Sunil’s report examines the kinds of tools that can support governance, based on his extensive experience as a Data Governance consultant.
Henrik’s post finishes with the questions “As data governance is very much about people and processes (bold added) and not so much about technology, do you need a tool at all? If you do, do you need a separate best-of-breed tool for the data governance part or will it be preferable to have it as an integrated part of the MDM solution?”
In my opinion the more relevant question is “Can data governance survive manual processes?” Tweet This
In a comment on Henrik’s post I suggested that you don’t need a tool for washing clothes – you can wash clothes by hand. It takes a lot of time and it doesn’t always get out the tough stains but it can be done.
A washing machine does the job better, and easier.
Similarly, appropriate data governance tools, correctly applied add tremendous value.
Henrik extends the question in a later post “Data governance tools: The new snake oil?” referring, once again to Sunil’s report.
Again he observes that “Traditionally data governance has been around the people and process (bold added) side of data management. However we now see tools marketed as data governance tools either as a pure play tool for data governance or as a part of a wider data management suite.”
Let’s examine Henrik’s initial questions in the context of Sunil’s report..
Question 1: As data governance is very much about people and processes (bold added) and not so much about technology, do you need a tool at all?
Sunil’s report focusses on a reference architecture for data governance, focussing on the six data management areas most critical to support governance. In Sunil’s view (and I would largely agree) these are:
- Data discovery,
- Data quality,
- Business glossary,
- information policy management, and
- Reference data management
Henrik’s posts point out the disconnect between these (largely technical) tools and the real life people and process issues that plague most data governance programs. While each of these tools has a role to play, I agree that this is not data governance. We can have data quality without governance, we can have governance without reference data management, and so on.
A data governance tool must address the people and process issues. Tweet this
I have witnessed the birth (and in many cases the death) of multiple data governance initiatives in multiple industries – including financial services, telecommunications, government, and mining. In each case the deployments focussed on structures and processes, and, in each case these deployments have been overwhelmed by the complexity and sheer scope of work required.
In the absence of automation the data governance processes failed in every case. Tweet this
Question 2: If you do need a tool, do you need a separate best-of-breed tool for the data governance part or will it be preferable to have it as an integrated part of the MDM solution?
This question assumes that data governance is only required for Master Data Management. While successful MDM certainly depends on data governance, data governance must extend beyond master data.
Can a master data management suites data governance capabilities extend to manage all critical data?
In most cases the so called data governance workflows provided by MDM vendors focus on simple use cases such as approving the merge or deletion of master records. In my opinion, these workflows do not address the data governance complexities of agreeing and enforcing the appropriate use of data, within context, across multiple business functions. This is not data governance.Assuming that the MDM suite enables proper data governance processes this would be normally bundled with a host of additional MDM specific capabilities that add cost but no real value from a data governance perspective. By definition, MDM tools focus on the management of master data – how does a master data centric tool extend to address compliance issues, to manage transactions, to identify key requirements for BI. and so on?
Similarly, data governance supports compliance, provides the framework for enterprise data quality, and creates the context for big data analytics.
To add value data governance platforms must provide the flexibility to implement commonplace Data Governance operating models to increase and sustain end-user adoption, to report on compliance, and to manage maturity and ROI metrics, and to make the management of master data practical and process driven.This is not the focus of master data suites but rather the domain of the dedicated data governance center.
Discerning buyers must distinguish between the snake oil sales teams, that are jumping on to the data governance bandwagon, and those that have a genuine value proposition.
A dedicated data governance center should be on the road map for any enterprise. Tweet this
It may not be necessary at an early stage, however information and meeting overload will play havoc with even the best manual processes. A data governance center is necessary to manage this overload and keep your data governance program alive.
Sunil’s report provides some context for companies looking at ways to automate data governance complexities. Other useful resources are Bloor Research’s Data Governance Market Update and the MDM Institutes’ Field Report on the Data Governance Centre