10 Nov 2011
Master Data Management: Does an effective solution exist?
Despite being a term that has only reached popularity in the last decade, Master Data Management (MDM) is really a new turn on an old problem.
Managing master data such as customer, product, employee, locality and partner data has always been a major challenge – ever since organisations have tried to share or integrate data across systems. For this reason, master management marketplace has grown significantly in the last decade and is predicated to continue to grow aggressively over the coming years. Organisations are spending very large amounts of money on their Master Management programmes and they want to ensure their investment is sound.
However, in most large organisations, managing master data is a very complex problem that technology alone will not solve. The majority of underlying issues are process and competency-oriented:
- Organisations typically have complex data quality issues with master data, especially with customer and address data from legacy systems
- There is often a high degree of overlap in master data, e.g. large organisations storing customer data across many systems in the enterprise
- Organisations typically lack a Data Mastering Model which defines primary masters, secondary masters and slaves of master data and therefore makes integration of master data complex
- It is often difficult to come to a common agreement on domain values that are stored across a number of systems, especially product data
- Poor information governance (stewardship, ownership, policies) around master data leads to complexity across the organisation
MDM solutions are often perceived by business and executive management as significant and costly purely due to infrastructure improvement efforts lacking well-defined tangible business benefits. MIKE2.0′s approach to Information Development is one method can help implement the necessary competences to address some of these issues.
Would you recommend any other solutions, and if so, why? What do you think are the key capabilities required for an effective MDM solution?
November 10th, 2011 at 7:06 pm
I would recommend Semarchy Convergence for MDM.
It is an “Evolutionary MDM” platform designed for non-intrusiveness, project acceleration and data version control.
In short, it solves, from a multi-domain logical modeling, the problem of getting data from various sources converging automatically toward golden records.
Have a look at: http://www.semarchy.com/convergence/
November 10th, 2011 at 7:40 pm
I would recommend SAP NW MDM. This solution offers support for MDM for customer, product, employee, vendors, and business partners and offers a great deal of versatility and provides consolidation and harmonization of data, thus giving “one version of the truth” while providing ability to automatically import and syndicate data to and from the backend systems.
November 10th, 2011 at 8:28 pm
Whilst not ignoring the separate subject of data / information quality, I believe that we have developed a means of deriving and storing all information required for any operation / organisation incorporating ALL data not just master data. And it requires only two tables based around memory (what do I need to know) and activity (what do I need to do with what I need to know).
So far it works in all environments tried, but there is a way to go…
It is not yet available commercially because it is not yet finished, but within a year, who knows?
November 11th, 2011 at 2:29 am
It all starts with defining the scope of the business. An interesting observation that I have made is that we have gone through an era where less and less money has been put into the top half of the lifecycle, analysis and design. Business has done well in focusing on security and privacy, however loads of money gets wasted because of what turns out to be necessary design changes after the fact. OK, let me get to answer your question—
1) ANALYSIS
a) Adequate analysis needs to be done for the purpose of defining, categorizing and prioritizing data. This will fall out based on the scope of the business.
b) The flow of the data will fall out based on the decision making criteria. (Cost, resources, infrastructure, demographics, etc.)
c) Organize the data flow into a logical hierarchy then define its level of sensetivity. (Private, Shared, Publc, etc.) Then define those levels. (BY law, by personelle, by position, by project, etc.)
d) Make sure you are noting where compliance comes into the picture depending on industry standard. (PCI, CCB, HCCA, etc.) Incorporate relevant process to accommodate compliance and privacy.
e) Determine necessary space needed for data and the propensities in data growth. Do some research on data models to determine innate complexities involved with housing, manipulating and overall storage of similar data. (Bring a few examples to the table, chances are you’ll end up with some combination thereof.
f) CLEAR COMMUNICATION THROUGHT THIS PROCESS IS A NECESSITY. You want to make sure that the powers that be are on board every step of the way and should be involved as champions. User Acceptance is an ongoing process not just an end state.
g) Research hardware and software with an eye on being “Green” and ROI. The IS departments of today need to be accountable for how their efforts may and should increase ROI. The latest and greatest is just that but is not always what is needed.
h) Throughout this process keep in mind “tradeoffs” because of other technologies, Smart Phones, Pads, Hot Spots, and anything else you are able to think of that will fill a need without burdening a grow able infrastructure. Dialogue with each other, business to business, and above all CRM (your life’s blood) should take a top priority in your design.
November 11th, 2011 at 2:30 am
Hi Brenda. There are several solutions in the Data Management market. One of the most suitable is Informatica MDM (as shown in the white paper indicated by Sandrine). But if you search in the web you’ll find other products.
Of course, an MDM solution is a part of the entire MDM process. There is always a human part (business and executive management as you wrote) that must be convinced that application of such process will lead to significant money savings.
November 11th, 2011 at 3:04 am
I guess the issue we are facing is the lack of understanding of the “how”s of MDM adoption, mainly in the technology division. As soon as we hear ‘MDM’, all the other buzzwords are thrown into the mix. Data Quality, data stewardship, or anything anybody is wanting to do but not getting funding for. I understand data quality, data governance etc are imperative for the success of MDM. But we need to know ‘when’ and ‘how much’. For that we need to know ‘why’. Ask the project manager of MDM or the evagelist of MDM to write down a list of specific issues they are expecting the help of DQ, DG bodies to resolve, on the first year of the MDM program . That will reveal the lack of clarity on the ‘why’.
And from the business side, many think an MDM solution is something we can simply buy and implement with some additional integration effort. And all the software vendors are happily supporting that view for their benefit.
Overall, I believe we have very good solutions(Most vendor softwares, like IBM, Oracle, TIBCO) in the industry are now good enough to be used as a base for the technical side of MDM implementation. Also need to be said, none of these may be able to handle all MDM domains together like customer, product, location etc. It could be wise to write down your strategic MDM domain needs and then map them to a shortlisted set of vendor products.
However that vendor software could only be say 5-10% of the story. The industry is not matured in adoption of master data management. That is the real issue. And it is a jourvey that takes many years, usually 3+ years to find real benefits. The business side should be willing look at MDM as a strategic initiative.
Also before starding any MDM initiative, it would be advisable to create an MDM roadmap for at least 5 years, which show the major milestones of MDM adoption in the given orgranization and sequences of them. Then derive major program phases from it.
November 12th, 2011 at 11:02 pm
Master data management is a solution in search of a problem. “Single version of the truth”, “Golden copy”, “and 360 degree view” are all time worn clichés used by industry to claim MDM provides some magic elixir to solve the ills of 30 years of poor designs.
Added to this are the unsubstantiated claims of the magnitude of the data quality problem. $600 billion a year is one outlandish claim.
More recently we’ve witnessed the popularity of data governance as a prerequisite to successful MDM.
ERP and CDI preceded MDM in attempting to establish a single instance of entities. These failed to deliver and so was born MDM.
Poorly designed legacy systems, inconsistent data models, poor data management practices and varying and volatile business needs are the norm in many organizations and MDM is projected to be a solution to this chaos.
Besides, there is no “effective MDM solution”. Each organization has its own “versions of the truth” which infrequently map to vendor off the shelf solutions. Customization of these solutions results in creating yet another “instance of master data”. Another database to maintain, which is ill defined and becomes another silo and legacy system.
Executive management is skeptical of MDM because of the failure of MDM projects and the inability to address the real business needs. Before asking “what’s the solution”, ask “what’s the business problem?”. Many times the business needs can be addressed without building yet another database called MDM.
November 14th, 2011 at 5:03 am
Never knew about this. But thanks to you i learned a lot!
November 15th, 2011 at 2:41 am
Nicely written. There is one additional issue. Over time, within one organisation within different divisions, same words, different meanings. This is clearest seen in Siebel. Their system is sales oriented. It shows limitations when looking at it from a support point of view (at least in the past). SCOPUS had similar issues. Those systems were therefor not completely correctly used to get these non sales issues registered. That knowledge is gone, and as such we would now wrongly align information. That is the problem, wrong alignment over alignment starts painting a wrong picture. The data loses value.
I see this as a direct consequence (with some companies), to get that desired ISO certification and then cutting corners as costs kept on running up (and in some cases one solution just would not correctly fit).
November 15th, 2011 at 2:42 am
…and to add to the comments – don’t forget to map your master data management objectives to a tangible business outcome. It is extremely difficult to determine the “benefits” of effective master data management, data governance, data quality or any other initiative unless the program is grounded in some expected business outcome. Technology is not the answer. It is simply a means (a necessary means) to the end. Without the process aligned to specific, desirable outcomes the “win” is suspect and difficult to prove. Align your data/information requirements with the business processes that consume that data. It’s really the only effective way to establish a value proposition for improved quality and management of data. So to answer the original question – Yes, effective solutions do exist. They are the ideal combination of people, processes, and technology which achieve the desired business outcome leading to an improvement in your organization’s overall performance.
November 15th, 2011 at 2:42 am
Mike has hit the nail on the head – completely agree with you – too many projects fail because the initial business outcomes were not clearly defined in a manner which could be specifically measured. One of the biggest issues with MDM is that there is the danger of trying to boil the ocean with an enormous program of MDM work, when projects should be broken down into manageable, measurable projects which address the pressing needs of the business. Like enterprise Business Process Management, MDM should grow organically throughout the organisation starting with a manageable first steps project to get the frameworks in place and for the Centre of Excellence to be established. From that initial project a program of projects should then be formulated and prioritised to address the needs of the business. As the projects come online, so the project teams build competency, the business are educated as to what is possible and the technology is proven.
January 17th, 2012 at 1:59 am
[...] her post, Master Data Management: Does an effective solution exist?, Brenda Somich writes about many of the cultural, data, process, and other complexities that plague [...]