Today, data governance is becoming one of the more popular corporate catchphrases. But what’s the real meaning behind it?
Google results will show an array of definitions which can confuse people even more!
Let’s break it down to a more manageable, bite-sized form.
What is it? — A collection of processes, policies, and rules that prescribe the management, handling, and access of data to uphold data quality. It encompasses the planning, implementation, and embedment of governance practices across the organisation.
Why do we have it? — By having high-quality data, people can trust and use it to gain eye-opening insights that inform their business decisions, support strategic goals, and drive companies to be more competitive and innovative.
How is it done? — Data management and governance solutions are used to support and execute activities within all stages of data governance. For example, defining business rules during the planning stage, profiling and cleansing data during the preparation and implementation stage, and active monitoring and remediation during the embedment stage.
It’s recommended to start with the governance of master data as it’s the building block of your operations and provides overall business context. Then you can move on to transactional data, reference data, etc.
Most forward-looking organisations have been striving to implement data governance. But a study showed that 80% of them have either failed or are struggling with it.
Some might think that it’s the strategy and planning stage that makes or breaks data governance, thus underestimating efforts for subsequent stages. It’s no surprise that failed data governance initiatives often originate from these components.
5 major Data Governance pitfalls
1. No involvement of business right from the start
Oftentimes, people mistake data governance for a purely IT-led initiative. This stems from the belief that data quality, data analytics, and everything else data fall under the purview of IT. So, it’s commonplace to see the CIO and their IT team going it alone, even at the stage of formulating the strategy and plans.
True, they’ll have the technical experience and skills like choosing the suitable technology solution, integrating it with the existing system landscape, and implementing data-related activities.
But let’s take a step back and think for a moment about why we have data governance in the first place. To improve data quality, specify how data is used and accessed, and most of all, turn data into valuable insights. And who will benefit from all this? The business, of course!
So, for your business people to benefit from data governance, they’d need to be involved at the earliest stage to align the definition and application with their respective job functions. They can provide inputs on foundational matters like prioritisation of data areas to govern, data quality rules and validations, who should have access to which data sets, data retention strategy, etc.
Make no mistake—The involvement of your business people is crucial to add more meat (and teeth!) into the data governance strategy and planning, which in turn will determine its overall success.
2. Absence of clear goal-setting
Most people are aware that data governance isn’t the endgame; it’s a journey unique to each organisation. But this mindset might backfire if it’s set in motion with no ties to business and operational goals.
In the absence of goals, your data governance would become a corporate idealism, not producing tangible outcomes and in danger of being perceived as useless and inconsequential!
Just like any other goal, those you define must be SMART—Specific, Measurable, Achievable, Realistic, and Time-bound. For example, you can focus on governing Supply Chain master data as your first goal and define milestones leading up to it before moving on to other areas. By building a list of manageable and successive goals, you’ll have better control in producing desirable outcomes.
It’s also worth measuring and keeping track of the business KPIs that are in direct correlation with embedding governance. Once your Supply Chain master data is managed and classified, you’ll have better visibility of materials and spares in inventory and can work towards optimisation. You can then measure the resultant cost savings and productivity improvements and share them across your organisation.
By highlighting this success story, you can convince more people of its importance. What’s more—you can secure support and funding from your top management to proceed with the initiative.
3. Poor follow-through on execution
While it’s true that a carefully laid-out plan that engages the right people leads to smoother execution, there’s also the matter of sustaining it.
Data governance is embedded via refreshed mindset, modified behaviours, and new ways of working. Involving business users at the start of the journey is only half of the battle won.
But if they’re expected to see through the execution by just referring to pages and pages of guidance documents without a solid support and monitoring structure, you risk losing their interest. With people on the ground left to their own devices, your data governance initiative gets left behind too.
That’s why you need to have change management to ease the transition of your organisation towards embracing data governance.
This starts with well-thought-out communication and engagement plans to reach out to people from different organisation levels, prep them for what’s coming, address their doubts, and get them onboard.
To address the practical part, you need to identify suitable people for data-related roles; those with the right authority, experience, and skillsets. It’s wise to map the roles against a RACI matrix, a.k.a. Responsibility Assignment matrix. This way, you’ll have clear visibility of roles that should be held accountable and responsible for ongoing data governance and data quality efforts.
Data owners should be appointed for each business area to define the corresponding process and rules right from the start. Their involvement doesn’t stop there; they need to ensure that the rules and policies are updated to suit current business needs. Mapped against the RACI matrix, they are the accountable ones.
The role of data stewards is more operational where they execute the data quality rules and policies as well as cleanse data when the need arises. In RACI matrix, they’re those responsible.
Another thing that’s often taken lightly is documentation. While it’s widely understood that documentation tells you how to do your job, change management should take it a step further.
Documentation should elaborate on the rules of engagement, rationales for implementing business rules, deviations, and exceptions—essentially the whats and the whys. This helps ease your people into new ways of working, understand their roles better, and sustain the practices.
4. No awareness of the benefits
Building upon the previous point, people generally don’t like to be told what to do. It IS human nature!
You could end up having more disgruntled and disengaged employees despite side-stepping the earlier pitfalls. Part of your engagement plans should include focus groups to explain to people the importance of having data governance. While they may have a clear idea of the benefits to their organisation, it may not be enough to inspire action in them.
You should articulate benefits that resonate with them, ones that answer their inner burning question “What’s in it for me?”. You’ll be surprised that a personal reason such as ‘reducing stress and depression caused by bad data’ could be a legit motivating factor to imbibe data governance in day-to-day jobs.
5. No support and commitment from top management
Instilling any kind of organisational reform such as data governance should start from the top.
It’s not enough for top management to give the go-ahead and continue doing business as usual. They should walk the talk too.
People need to see first-hand that the executive leadership team uses data in everything they do—be it examining a single KPI metric or strategising the company’s future roadmap. By demonstrating this behaviour, it’s easier for them to rally people to get on the data governance bandwagon and apply it within their job functions.
It’s also a strategic step for top management to appoint data champions. They are responsible for promoting data governance and building awareness of its importance and relevance within the organisation. Most importantly, data champions should be able to identify and remove barriers that hinder its implementation.
Do NOT downplay the role of data technology
Yes, all these pitfalls allude to people and processes. But you shouldn’t underestimate the role of technology in solidifying the success of your data governance implementation and sustaining it.
Suitable data management and governance solution serve to facilitate and expedite data governance activities. Some examples are automating master data processes, doing data profiling and cleansing, and monitoring data status and health.
Without it, your people will have to spend time and resources doing these tasks by themselves. It’d be like a cottage industry of spreadsheets and in-house apps to manage and govern data. Over time, these siloed systems and disorganised ways of working will spiral out of control, creating chaos, stress, and unhealthy work environment.
Manual/semi-manual work will lead to more errors and people have to spend more time checking and rectifying them. This will divert their time and attention from doing strategic tasks such as encouraging data adoption and usage among their peers.
It’s hard for data governance to thrive in this situation.
Look no further than MDO
Master Data Online (MDO) is your ideal data management and governance platform. MDO doesn’t just automate and simplify your master data processes—it helps you implement data governance and execute the embedment activities with minimal human intervention.
It adopts a comprehensive governance framework that encompasses all data scenarios. Active governance ensures your data is cleansed, deduplicated, and enriched before entering your system. While passive governance supports ongoing validation and remediation of data within your system.
MDO helps you implement rules-based governance policies, enhanced with approval workflows and audit trails. This drives data ownership, builds trust in data, and increases its usage across your organisation—essentially the goals of data governance.
Look no further than MDO to complement and support your data governance journey.
Author: Shigim Yusof