Has Your MDM Vendor Found “Multi-Entity Religion”?
Subtitle: Is Your MDM Road Map “Strategic” or “Myopic”?
Monday, March 24, 2008
(Note: If you have trouble accessing the hyperlinked articles, please go directly to MDM Alerts)
In the past 2 ½ years, the MDM Institute has hosted more than 4,000 attendees at our worldwide MDM SUMMIT series (formerly the CDI-MDM SUMMIT). These highly-regarded events have taken place in major metro areas ranging from New York, San Francisco and Toronto in North America to Frankfurt, London, Madrid, and Moscow in Europe and all the way to Sydney. Our MDM Alert subscriber base has grown to 7,500+ subscribers. Our MDM Advisory Council now includes such organizations as: 3M, Cisco Systems, Honeywell, Microsoft, Norwegian Cruise Lines, Westpac, Weyerhaeuser, Woolworths Australia, and forty other Global 5000 size enterprises who have turned to us for assistance with their MDM initiatives. And we have been invited to speak at or attend most every MDM vendor’s national or international user group meetings.
Why the bravado? Most analyst firms have not made the leap of faith to get from CDI, PIM, EIM, et al to “multi-entity or multi-domain MDM” and instead remain quagmired in the same “mud wrestling” mode of recommending discrete mastering capabilities for both party and product.
So let’s put the stake in the ground. Recently, we interviewed 25 prospects of the leading MDM providers about their enterprise MDM strategies. According to these decision-makers/influencers (e.g., enterprise architects), clearly multi-entity or multi-domain MDM is their focus more so than discrete customer data hubs (a.k.a. CDI) or product data hubs (a.k.a PIM hubs). Moreover, these MDM evaluation teams are intent on investing in vendors’ products whose road maps are clearly tracking to the next (4th) generation hub requirements.
Accordingly, every CDI vendor (and many a PIM hub vendor) has found “multi-domain MDM” religion. So the recent buzz around MDM is rivaled only by intensity with which CDI vendors have "found multi-entity MDM religion". And they are not alone. Being the brilliant analysts and marketers that we are, we renamed both our business and our conference series to map to this trend (we are now the “MDM Institute” and our conferences are now the “MDM Summits”).
Clearly, enterprise MDM is a major IT initiative. Most enterprises and solutions vendors are finding modest near-term success (relatively quick time-to-value) with the single-faceted approach inherent with the third generation of MDM solutions. Increasingly, however, these same enterprises are determining that this myopic strategy of focusing solely on a single data domain and usage style is detrimental to the longer-term business strategy of integrating supply, demand and information chains across both product intra- and extra-enterprise boundaries. Coming to market in 2008 are a wide array of multi-entity MDM solutions, which we have historically characterized as the 4th generation of MDM solutions.
Clearly, the future direction is to also grow all reference masters into operational masters, e.g., pricing and location style masters into transactional support roles via operational, analytical, and collaborative MDM linkages.
Through 2008-09, Global 5000 enterprises will broaden their MDM business initiatives from single use case, single entity to multi-style, multi-entity. By 2010-11, enterprises without a long-term multi-entity MDM strategy run the ironic risk of building “MDM silos” which will need to be patched/fused together via middleware – in effect, recreating the original MDM problem associated with master data segregated and isolated within ERP and CRM instances.
Inside this MDM Alert are more detailed discussions of:
What are the “Vital Signs” of a 3rd Generation MDM Solution?
Why “Multi-Entity MDM”? Why Now?
What are the Necessary Features of 4th Generation MDM Solutions?
Which Vendors Have Found “Multi-Entity” Religion?
BOTTOM LINE : Contemporary MDM solutions must provide support for multiple types of master data (domains or entity types) else risk marginalization as large enterprises are increasingly mandating such capabilities. Specifically, the overarching concern is to avoiding "repaving the cow paths." This occurs when an IT organization executes on a shortsighted strategy of mastering the master data in one given business area with a specific brand of MDM solution and then discovers that another division or line of business has chosen a different brand (and architecture) to solve their other “mastering” issues. All too often, these different product-specific MDM solutions do not offer the capability of integrating master data across the great divide between party master data (customers, suppliers, employees) and product master data.
“Heads up” from the MDM front lines and see you at our third annual San Francisco event (MDM SUMMIT – Spring 2008) this March 29th – April 1st.
Aaron Zornes
Chief Research Officer
Feel free to pass along a copy of this newsletter to colleagues who may be interested.
Overview of Multi-Entity MDM
- MDM has matured from “early adopter IT project” status to become a mainstay “Global 5000 business strategy”
- SOA-based MDM with multi-entity support is desired to manage master data domains (customers, accounts, products, etc.) that have significant impact on (and span across) the enterprises’ most important business processes – e.g., compliance, cross-sell/upsell, and customer service
- MDM is increasingly concerned with the notion of “multiples” – multiple data domains, the multiple relationships among them, and the multiple usage styles
- Most vendors approach MDM from either specific usage/domain pairing, or broad tool – e.g., a rudimentary data model and set of tools for data quality, workflow, etc. to build out the enterprise’s own MDM infrastructure
- Conversely, an example of product-centric MDM evolution broadening into master “party” data (customers and retailers) for build-to-order product management and more effective marketing
Clearly, enterprise MDM is a major IT initiative being undertaken by a large number of the market-leading Global 5000 size enterprises. Most enterprises and solutions vendors are finding near-term success with the single-faceted approach inherent with the third generation of MDM solutions. Increasingly, however, these same enterprises are determining that this myopic strategy of focusing solely on a single data domain and usage style is detrimental to the longer term business strategy of integrating supply, demand, and information chains across both product intra- and extra-enterprise boundaries. Coming to market during 2008 are multi-entity MDM solutions we characterize as the fourth generation of MDM solutions which address the requirement for multiple domains and styles as well as the roles of the consumers.
To help IT organizations and their business partners focus on the more desirable longer term MDM strategy, vital issues that this MDM Alert addresses include:
- What is multi-entity MDM?
- Why is multi-entity MDM considered “evolutionary” while domain-specific MDM data marts viewed as “myopic”?
- When will multi-entity MDM graduate from “early adopter IT project” to Global 5000 strategic business initiative?
- Which use cases are most amenable to benefit from multi-entity MDM?
- How does an organization plan for multi-entity MDM deployment?
The value of multi-entity MDM can be intuitively recognized in a range of business initiatives – from short-term fixes to a narrow set of problems such as capturing customer privacy preferences across product lines to long-term enterprise-wide initiatives to delivering infrastructure agility by embracing SOA.
During 2007, MDM solutions matured from “early adopter IT projects” to become “Global 5000 business strategies”. The industry consensus is that “multi-entity MDM” is a software solution to concurrently manage multiple, diverse master data domains (customers, accounts, products) across intra- and extra-enterprise business processes. Additionally, enterprises are determining that master data must be presented via multiple views to accommodate the wide range of data consuming and creating roles that exist across an organization. By centralizing the most critical master data to a single trusted source, and managing this within the context of governance-driven data lifecycle, multi-entity MDM provides flexible business process integration across multiple data domains and usage types. Multi-entity MDM solutions deliver such complex and collaborative business processes such as identifying the most valuable customers, introducing new products rapidly, crafting new product bundles more quickly, and managing threat and fraud risk more effectively.
Global 5000 size businesses are rapidly ramping up plans to consolidate “master” data into data hubs using a combination of off-the-shelf data hubs, EAI/EII/DQ toolkits, and even custom-built IT projects. The current commercial off-the-shelf solutions available to enterprises are commonly characterized as 3rd generation solutions.
What are the “Vital Signs” of a 3rd Generation MDM Solution?
In our experience with the MDM Institute Advisory Council membership, “type A” MDM project leadership within very large scale IT organizations advise of these five “DNA markers” as good indicators:
- SOA/Shared Services Architecture Evolving to “Process Hubs”
Rather than re-invent the “data hubs” that are inherent within enterprise CRM and ERP vendor solutions, savvy IT organizations understand that the real need is to centralize and manage “business policies” such as pricing discounts, privacy preferences, etc. In turn, this will require major attention by the vendors to such issues as business process management/workflow API compatibility – i.e., BPEL compliance alone is not enough to enforce BPM compatibility. See also “Dysfunctional Data Hubs - 2006-07 Strategic Planning Assumptions for CDI-MDM and Business Services” January 19, 2006. - Sophisticated Hierarchy Management
The management of data about organizational structures such as subsidiaries, business units, and sales regions, etc. is a complex and costly endeavor for all businesses but especially critical for businesses selling to other businesses. Because it is difficult to "know" these complex customers and to consolidate information about the business relationship, “hierarchy management” is fundamental to such CDI business initiatives. CDI and MDM are inherently about managing the “relationships” among parties such as customers and suppliers. While is somewhat reminiscent of the OLAP data mart trends of years past (everyone creating their own slice of the data with a drillable view based on their dimensional views), such “hierarchy-by-anarchy” philosophy counters the basic premise of MDM wherein enterprises are looking to capture, manage and leverage “standard” master data and relationships (hierarchies). Nonetheless, it is vitally important that a 3G MDM solution support flex-hierarchies to both manage relationships across data hubs (different ERP general ledgers for example) as well as import and map to industry standard hierarchies such as D&B’s DUNSRight legal entity hierarchies. See also “Customer Data Integration: MDM Milestones, Part 2” March 2006 DM Review. - High Availability Identity Management
The essential identity matching and linking capability is necessarily high-RAS (reliable, available, and scalable). This is because the ability to positively identify a customer, for example, is critical in online financial services transactions. Another example would be the initialization of a call center inquiry based on the inbound phone number used to dip into an ANI database and determine the caller’s account number, credit rating, next-best offer, etc. – all at speed-of-thought. In today’s increasingly 24x7x365 business climate, an enterprise cannot go offline for maintenance nor suffer outages when it comes to identifying customers or suppliers – or when serving up pricing policies, etc. A vital distinction is that this functionality is more than match/merge capability as such identity matching capabilities must also materialize and match on hierarchy and other relationship attributes. - Data Governance-Ready Framework
Every MDM vendor and consultancy seems to have gotten “data governance religion” yet there remains the huge disconnect between methodologies and processes with the actual MDM software that enforces such MDM policies. To be generous, MDM solutions evaluators need to acknowledge this age old metadata problem still challenges the software vendors, however, 2H2008 will see pressure on the mega vendors to provide at least a “lite” methodology with integration to the underlying MDM software stack given that such solutions will be offered from best-of-breed or 3rd party vendors. See also “Corporate Data Governance - From De Rigueur to De Facto to De Jure” November 2005 DM Review. - Registry, Persisted, & Hybrid Architecture Flexibility
Architectural forms vary in terms of the amount of instantiation of master data – varying from a fully virtual registry style cross reference index (e.g., web portal) to a fully persisted physical data hub (e.g., enhanced operational data stores). More common is the hybrid model which allows both in one solution to better support operational, analytical and collaborative “use case” MDM styles.
Why “Multi-Entity MDM”? Why Now?
During 2008, “party” and “product” data interdependencies will quickly broaden MDM requirements across data domains and the relationships among them – i.e., from “customer” to “product” to “vendor”. See Figure 2 – “The PARTY:PRODUCT Conundrum” for an overview of the attributes that are commonly shared between “party” entity and “product” entity. Currently many MDM solutions vendors are focused on one or the other major entity – hence our use of the descriptive term conundrum to describe this riddle.
Figure 2 – The PARTY:PRODUCT Conundrum
Additionally, the future MDM landscape will be influenced by these “multiples”:
- Multiple data domains
- Multiple relationships
- Multiple usage styles – analytical, operational, and collaborative
- Multiple views/roles of master data consumers including linkages between operational data domains (e.g., using collaborative or analytical MDM to integrate the processes)
- Multiple phases of implementation that are suited to the enterprise’s unique needs -- ranging from quick time-to-value “registry” solutions to full function “transactional hubs”
Clearly, the future direction is to also grow all reference masters into operational masters, e.g., pricing and location style masters into transactional support roles via operational, analytical, and collaborative MDM linkages.
Through 2008-09, Global 5000 enterprises will broaden their MDM business initiatives from single use case, single entity to multi-style, multi-entity. By 2010-11, enterprises without a long-term multi-entity MDM strategy run the ironic risk of building “MDM silos” which will need to be patched/fused together via middleware – in effect, recreating the original MDM problem associated with master data segregated and isolated within ERP and CRM instances.
Figure 3 – Myopic vs. Strategic MDM Road Map
What are the Necessary Features of 4th Generation MDM Solutions?
During 2008-09, both mega vendors and best-of-breed MDM vendors will not only have embraced and delivered the five key 3G MDM capabilities but will also be well on their way to the next generation of MDM solutions. These 4G solutions can be characterized as “full spectrum” hubs due to their support for both structured and unstructured information.
Additionally, we expect to see greater emphasis on extreme “enterprise scalability” while concurrently delivering “master data search” capability. The latter is a relatively new MDM ecosystem category furthering the utilization and ROI of such enterprise information management (EIM) by incorporating “search” for both structured and unstructured info across a variety of applications such as catalog management, deep web search, and enterprise search.
Our five key “DNA markers” for 4th generation MDM solutions focus on:
- Multi-Entity MDM
An MDM solution will need to concurrently manage multiple, diverse master data domains (customers, accounts, products, etc.) across intra- and extra-enterprise business processes. Moreover, a 4G solution will provide the capability to expand on relationships among entities – i.e., evolve from the 3G requirement of a single primary data domain and other supporting domains into becoming the system of record for multiple data domains and relationships among them. - Multiple “Use Case” Styles
4G MDM solutions should support all users and usage requirements for master data – e.g., different functions to define and create data (collaborative), use and maintain (operational), and derive insight (analytical). Additionally, this implies multiple deployment capabilities including the ability to start as an index for one domain and grow into full multi-entity over time. - Process/Policy Hub Architecture
Clearly, BPM workflows are critical to achieve value from MDM and to ensure that the outcome of such data governance infrastructure is actually orchestrated across business units and master data hubs. Just as clearly, there are major ROI and other benefits from centrally managing such policies within a single trusted policy/process hub. In short, 4G MDM will support the linkage of MDM styles into the actual business processes. - Integrated Data Governance
While relentless near term business drivers (such as compliance in Financial Services) are now requiring enterprises to institutionalize data governance, the longer term goal is to integrate, measure and manage data governance metrics within the context of the master data lifecycle. Clearly, effective data governance is integral to delivering reliable and usable MDM to develop master data as a corporate asset. - Enterprise Search & Support for Unstructured Info
During 2008-09, semantically-enabled metadata will enable “search” for both structured and unstructured information across a variety of applications such as catalog management and deep web search, and enterprise search. By 2009-10, enterprise semantics and SOA-enabled data services will further provide the technology foundation for policy hub. While the majority of contemporary MDM solutions focus on the structured data held in CRM and ERP applications, the reality is that a plethora of valuable customer, product, supplier, employee, etc. information resides in what is characterized as “unstructured” information, e.g., emails, instant message log files, voicemails, etc. To provide a robust “universal customer view”, etc., it is clearly desirable to incorporate these valuable information sources as part of the composite view.
Which Vendors Have Found “Multi-Entity” Religion?
Most vendors in fact approach MDM from either specific usage/domain pairing, or broad MDM tool – e.g., rudimentary MDM data model and set of tools for data quality, workflow, etc. to build own MDM product.
Currently, the mega vendors for the most part offer one center of gravity or the other. In addition, by offering the full “application-like” functionality of “master product catalogs” or PIM data hubs, these mega vendors (IBM, Oracle, SAP) both benefit from and suffer from the increased power and complexity of their approach to multi-entity MDM (relative to best-of-breed vendors such as D&B/Purisma, Initiate Systems, and Siperian who do not provide PIM-like application data models or processes)..
- DataFlux. While DataFlux users report that they are able to flexibly deploy multiple entity types with the models provided, we believe that the near term focus (rightfully so) will be on data governance processes that augment the other hub vendors’ lacking capabilities in this area.
- D&B/Purisma. The current D&B/Purisma offerings focus more on the integration of enterprise data with B2B hierarchy data than supply chain data. With the usage of multiple attributes with its hierarchy management, Purisma users have broadened out beyond B2B and B2C party data into location mastering, etc. but do not reach across the chasm to supply chain or manufacturing data (yet).
- IBM. With the February 2008 release of InfoSphere MDM Server, IBM has blended certain aspects of its MDM acquisitions (DWL and Trigo) such that the data models, the SOA services, and underlying middleware all share a common stack that enables both “party” and “product” to reside co-equally. Concurrently, however, IBM WebSphere Product Center will continue on its own line of development due to the nature of PIM requirements.
- Initiate Systems. During 2008, Initiate Systems launched their Master Data Service for certain new products along the party dimension while also claiming deployments of product and location type data (vehicle, incident, and location). Their focus is on the high performance matching algorithms and quick time to value of the registry model. Once Initiate Systems undertakes its IPO and attains its revenue objectives, accordingly the vendor will recalibrate for the long haul and focus R&D on process hubs, data governance, and other areas where the market communicates its needs such as multi-entity MDM.
- Microsoft. Through YE2008, Microsoft MDM will remain essentially the Stratature Enterprise Dimension Management product as Microsoft acquired last summer and will continue to focus on the low end of the enterprise spectrum albeit with a solid multi-entity foundation.
- Oracle. Oracle on the other hand has a native Product Information Management (PIM) Data Hub that provides solid synergy to its counterpart Customer Data Hub, and also has from its Siebel acquisition a Universal Product Manager solution that is a B2C customer-centric view of product (as seen from the Siebel Universal Customer Master solution). The combination of CDH and PIM Data Hub provides multi-domain MDM in large part due to Oracle’s Global Single Schema they reside on. Additionally, Oracle’s Site Data Hub is based on this schema and set for 2008 release. Additionally, Oracle provides a road map for 2008 that includes Fusion MDM as a semantic layer to unify both the data and process models across its MDM family (Oracle CDH, Siebel UCM, Hyperion Data Relationship Management, Oracle PIM Data Hub, etc.) to deliver multi-entity MDM across all its MDM products.
- SAP. For the present, SAP offers a product-centric MDM solution (SAP NetWeaver MDM v5.5 sp6) with its “customer” model being the product-centric view of “business partner” B2B rather than the pure play B2C customer. Version 7.1 (planned for 4Q2008) will remove the party model limitations (“N” level hierarchy) and give SAP full multi-entity capabilities for operational MDM.
- Siperian. Siperian has the deepest multi-entity penetration rate among its deployments based upon our interviews. While its approach to MDM mastering capabilities appears to be a tool kit at first brush, upon closer investigation, the Siperian MDM Hub multi-entity capabilities currently address customer, location, and product data while providing both data model flexibility and BPM/workflow integration across these entity types. Rather than provide the full CDI or PIM application functionality of IBM, Oracle, or SAP, however, Siperian provides an advanced distributed architecture and faster time-to-value.
Bottom Line Redux
With a 4th generation MDM platform, an enterprise will be better able to:
- Identify and provide differentiated service to its most valuable customers via their relationships (households, hierarchies); also cross-sell and upsell additional products to these customers
- Introduce new products and product bundles more quickly across more channels to reduce the cost of New Product Introduction (NPI)