We normally talk about the 3 dimensions of an Information management or an MDM initiative – People, Processes & Tools.
People represent the “who”; organization, groups and individuals who create, update or somehow consume data from the Information Management platform.
Processes represent the “how”; i.e. business processes for initiating, reviewing, approving and finally synchronizing with the target systems.
Tools represent the “where”; i.e. the technology platform, including the user interfaces, the workflow engine, data model, interfaces and such.
What is probably omnipresent, but still doesn’t stand out is the 4th dimension of ‘Data’ itself.
So, Why talk about Data as a ‘Dimension’ – after all, the IM or MDM project is for managing the data !!
Here’s why…
First, the philosophy :grin: - Analogous to the ‘Time’ axis; Data provides Life and Motion to an IM / MDM program. Without ‘Data’ it’s just a skeleton.
Now.. Although ‘Data’ tends to be implicit to other three dimensions; it deserves a ‘dimension’ of its own to emphasize the significance of handling it correctly.
How we handle ‘Data’ could be the factor that sets a ‘Transformation Program’ apart from a 'Project Implementation'
Following are probably the most significant aspects of data that need to be considered when beginning a transformation journey:
- Data Scope – Define the principle for scoping the data like a ‘Reagent’; and test-run the data through it. If it doesn’t pass the test, think really hard about all the points below before including it in scope. Think about a Centralization scenario; it is a must to ensure unidirectional flow of data between the hub and target systems – which in turn means that each data item in scope must be part of integration and must enforce edit restrictions in the target system. Correct scoping provides the right focus and the right emphasis.
- Data Standardization – Non Standardized data is a nightmare, but unfortunately a reality in global organizations, especially in those that are a result of large scale inorganic growth. Non Standardized data results in multiplied complexities in data model, business logic and even data maintenance. Organizations that are able to drive standardization of data can expect to have a smoother run in achieving the objectives of an IM / MDM program – i.e. with limited friction in areas of establishing data standards, processes, turn-around-times etc. Also think about multilingual requirements – Is the platform and the data needed in multiple languages. Non Standardized data could also mean Non-standardized user training.
- Data Volume – To start with, Volume of data determines the sizing of system. However, further down the line it could multiply the complexities involved with Data Consolidation and Data Standardization; directly influencing the effort and duration. Another ‘not so subtle’ difference that Volume can make is with respect to the design of user interfaces. As an example : Think about searching Five thousand records vs. searching Five Million records.
- Data Consolidation – Consolidation is probably the most important activity when trying to achieve business transformation. Again, a very sound ‘definition’ of the data entity is needed. A definition, which is not a few phrases or sentences; but specifies how the data entity fits into the organization’s ecosystem and is in-line with the transformation objectives. As an example: Think about what uniquely defines a Supplier – is it the Name, or Name & Location, or Name & Location & Relationship. Subsequently, think about the tools that will help define business rules and for carrying out Consolidation activity. Specialized tools that can integrate with the IM platform will make things easy but the cost must be measured against the benefits.
- Data Integration – data integration will deliver some of the lowest hanging fruits – e.g. reduced cycle time for data processes, reduced data errors, elimination of data inconsistencies & single point of data entry – not to say in any way that these are less significant. In fact, without these foundational benefits, business transformation would be unachievable. What is important to note here is that the complexity of integration increases if the target systems have different data model. One way of determining this is by comparing the definitions of data entities across target systems.
- Reference Data – reference data works like a conduit to guide the use of only valid values for certain fields. Reference data standardization is the cardinal rule for any data item that will be used globally across multiple systems. For local data items; not having standardized data would only lead to increased complexities in the system – just like for any other data; however, Standardization of Reference data poses a bigger challenge due to the fact that it has be down at the data values and not just data fields.
It must be realized that there is no single formula that will overcome data issues in a transformation journey. However, I hope that this will help trigger thoughts around the key aspects and help to better prepare for managing the challenges.
Have you encountered any other Challenging data aspects?
