And this is exactly where companies face challenges nowadays. Mergers & acquisitions and the autonomous expansion of companies in recent decades have led to multiple customer systems with duplicate and often inaccurate and incomplete data. This has impacted effective and efficient business operations in numerous ways: ineffective customer contact because of non-distributed data, inefficient operations caused by erroneous or incomplete shipping
“The purpose of business is to create, and keep, customers”
Peter Drucker
Customer Data Integration
and billing data, compliance issues caused by incomplete and inconsistent customer data in systems, and so on. The increased use of the internet as an additional sales channel, and the abilities to capture the customer’s preferences via this channel, have also created an additional driver for creating that single view of the customer.
Customer Data Integration can be described as the mix of processes, governance, data standards and
technologies that consolidate data from across the organisation to create one centralised view of the customer.
A KPMG Round-Table on 13 October 2014 on this topic zoomed in on the ‘How’ question: “how do companies approach the issue of creating this single view of the customer?”. In his introduction, KPMG partner Ronald Jonker underlined the need for many businesses to improve their customer data quality and embed sustainable customer data management into their organisations. As Ronald said: “the battle for the e-Customer has begun. Only businesses that are truly able to provide e-Commerce services as a ‘one company-one portal-one view’ experience, showing they have captured the customer’s preferences, and that are able to use this information in a trusted way, will survive.”
ADVISORY
‘Customer-centricity’, ‘the perfect customer journey’ and ‘seamless
ordering and delivery’ have all become guiding principles for companies
wanting to improve their top-line performance. It all focuses on knowing
your customer, providing them with the right products or services, of the
required quality, for the agreed price and in a timely manner. This isn’t a
new concept. It has always been the mantra for successful executives,
no matter where they operate, no matter what products or services they
offer. And all it requires from a data and information perspective is that
companies are able to capture their relationship with the customer in a
single, accurate and complete view.
“A QUALITY CUSTOMER VIEW:
from the Vision to the Reality”
For a bank with a mission of ‘empowering people to stay one step ahead in life and in business’, it is crucial to have a single, complete and highly accurate view of its customers if it is going to offer them tailor-made products and services. To be able to provide this high-quality customer view, Jeroen and his team undertook the mammoth task of designing and implementing an entirely new operating model for ING (of which the customer view is only one aspect). Among their objectives: to completely digitise all customer data processes, to create one single source of the truth with quality master data, and to significantly simplify the IT landscape.
Not sustainable
If this looks ambitious on paper, imagine the challenge in real life at an organisation as large as ING. Modern banking customers increasingly expect transparency and real-time information on their personal data and financial status. Customers are also using a
Jeroen van Dullemen, head of Customer Data Management at ING Netherlands,
as well as the chairman of the Netherlands Platform for Information Quality (NLIQ),
discussed what it took at ING to create a single, high-quality customer view.
“You need to find ways to get on people’s priority lists by making the benefits
your project will bring them clear.”
growing mix of channels to interact with their bank. And when Jeroen and his team started on their journey in 2008, the bank’s infrastructure was certainly not prepared for this. Back then, ING was a bank that sold millions of products to millions of customers, a bank that used a patchwork of IT systems but lacked a central email address or phone number, and showed signs of the silo-mentality typical of many large organisations. It also worked from a once successful but now
outdated ‘pre-internet’ operating model that still used a lot of manual, paper-based, batch- based and channel-specific processes. Jeroen’s verdict was clear: “Our operating model simply was not sustainable for the next 20 years.”
Carrot and stick
Even so, it took a fair amount of work and perseverance to convince the organisation to give priority to a thoroughly redesigned, future-proof operating model. Jeroen saw how bank employees were contacting ING’s call
centre to retrieve information while attending to customers in front of them. Even so, he found that theoretical, rational arguments alone were not sufficient to obtain priority status for his project. He attributed this to the fact that everyone has their own priorities and problems and innovations always seem to outweigh a ‘general interest’ project like his. “To overcome this, you need to find ways to get on people’s priority lists by making the benefits your project will bring them clear. You need to figure out and really understand what drives them, what problems they’re facing, or find a suitable carrot and stick approach. Of course, that’s part of what makes it fun.”
Customer data as a business
Some of the highlights of ING’s approach centred around the organisational set-up and data governance aspects. “We have created a business department for customer data management, called the Customer Domain, which takes business
responsibility for all policies, rules, standards, requirements and so on related to customer data. The department also initiates and steers projects for change. Creating a separate business department was a significant step at the time, because previously everything related to data was covered by the IT department.
In response to a question from the audience on how ING arrived at a single, integrated data model, and how it decided on what should be included in or excluded from this central data model, and which data attributes should be maintained centrally, and which locally, et cetera, Jeroen informed the audience: “We basically made a distinction between data used throughout the organisation, and data useful only for specific purposes. There we touched up on the silos, the departments that are responsible for different channels or products, and that use their own specific data models. And we claimed responsibility for those parts of their customer data that are useful for the entire bank, combining this data into a master data management system. From that single source of truth, we distribute the data throughout the entire bank, and we assume responsibility for all the requisite data management and data quality processes. This very clear definition of roles and responsibilities has helped us a lot, because it gives us something to steer on.”
Customer data quality is partly managed via a ‘self-service’ concept: where appropriate, the customer view is also
KPMG Customer Data Integration | 3
provided online to clients, who can then check and, if the relevant policy allows, change the data themselves. “And regardless of whether clients make changes online or need to visit a branch office to have their data changed, the way these changes are processed is now identical and immediate. What some years ago might have taken a couple of weeks, is now performed in almost real-time. And this ‘self-service’ concept really works. A recent
questionnaire showed that a significant number of our customers had visited their personal homepage or even made data changes at least once, even without us offering an iPad or other such incentives for them to do so.” In response to a question from the audience, Jeroen explained why ING has chosen not to install a data governance board: “I agree that we need support from senior management, and for a while we did have a data council. But we quickly concluded that nobody was really interested in data as such, and that a separate data council was inefficient and a waste of everybody’s time. Instead we make as much use as possible of existing structures or value chains if we need to address data management issues. The key here is that we always try to treat processes and data together, and almost never just data in isolation.”
Achievements
Today, ING is busy implementing the plans that started as a simple
PowerPoint presentation five years ago. “We are well underway to creating a
single source of truth and a single customer view. Our aim is to digitise all the bank’s processes. To date, 95% of our products are coupled on a real-time basis to the central customer administration, enabling a truly 360-degree customer view. Finally, within ING there is now a lot of awareness around our new operating model. Everybody understands what needs to be done and is cooperating; these days, almost nobody needs to be convinced.”
Lessons learned
• Data is a business asset and therefore a business responsibility. For ING it has been crucial to establish a separate business department responsible for data management and processes.
• Changing the information architecture touches on the architecture itself, the processes, the data model and the data quality, and you always need to be aware of the interconnections between them.
• ING has come to realise it is part of various value chains; the key lesson in this area has been the need to understand what data does in the company and what it does for the customers.
• Programmes such as this require perseverance: always keep going, even if you’re not always getting what you want at the time you want it. The chance will come to achieve what you need at a later time.
KPMG Customer Data Integration | 4
The first case involved a German apparel and footwear company with serious issues in its CRM systems and processes. These issues were further complicated by the lack of an appropriate data governance model and clearly defined organisational roles and responsibilities. Customer master data was spread over 25 databases, 21 of which did not interface with any other system. As a consequence, its marketing and sales campaigns yielded poor results, and since feedback from these campaigns was not entered into the customer master data, were bound to disappoint again.
Mickey Mouse
The approach to remedy this situation involved the definition of a new data model that enabled the use of much improved customer profiles that were also enriched with data from external sources, and a rigorous cleaning of the existing data. After weeding out duplications, incomplete entries and scores of customers named Mickey Mouse or Pipi Longstocking, only 2 million of the original 6 million records survived. In addition, processes for master data governance and data quality maintenance were introduced, and the number of databases in the new IT landscape was much reduced.
The results, after only seven months, were quite spectacular. Campaign effectiveness, as measured by the percentage of emails opened or the number of links clicked on, had increased by some 25%, while web sales effectiveness was up by 10%. Significant savings were achieved because the number of FTEs involved in customer data management was reduced dramatically, by a factor of 10 even on a central level. Last but not least, with much better insight into campaign effectiveness available, around a quarter of all campaigns could be discontinued.
THE SHERLOCK HOLMES PRINCIPLE:
Good-quality integrated customer data
is hard to find. It is even more difficult
to create tangible business value.
Antje-Kathrin Schumann, head of KPMG Germany’s Customer Management
Group, presented two cases in which companies had to deal with a number of issues
related to customer data management in order to achieve excellence in this area.
A number of lessons learned were drawn from these and other, similar cases.
24 million records
The second case described how the data quality issues at a German automotive company were tackled. Across its two divisions, Automotive and Financial Services, 24 million users were registered on the website and on the online car configurator. The records included vast amounts of incomplete and fictitious data entries, nevertheless these records had been downloaded to the company’s SAP system. A huge data cleansing programme was started, with thousands of employees involved worldwide, which would in the end leave only 13 million of the 24 million records in the system. The remaining records were enriched with credit and household data from external suppliers, and transactional data (e.g. car configurations, complaints, leasing contracts) was reassigned. Overall, data quality was increased by 20%, resulting in clear business benefits: processing costs for contract creation were reduced by 15%, and payment defaults by 30%.
Lessons learned
A number of ‘lessons learned’ have been distilled from these two cases, as well as from other projects focused on achieving customer data excellence, and these were presented during the round-table in the form of ‘key enablers’ and ‘critical success factors’.
The key enablers were grouped into four categories:
Data governance
Focusing among other aspects on data management strategy, policies and procedures. Subsequently it involves data ownership and management commitment and involvement through a data governance council or a similar body.
Data quality
Focusing among other aspects on the definition of quality criteria for key data fields, on the measurement of data quality and on data cleansing. Data enabling processes
Focusing among other aspects on data lifecycle processes and related tasks and responsibilities.
Technology
Focusing among other aspects on the IT architecture that facilitates all of the above. It ranges from tools for data governance, data quality and workflow to master data management systems. The following were identified as critical success factors:
Management commitment
This includes the explicit and active involvement of management in data quality and data management councils. On the part of management, this commitment requires an understanding of how data management and data quality can affect the efficiency and quality of operational processes, and ultimately the overall success of the company.
Structured and goal-oriented governance This includes defining responsibilities along the chain of operational processes (e.g. marketing, supply chain management), and developing and communicating the appropriate guidelines and rules to ensure the governance changes are embedded throughout the entire company. Optimised processes
Excellence in master data management (MDM) can only be achieved if the relevant processes are organised and executed at a sufficiently high-quality level. Any gaps, duplications or
inconsistencies in MDM processes may lead to poor data quality, operational inefficiency and compliance issues (e.g. missing data histories).
These cases triggered lively discussions on various topics, such as the vital importance of customer hierarchy management, the do’s and don’ts when engaging customers in assembling customer master data, data migration strategies and the development of incentives that immediately and directly benefit those involved in data entry and maintenance. These discussions certainly suggest there are more than enough topics left to address in future round-tables on Enterprise Data Management. After the presentations, participants took the opportunity to join in further discussions and expressed their appreciation to KPMG for organising what had been a valuable round-table event.
The information contained herein is of a general nature and is not intended to address the circumstances of any particular individual or entity. Although we endeavor to provide accurate and timely information, there can be no guarantee that such information is accurate as of the date it is received or that it will continue to be accurate in the future. No one should act on such information without appropriate professional advice after a thorough examination of the particular situation.
© 2015 KPMG Advisory N.V., registered with the trade register in the Netherlands under number 33263682, is a member firm of the KPMG network of independent member firms affiliated with KPMG International Cooperative (‘KPMG International’), a Swiss entity.
Contact
Ronald Jonker
Service Leader
Enterprise Data Management KPMG Advisory N.V.
tel. +31 10 453 4113 mobile +31 6 512 05 427 e-mail: [email protected]
KPMG
Laan van Langerhuize 1 Amstelveen
P.O. Box 74500 1070 DB Amsterdam
