Master Data Management
Improving Corporate Data Governance
11th May 2005
Registered in England and Wales: Company No. 3426103
Registered office: 8 York Road, London SE1 7NA VAT Reg. No. GB 815 0730 56
Master Data Management
Improving Corporate Data Governance
Product Manager, Kalido
Welcome everybody. Thank you for attending this conference call on master data management.
I am Andrew Davis, the Product Manager for Kalido. With me on the call today is Jane Griffin, a
Partner at Deloitte Consulting, and Will Reilly from Ascential Software, which is an IBM company.
Today Jane will go through her experience with master data management, followed by a brief
introduction to some of the technologies from Kalido and Ascential that can be used in this area,
followed by a question and answer period. Let me start by introducing our participants in a little
Jane Griffin, our main speaker today, is with Deloitte Consulting. She is an IT professional with
over thirty years of experience. She founded her own company System Techniques many years
ago, which was later sold on to Prism Solutions. She also serves as a Partner on several advisory
boards for data management vendors. She has written a number of articles for DM Review and
speaks globally on issues around master data, data warehousing, and data management.
Will Reilly, who has joined us from Ascential Software, is a Solutions Executive in their
Industry Marketing Team. He works extensively in Retail and CPG and was recently named a
“Pro To Know” in the Supply & Demand Chain Executive Magazine. Prior to joining Ascential, Will
was a Supply Chain Consultant with IBM.
To introduce my role in a bit more detail, I am Andrew Davis. I am the Product Manager for
Kalido. I previously worked at Shell International, where I was involved with industry standards
for the process industry on data management and data modeling. I have been with Kalido since
its inception as a company approximately ten years ago.
These are the people on the call. I would also like to welcome the people attending the call; we
had over 600 attendees registered for this event, and we suspect there may be up to 300 people
on the call. Therefore, we may not be able to get to all the questions people have as we go
forward. I will now hand over to our principal speaker Jane Griffin, who will provide us an
introduction to master data management.
Master Data Management
Partner-Technology Integration/Information Dynamics,
Thank you for that introduction, and thank you Ascential and Kalido for sponsoring this event. It
is a lot of work to make this happen, and I appreciate the invitation to participate. All of you who
are on the call, thank you for your persistence in getting into the call and for taking time out of
the middle or end of your day to participate. We hope that this is informative, educational, and
enlightening, and may straighten out some of the discrepancies in the planet around enterprise
data management and master data management.
Defining Master Data Management
I have struggled a bit to define master data management because as I look at what Gartner and
some of the bigger vendors say about enterprise information management there are a variety of
different definitions for master data management. I have attempted to define it for you; because
this is my definition there will certainly be discrepancies as you talk to other people in the market.
I define master data management not as a product or a process on its own, but as a collection of
processes, technology and governance that spans all our organizational processes and application
systems, which gives us the ability to create, store, maintain, exchange, and synchronize a
consistent, accurate and timely record of our core business entities.
Those core business entities are tiered in three different levels, master data being the key and
core subject areas: Customer Product, Legal Entity, Chart of Accounts, Employee Vendor. There
are a couple more categories, and depending on what classification scheme you use those
categories could fall into aggregate data stores. Those are things like Bills of Material. Some
people categorize Bills of Material as master data; I would categorize that as an aggregate view of
master data, in that it combines one or more of your master data stores, such as Vendor Product
or multiple instances of Product, to form a new result.
There is also Transaction data that occurs against your master data. All of those things could be
considered to be synchronized. Enterprise master data management also supports those
transactions, and is enabled by a series of tools, techniques, processes, governance, and so forth
to give you a consistent view of your enterprise.
Master data also has different levels of consistency. In implementing global master data
management for large organizations, which I have been involved in for the last several years,
they would often pick tiers of master data definitions, deciding they wanted to aggregate
information about their product types, geographies, customers, or the vendors they do business
with. They wanted to set global standards for master data. There may only be ten or twenty
data elements - in one instance I had a client who had sixty - that are considered global across all
their master data stores. You would also have divisional and possibly local stratospheres of your
master data definitions.
Growing Importance of Master Data Management
With that in mind, I want to set the record straight on what master data management is. Why
are we concerned about this all of a sudden? For the last several years it has become one of the
number one questions that I get asked by my clients. I visited with Gartner, META and Forrester
a month or two ago and they also confirmed that one of their number one topics was how to
manage information assets in a business.
As I looked back on some of my projects over the last five to seven years, I discovered one I had
done to assist a client with the development of an Enterprise Application Integration Strategy. As
I looked at the study, it showed that their Product information was out of synch; they could not do
line-of-sight on the product development process because they could not see the original product
engineered through product manufacturing, sales and distribution, etc. They had selected the
EAI Enterprise Application Integration Technology to help facilitate that. As we were tying all
these messages together we found that if the sender and recipient of the message were not
speaking the same language, the message did not compute. We needed master data; we needed
a consistent nomenclature and a consistent language in our enterprise, so that I receive a code I
can interpret that code appropriately.
Drivers of the Move Toward Master Data Management
Enterprise application harmonization is a big need and a big driver in master data management,
as well as consolidation of applications. Most organizations are taking costs out these days and
trying to sunset 100 ERP systems and move to a few; usually less than a dozen, and maybe less
than a handful. They have found in previous sunsetting activities that if they do not get the data
consistent they have not realized a real advantage. Master data management is certainly being
driven by the need for harmonization of messaging and consolidation of applications. That is on
the process side.
On the business intelligence and decision support side, companies are finding that if they do not
have consistent data they cannot aggregate that data at the level required to manage their
enterprise without significant pain and manpower. I just finished a strategy project with a client.
They have 1,800 Financial Management people; 500 are in their shared service center, and 1,300
of them are master data management people doing this by spreadsheet. They categorize, re-key
and aggregate data because it is not coded properly for their consolidation and financial reporting
process. That is a lot of manpower, and I am sure there are a lot of things going wrong with that
data as it is reported. It is incredibly important that we get the master data problem solved.
The third big driver of the move toward master data management may in fact be number one on
some people’s list. As public companies we now have to sign off on assuring we have compliance
and good data quality. That means I need transparency into the data movement in my
enterprise. That is a big problem in most enterprises.
Key Processes Dependent on Master Data
The key processes in our businesses that rely on master data are significant. I have worked with
organizations that are just trying to rationalize their financial or operational information, but that
always leads them to the following question: If I am not getting the right information from my
procurement side, my product development side, or my sales and marketing side, am I really able
to standardize that? The answer is yes in terms of Charge of Accounts and perhaps in terms of the
overall Legal Entity. However, I still need to synchronize that back to all parts of the business.