook in the period of November/December 2012, and that he/she might also be interested in an XXL size photobook.
The forecast results are fed directly into the operational CRM system and the planning process for newsletters and mailing campaigns – so, for example, “customer A” will be assigned to the customer group that receives a newsletter on the subject of “large-size photobooks” towards the end of the year. As a result, the newsletter gets a more positive response, because it corresponds to the recipient’s preferences.
“With the new solution, CEWE COLOR is in a strong position to achieve its goals – making the relationship with its customer base more profitable, increasing cross and up-sale potential, and also boosting loyalty,” says Neigel. “We can achieve this by making the right offers at the right time. Our ultimate goal is to reduce the defection rate by one percent. With 12 million active customers and an average purchase order value of €15 to €20, you can easily work out how that would pay off over the years.”
Sales cooperates with retail partners
In addition to extending the company’s marketing insights, IBM SPSS supports the profitable sales of CEWE COLOR products via its retail partners. With the help of the data mining technology, the sales team can compile detailed shopping cart analyses of individual customer groups and provide these to their retail partners. The resulting insight into customer preferences and product affinities helps drugstores, online retailers and specialist markets to purchase the correct quantity of CEWE COLOR products and optimise their position in the market.
“For example, our retail partners might discover that in the coming months the demand for a certain photobook model or gift item is likely to increase,” says Neigel. “This might lead them to promote CEWE COLOR products via their online channels or at the point-of-sale in a targeted manner, enabling them to gain much higher returns. So both sides profit from this cooperation: CEWE COLOR strengthens its relationship with its retail partners, which in turn increases end-customer loyalty.”
Outlook: expanding the database
IBM SPSS is already supplying CEWE COLOR with deep insight into the aspirations and desires of its customers. Following the motto, “Think big, start small, grow fast”, the company wants to continue the data mining project and further exploit the potential of the solution. “For CEWE COLOR, IBM SPSS is definitely a quantum leap in analytics,” summarizes Eugen Neigel. “In the future we will also integrate online customer data on surfing habits and click-rates into the solution, and this will continue to sharpen our insight into customer needs.”
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.
© Copyright IBM Corporation 2012. IBM Deutschland GmbH, 71137 Ehningen, Deutschland. ibm.com/de. IBM Österreich, Obere Donaustrasse 95, 1020 Wien. ibm.com/at. IBM Schweiz, Vulkanstrasse 106, 8010 Zürich. ibm.com/ch. Produced in Germany. August 2012. IBM, the IBM logo, ibm.com, Let’s Build A Smarter Planet, Smarter Planet, the planet icons 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: ibm.com/legal/copytrade.shtml. 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.