Building Data Governance Strategy and Road Map

Data governance refers to the practice of identifying important data across an organization, ensuring its quality, enhancing its value, and making it reusable by people in the organization in an effective and compliant way. My previous article Increase Trust, Value, & Security of Your Data with Five Key Elements of Data Governance discusses the data governance concepts in detail.

By defining the essential components of the data governance program, a data governance framework supports the execution of data governance. As part of this process, it is necessary to improve and manage data quality, identify data owners, create a data catalog, create reference data and master data, ensure data privacy, enforce and monitor data policies, promote data literacy, and distribute data.

The Data Governance strategy and vision should encompass the following critical factors to assess data governance readiness and maturity:

Data governance starts with this vision behind the data governance. This includes a vision statement by the stakeholders backing the data governance program and the objectives of the program.

It is critical that the vision behind the data governance is commonly understood and communicated at the the enterprise level — which means that all the stakeholders are at the same level of understanding of the vision and the objectives of the data governance program.

After the common understanding of the vision is established among the stakeholders, the next step is to define WHAT is needed to make the data governance program successful — The goals and the KPIs

There is a need to consider two important metrics — Risk Management & Compliance Metrics, that would measure the improvements in data quality, security, privacy and data retentions.

The second important metric to consider is what value is being created with the data governance program. This would help to monitor how data governance program is contributing to improve business value through the creation and use of trusted data.

The business values could be:

Other goals could be to have a focus on:

Examine the architecture of the data flows from the first point of contact throughout the entire data lifecycle. And to Identify those areas where data gives the most value, and where data creates the most risk.

Once agreed on the vision, and have set the goals in terms of key objectives and measurement criteria, next step is to plan which KPIs need to be addressed first (i.e, basically addressing the biggest pain points). A good starting point is to identify and describe the business problems caused by ungoverned data to systematically identify candidate business cases.

This way it would be easier to rank the business problems in order of severity and return on investment (ROI) if the problems were solved.

This will also include Prioritising problems where ungoverned data has the greatest business impact in terms of greatest opportunities to drive value.

As a final step, the data governance strategy would then include the projects / initiatives that are needed to achieve the business objectives, meet the targets set and deliver the ROI identified in business cases.

However Data governance cannot happen as a big bang, and it would make sense to use the ‘divide and conquer’ approach through out the data governance lifecycle i.e., starting small as an MVP (Minimum Viable Product) and then expand it to across multiple groups, where each working group is responsible for a particular domain or data entitiy e.g., Customer or a data subject area that consists of multiple entities

For Example, One working group would be working on the entire data governance life cycle for one particular domain or data entity starting from data discovery to understand the data landscape, where the group would be going through the entire process of data discovery, mapping, cataloguing and profiling.

Next, the working group would establish a business glossary to provide a common business vocabulary. This would require the definition of data entities (for example, master data), data attributes names, data integrity rules, and valid formats. The working group would:

They would then be setting up a procedure for automating the application and enforcement of data governance policies and rules, as well as establishing a manual procedure for applying and enforcing policies and rules that can be invoked to govern:

Since data governance is not a one time implementation project, it needs to be continuously monitored and improved- The working group responsible for one domain entity would be monitoring and auditing data usage activity, data quality, data access security, data privacy, data maintenance and data retention and also Monitoring policy rule violation detection and resolution.

The data governance road map helps as a starting point for a green field data governance project. It is an iterative process and would look something like this:

As a first step, understand the short and long term goals of the organisation i.e., what are we trying to address here, why do we need the data governance here? This would help us: define the data governance strategy and align it with the business strategy.

Also to understand the high level inventory of data sources, then documenting the pain points with the stakeholders, trying to learn what are they actually struggling with e.g., with data, processes, lack of resources etc.

First need to develop a program charter i.e., mission, vision, and the scope. Of course we cannot tackle everything, therefore, need to restrict the scope for for-example for 6 months — 1 year.

Also, team building, maybe starting as data stewards, data stewards because they are subject matter experts, they have been with the company for sometime, they understand data, business rules, and maybe already governing the data somehow, so start identifying them. Also at the same time, setup a Data Governance (DG) council. The council decides on the first set of priorities, which could be done after detailed workshops and looking into pain points and objectives — and for starting should be on the one key dataset — smaller scope to start.

Next, it’s important to outline the policies and processes for managing data, and map them to the roles and responsibilities. Clear processes will also be required for data storage, formatting, metadata, security and compliance, for example. It is time to define the metrics and a measurement method. Once we have a framework in place for measuring and monitoring data governance, we will be able to select the right tools for the job.

Also use the defined metrics and a measurement methods to track progress and measure what is working.

As the Goal is to develop a framework for the data governance lifecycle, along with the short term and long term improvements tide with the long term and short term goals of the organisations.

It is important to regularly check the performance of the data governance against the business objectives, in the spirit of continuous improvement. Goal here is to make at-least one change, that could really have biggest improvement impact and to have a further buy-in from the council — that should be all around data quality.

Related posts:

Things can change in 10 years, and they changed. If you want to understand how this got here, these moments are a good place to start.

I see you in an ethereal light. “Wedding Vows” is published by Ryan Wheeler in The Creative Cafe.

Luc mengeluarkan sebuah kue kecil dari keranjang pikniknya. Ia menghirup aroma yang keluar dari kue itu. Setelahnya, Luc mengeluarkan garpu dan mulai menikmati kue buatannya. Cuaca hari ini cukup…