So you’re convinced it’s time to start a data governance program. That’s great! Data is a vital asset to organizations today, and governing data is a strategic necessity (and often a regulatory necessity as well).
But where do you start? How do you launch such a program from scratch and ensure its success?
Steps to Get Started
Here are the six steps you must take to start a data governance program:
- Find your why.
- Gain executive commitment.
- Sell the need for data governance across the company.
- Pick a starting place.
- Establish your data governance program structure.
- Ensure data governance by design.
1. Find Your Why
Data governance is expensive—there’s no way around it. To be frank, all governance is on paper. It adds little to the bottom line, and it requires attention at all levels of the company, including executives. That said, there are definite benefits to data governance.
Before starting on such a journey, be clear about why you’re doing it and what the company will look like as a result. Your reasons and vision frame everything else about the program launch. Executives demand that you demonstrate alignment between the program and the company’s strategy. Your vision establishes that linkage.
No matter how good an idea is, some colleagues will need you to persuade them and sell them on the idea. Clarifying your vision lets you promote the benefits and not the features.
Launching the program will take a lot of effort. That means you’ll need a clear vision to keep the momentum going when things get hard.
2. Gain Executive Commitment
Why do you need executive commitment? Data governance feels like the realm of IT. IT runs and supports the systems, and IT is involved in changes to enhance the systems. So, IT must drive data governance, right?
Business units are the ones that own data, not IT. IT is certainly an involved and committed partner, but the decisions on what data to collect and use within the enterprise are business decisions. These decisions introduce business risk, and executives need to understand and ensure that risk is managed.
Any governance effort lives and dies through its executive support. Governance occupies staff time—in some cases, quite a bit. Some staff will be on data governance working groups, while others will have more formal data governance responsibilities. In addition, data governance policies and standards force updates to IT procedures, which someone must manage appropriately.
In terms of money, some tooling may become necessary to run an effective program. Like all tooling, you’ll license it and then pay its ongoing operational and maintenance costs.
How to Start
Here are the three things you’ll need from executive leadership when starting a data governance program:
- Resources to staff and execute the program, in terms of both dedicated and shared personnel
- Executive leaders’ regular involvement in monitoring and overseeing the program
- A solid promise that managers will adhere to data governance policies and standards within their own departments
Your data governance program won’t get off the ground if you can’t get these critical executive commitments at the start and on an ongoing basis. In particular, you’ll need continued presence from executives on communicating the data governance vision throughout the company and to serve on the data governance program’s overall steering committee.
3. Sell the Need for Data Governance
Once you’ve gained executive commitment, you’re done, right? That sounds good, but it takes a lot more than that to drive the necessary changes.
Instead, use organizational change management practices to sell the entire organizational body on data governance. Don’t underestimate the need to bring the rest of the organization along with you. Since for some (or even many), their roles may grow or expand to support data governance, you’ll need them to be committed and not merely compliant.
Start by communicating the data governance vision from the top of the company, and then cascade it throughout the organization.
Next, hold smaller discussion forums and Q&As with key staff and stakeholders. There, you’ll hear their feedback and concerns and position yourself to address them from the outset.
Last, identify those staff that are enthusiastic, and enlist their services as your champions. In particular, find those staff that have key process or data ownership responsibility already, and actively recruit them to be data stewards, which we’ll cover in detail below.
4. Pick a Starting Place
Chances are if you’re reading this, your company already has lots of business data in use. So much data, in fact, that starting a data governance program to oversee it feels like boiling the ocean.
The overall goal of your program is covering every piece of data you’ve got. But “everything” isn’t a realistic starting point. Let’s start by considering a few candidates.
Candidates for Beginning a Data Governance Program
Here are the top two candidates for where to start your data governance program: most critical business data first, and passionate business champion first.
Most Critical Business Data First
No matter how much business data you’ve got, I’ll bet you that some of it is more critical than the rest. I’m sure right now you can think of data in your company that must be protected at all times because any threat to that data is a threat to the company as a whole. If so, you’ve found a candidate for where to start your data governance program.
Passionate Business Champion First
You’re already working on selling your data governance program across the company, and allies and champions are slowly coming forward. Use early allies who also have data ownership responsibilities by enlisting them as your first data stewards.
Your stewards make up the backbone of your data governance efforts. Since stewards have direct accountability for the business data, you absolutely need their support and involvement. As such, selecting one who has already signed on to your vision goes a long way toward establishing the program structure and carrying it forward.
Regardless of how you choose your starting point, rely on your change management efforts to communicate your progress throughout the organization. You’ll show the seriousness of the launch, and you’ll continue to identify champions and allies.
Data stewards will form a critical part of your data governance program, and they’ll have specific responsibilities. Decisions about the use of data within processes, as well as its safeguards, protections, and risk mitigations, will fall to them.
It’s best to select as stewards those staff that already exercise process ownership because they’ll already have both the incentive and involvement necessary. The steward must be high enough in the organization to have authority while being low enough to understand the processes, data, and risks.
After the executives, your potential data stewards are the most critical in the organization to recruit to your vision because they’ll be the engine that drives your success.
5. Establish Your Data Governance Program Structure
Once you’ve picked your starting point, you’ll need to set up the right data governance program structure. The right structure puts decision makers in appropriate spots to administer, monitor, and execute the program. You’ll need to evolve this over time, so a good starting point is to consider having two levels.
The first level includes some executives plus a critical line of business and IT leaders. This group sets high-level strategic objectives and oversees the program.
The second level contains your data owners and data stewards plus similar managers in IT that handle the typical data governance activities. Sometimes you can rely on one of those members to facilitate the data governance program and coordinate its activities. That said, consider hiring a dedicated data governance program manager solely to manage the program. This is certainly a bigger resource investment, but it ensures you place accountability for the program is squarely on one empowered individual.
Let’s now look at some of the most important activities in data governance.
Key Data Governance Activities
Your data governance team makes official decisions related to data. Here are some common examples:
- A business unit wants to start collecting customer data in one of its processes.
- A new product or service is being launched, and it will require new business data and processes.
- A recent audit or risk assessment identifies business data risks that need to be mitigated.
In the above cases, the data steward who owns those areas must present or champion this work to the data governance team. In turn, the team will discuss and approve its usage. This includes items like ensuring no duplication across units, establishing the metadata and usage of the data and its fields, and outlining the key risks and mitigations that will be needed. Once data governance has approved the decision, only then will IT make application or system changes related to the data.
6. Ensure Data Governance by Design
Governance is hard to do well when it’s tacked on to people’s jobs. Think about it: Everyone in your firm is already working hard, perhaps at or over capacity. Now you want to add a new data governance program and increase the responsibilities of some key staff.
It’s unfortunately common in IT to launch an initiative intended to revamp how IT works. It’s a sensible goal, since IT serves the business and needs to ensure it’s optimized for that purpose. However, great plans go awry when IT underestimates the level of involvement needed from business partners. Or even worse, things go wrong when IT needs business partners to act or behave differently, yet IT begins its transformation without the necessary conversations and commitments.
With these twin challenges in mind, let’s look at how to apply data governance by design to maximize your chances of success.
Data Governance for Business
We’ve already discussed the purpose of data stewards. For those individuals, it’s critical that their job descriptions reflect these duties as part of normal organizational performance management. In addition, you must reflect related data governance duties in the roles of any staff member who’s handling, processing, or managing organization data. These job descriptions don’t have to be elaborate, but you must have the appropriate incentives in place.
Data Governance for IT
Whether your data governance is formal or informal, you’ve already got some concept of IT governance in place. Data governance on the IT side requires you to align both governance efforts. Most commonly, you’ll ensure IT governance approves changes related to data only after data governance staffers have provided input and decision as needed by your policies. For example, if you have a governance team that approves database designs, make sure someone from that group is also on the data governance team and can drive the right conversations.
Starting a data governance program is a long and important journey. You’ll launch your program on the right foot by following the steps above. Check out Why Is Data Governance Important? 3 Major Benefits if you need help selling the vision and benefits of your program. Learn more about the technical specifics around data management by reading What Is Data Management and Why Is It Important?
This post was written by Daniel Longest. With over a decade in the software field, Daniel has worked in basically every possible role, from tester to project manager to development manager to enterprise architect. He has deep technical experience in .NET and database application development. And after several experiences with agile transformations and years spent coaching and mentoring developers, he’s passionate about how organizational design, engineering fundamentals, and continuous improvement can be united in modern software development.