Process Data Management: Like Your Brain And Your Heart, BPM and MDM Can’t Survive Independently

 

Clay-Richardson By Clay Richardson & Rob Karel

 It's an age old question:  Which came first:  the data or the process?  Okay, not an age old question, but an interesting one to ponder nonetheless!  Of course the answer depends on who you ask. IStock_000009712913XSmall  Ask most Business Process pros and the reply:  "Process existed even before computers. In fact I saw on The Discovery Channel where they found a process map in King Tut's tomb. No joke..."  Ask a Data Management pro and you're likely to hear: "Data was invented by the Pharaohs - you ever hear of a little thing called 'numbers'?"  

In upcoming research we tackle this vexing question and conclude that process and data are of course inseparable (shocking we know).  Unfortunately, most Business Process pros, Data Management pros, and vendors we interviewed couldn't see the forest for the trees.  In fact, in our recent MDM Survey, only 11% of respondents said that their MDM and BPM initiatives share the same cost center and team members that work together on a daily basis to develop solutions for the business.

In many ways, master data management (MDM) and business process management (BPM) represent two different sides of the same coin when it comes to business optimization and transformation.  So, why is there so little collaboration and interaction between Business Process and Data Management professionals?  There's plenty of blame to go around:

  • Business Process pros don't find data sexy. While most process improvement initiatives pay lip service to data quality, only a small handful take data seriously and incorporate data modeling and data mapping into upfront discovery efforts. BPM teams typically focus on process first, user experience second, and data is an afterthought - no one ever asks if the data is clean until it’s too late!
     
  • Data Management pros value clean data over process context.  "If we create clean data, then they will come" is often the mantra of Data Management pros. Unfortunately, the person coming is the CFO - asking "why are we spending so much money on something when we can’t articulate its business impact?"  In other words, Data Management pros often fail to provide cross-enterprise context for MDM.
     
  • BPM Suite vendors conveniently MIA on the issue.  Some argue that SOA is the silver bullet to bridge the gap between MDM and BPM initiatives.  While SOA can definitely address reusability and architecture issues, many BPM teams come to realize the importance of master data too late into their initiatives.  While BPM suite vendors provide basic tools to define process variables, only a few (Lombardi/Siperian, Appian, and Polymita) accept some responsibility for keeping process data in sync with master data sources.
     
  • MDM vendors focused on data governance and stewardship only.  The good news is these MDM vendors do in fact recognize that business process plays a major role in their success. The bad news is they are only actively addressing one side of the problem: the business processes and stewardship roles that govern the administrative workflows within the MDM tools themselves. The MDM vendors are not actively influencing the business processes that consume and depend upon the master data these vendors provide.

Although process and data professionals may not see it, like your brain and your heart, BPM and MDM are interconnected and one cannot survive for long without the other.  Our upcoming research, "Warning: Don't Assume Your Business Processes Use Master Data," highlights the fact that process improvement initiatives face a vicious cycle of deterioration and decline if master data issues are not addressed from the outset. And MDM initiatives face an uphill battle and certain extinction if they're not connected to cross-cutting business processes that feed and consume master data from different upstream and downstream activities.

Earlier this year, Forrester highlighted the emergence of "process data management" in our BPM Tech Radar report.  This new category in the business process landscape organically merges BPM and MDM disciplines and capabilities to provide the enterprise with one version of "process and data truth".  

Over the next several months, we will publish a body of research that scopes the process data management challenge, provides best practice for combining process improvement and data management initiatives, and presents case studies highlighting teams that have successfully integrated these two critical capabilities.

Why It Matters?

Call it what you like, the consolidation of process improvement and data improvement is inevitable.  Organizations looking to minimize operational risk and boost adoption on BPM initiatives must incorporate data modeling and data management activities for cross-functional process improvement projects.  Additionally, organizations looking to increase the visibility and importance of master data must identify and harness the most critical business processes that generate and consume mater data.  These are just starting points for bringing process and data closer together.  Most importantly, senior management must foster collaboration and provide cross-training between these two siloed disciplines to begin this long overdue paradigm shift...

What’s Your Take?

We want to hear from you.  Let us know whether you think there is a real need to synchronize MDM and BPM activities? Do you think BPM vendors should accept some responsibility for keeping process data synchronized with master data sources? Alternatively, do you think MDM vendors should be doing more to educate the marketplace? Also, let us know if you have a process data management success story, or lesson learned that you want to share.  Post your thoughts in the comment section or feel free to shoot us a quick e-mail at crichardson@forrester.com or rkarel@forrester.com.

Comments

re: Process Data Management: Like Your Brain And Your Heart, BP

There's lot of truth to what you say - the integration of MDM and BPM has been well defined in the inbound - the creation of master data. But in the outbound - synchronizing master data (from the MDM hub) with downstream systems - is BPM the answer?Inbound: There's no argument that that master data needs to be reliable. While the MDM hub automates the merging of a large volume of duplicates, the exceptions need to be handled by the data steward in "collaboration" with the data owner/ business user using "BPM/ Workflow". We (Siperian) is seeing a several customer use cases in the creation of master data. Read this blog to learn further - http://bit.ly/j4Juf.Outbound: We have a couple of use cases here - data synchronization or data provisioning. Once the master data is created (in the MDM hub), you can do a couple of things - (1) Synchronize the reliable master from the hub with downstream systems (e.g. CRM/ ERP) that have incorrect master data. With the now reliable master data, the business process within these systems can function better (like getting an engine tune up). (2) Provision the data into a new system based on some business rules. E.g. when a new customer is created in the SFA system (which will make its way to MDM hub), and if the customer has more than a $1B in revenue (business rule), you might want to MDM system to populate the new customer into the Marketing Automation system (data provisioning) to perform direct marketing campaigns. While you can do this using BPM, do you need it? We have customers doing this using EAI, instead of BPM.Clearly, BPM has role to play in the inbound use case for human interaction workflow; but, in the outbound, where system-to-system interaction is required (without human interaction), BPM seems to be losing to EAI. That's our observation.

re: Process Data Management: Like Your Brain And Your Heart, BP

You hit the nail on the head when you day "the consolidation of process improvement and data improvement is inevitable." Look forward to your upcoming posts on this topic.My guess is that MDM will be the discipline that moves quickest to the common middle. Newer data applications seem to be built more around business drivers.... we'll see. Was reading in a recent article where Navin Sharma at PBBI even took it to the extreme that data governance is everyone's business. http://ebs.pbbiblogs.com/2009/08/03/data-governance-its-everbodys-business/

re: Process Data Management: Like Your Brain And Your Heart, BP

Hey guys - Well said!This is an issue that goes back to my early days in technology (back in the early 80's).You correctly addressed the number 1 problem: that of specialization. Data folks are trained to ignore process, since it's quite dynamic, whereas data are more stable. Process folks are taught basics of data structures but don't have an appreciation for the nuances in "getting data right."Might I add another contributing problem? It is this: Scope management. I address this in a follow-on post (http://tinyurl.com/y8vg2yl). Our Agile Data Governance method helps both data and process practitioners focus on solving "biggest bang for the buck" kinds of problems, and ensure that neither data or process are left out of the discussions.I believe that as we succeed in Data Governance, we'll see these two worlds come together in very practical, very focused ways. At least that's what we're seeing in our customer accounts.Keep it coming!