Unlocking the Potential of Data: The Importance of Business Involvement and Governance
Are you struggling to get your company’s data initiatives off the ground? Your data projects are asking for more inputs and more commitment from the business?
It’s a common problem. Even if your company is fully invested in becoming data-driven, it can be tough to get the key business stakeholders on board.
You might have all the tools, expertise, and resources you need, however if the decision-makers in your company don’t see their role when it comes to data, it can be an uphill battle.
So, how do you get the business involved?
In this article I will present why business involvement and data governance are essential, and some of the key functions needed to have data governance running.
The overlooked role of data governance
I know, the term “data governance” can be a bit of a buzzword and it’s often misused or misunderstood. Many confuse governance with control.
However governance’s primary aim is not about control. Governance is about optimizing the realization of the value of an asset. Yes, you will need some level of control to ensure that this realization happens. But first governance is defining, or finding, what the optimum is. And if we focus on strategic governance subjects the key questions are: what is worth building or doing with data? what would make a real difference, and bring strategic opportunities? or really facilitate operations?
“Governance is about optimizing the realization of the value of an asset.
Finding this optimum requires communication. And data governance, or better data governance, is exactly what you need to facilitate this communication between the people bringing new technologies or ideas, and the people who can evaluate the business impact of an innovation, and eventually adapt to actually use it. Governance must take all aspects of the equation in account: value and opportunities of course, but also risks and costs. And among the costs, besides financing, we have the fact that business stakeholders will have to accept changes: business processes, tools, maybe activities and roles will have to be adapted in order to integrate the change and realize the expected value. Your initiative will often depend on enhanced data capture and quality. But data capture, data quality and curation most often depend on business processes, organization and expertise. The ultimate determination of whether a promising idea presents a real opportunity is thus almost always made by business stakeholders.
You should now see better why business involvement and data governance are at the heart of many issues related to data initiatives or projects: Data is typically managed by technical experts, mostly IT people. The business stakeholders see data as an IT-managed subject, most often as an IT problem.
So IT experts are given the task of carrying out data initiatives or projects. Without proper governance, you will often reach a situation where they cannot get business impact, even if they invest time and do their best to deliver and communicate. If their work did not hit the right target, the point where business value clearly outweighs the costs as well as the challenges associated with change, or if this is simply not the appropriate time or place for the business to make a change, then it is unlikely that the initiative will be successful.
With a running data governance, business decision-makers are involved upfront. They understand what data can do and bring. They understand the proposition, and can even be its true initiators and sponsors. They trust and value technical experts, knowledge workers, participate in the establishment of the vision, in the definition of the product, and are reliably committed to the projects to facilitate their rollout and success.
Setting the right basis for your data initiatives
I won’t deny it: setting up data governance can be a project in itself, we even talk about a program.
I have conducted consulting missions for establishing the foundation of data management programs, ranging from developing a new organizational blueprint around data management to identifying and addressing organizational and process gaps that impede the introduction of successful governance, as well as providing support to organizations facing challenges in initiating the first steps of their data strategy. I have seen very different situations. Defining the right target governance model depends on many factors, including the company’s complexity and culture.
Ramping up data governance properly will need communication and an analysis of your priorities and needs regarding data.
So I won’t present all you need to set up a running data governance here. I will just try to help understand what it can look like by presenting two key functions of a running data governance.
Covering these functions can represent more or less time investment depending on your company’s complexity and size. But we often observe that even without governance, people are already investing effort to try to fill the gap. Their work is simply made difficult, and frustrating, by the fact that they are missing the organization where they could collaborate with their colleagues to make the difference.
Function 1 — Improve understanding and support well-informed decisions
Informed decision-making requires a shared understanding of the full picture regarding data: the value or opportunity it brings, the business risk which might be bound to it, and the cost. You will thus regularly have to bring several people to the table to get this full picture, technical and business experts, and decision-makers.
Now if you want to facilitate communication, spare some of your decision-makers' time, and ensure consistency and follow-up, you will need to bring data architecture to the process: Data architecture is not only about technology, or even data modeling. Data architecture has a key role to play to bring the business and technical worlds together. Data architecture helps to structure and rationalize the decision-making process, facilitate informed and consistent decisions, as well as share and implement the established plans. It is therefore an essential facilitator of data governance.
Data architecture practice does not have to be complicated. Data architecture is not about tracking all the nasty details of your systems or databases. Data architecture’s main value will come from high-level considerations; subjects that have a strong impact on the way the company will run and realize, *or miss*, its data-related opportunities.
Regarding governance, the data architects have a clear mission statement: bring transparency to the current state and future plans and help highlight data-related impacts.
On this basis, the data architect is involved in prioritizing the project portfolio and validating project concepts. The data architect has a clear view of how data is handled by systems and projects; the data architect supports and monitors project preparation; is typically involved in quality gates. The data architect defines the standards for data management from a technical point of view based on the business strategy. The data architect records and maintains the “data registers” (high-level models, which have a role similar to land registries and maps in land management).
In many companies, you have experts working to cover the data architect role without the name. What is often missing, in this case, is the involvement of the data architect in the decision-making processes, a simple update in the governance process.
Data architecture closes the gap between data and business. Everyone has heard of the success of data-driven organizations. But not so many see the link clearly between data and business. For a modern company, business and data can be seen as the two faces of the same coin:
A company operates by applying processes. A traditional company’s processes feed on raw materials, inventory, energy, utilities, financial resources, human resources, data,… to generate products or services, customer satisfaction, structural improvements, financial results, and so on, including again data.
With digitization, the resources bound to business processes are increasingly dematerialized, dematerialized into data. For example, in the financial sector: you likely rely less and less on cash. Transactions are handled in electronic form, exchanging scriptural money. Your bank card is only a token to hold digital keys, making sure you can be reliably identified and sign electronic transactions. You handle most of your interactions with your bank through an online platform. Banks themselves can trade dematerialized stocks and bonds. Even insurance companies can offer their services through electronic platforms. You subscribe to insurance policies, pay premiums, declare losses, and receive compensation through an exchange of information, data, which settle transactions.
Other sectors such as retail and e-commerce are also heavily dependent on data. Retail companies use data to track customer purchasing patterns and optimize their inventory. E-commerce companies use data to personalize the shopping experience of their customers and improve their marketing strategies. In the manufacturing industry, companies are using data and digital twin technology to optimize their production processes and improve efficiency. By creating a digital replica of their production line, they can monitor and analyze data in real-time to identify bottlenecks and improve overall performance.
Some sectors will always handle physical resources. But still, more and more processes are controlled through information systems, where these resources are modeled/encoded as records of data. With innovations like IoT, the digital twin, AI and ML, the link between the physical and digital world is getting tighter, leading to the point where all your business is getting reflected and modeled by your data, and your business processes eventually consume and generate data.
With digital transformation and innovation, business and data are becoming bound together, representing the two faces of the same coin.
Function 2 — Support the data owner
As you can imagine the time of your key decision-makers is precious. And following-up the proper management of data can be nasty. So to let your data beneficial owners govern, you will need to assign them reliable and efficient stewards.
I like to illustrate the function of data stewards by referring to the role stewards were playing to manage the interests of big landowners. Think about the responsibility: manage tenants, workers, and resources such as land, crops, and animals on a daily basis, financial management of the estate, including collecting rents, managing budgets, and keeping records, or handle legal matters, such as disputes with tenants or other landowners. This was an enormous task. Luckily data stewardship is likely not that complex, but still it is a very important responsibility and an invaluable function to support the owner. The data steward must ensure responsible management of the asset (the data), integrating the owner’s objectives and strategy. Given the importance of the role, the (lead) steward must have the confidence of the owner.
Data stewards manage the data from the provider, or point of entry into the source application, through transformation, to the point of consumption. Data stewards define and enforce the business rules (entry rules and constraints, quality rules, transformation rules, life cycle, data cleansing, usage policies, etc.) and participate in the valorization of the data. A (lead) data steward coordinates with the data owner to make sure the stewardship complies with the owner’s objectives and strategy.
You can consider three types of data stewards:
- Business data stewards: belong to the business department and ensure the application of data management principles by the business (e.g. reference and master data entry, data curation, entry rules)
- Technical data stewards: often belong to the IT department and ensure the application of data management principles by the systems, in particular modeling/constraints, integration, persistence, and data transformation. (You can think of data engineers, data modelers, master data management/data quality experts, BI experts, data scientists, etc.)
- Lead or coordinating data stewards: ensure the responsible management of the data by the data stewards. The lead data stewards ensure alignment with the owners. (I write here lead stewards and owners: Remember, you can manage different data by different people. Responsibility can be distributed by expertise and affinity.)
The important thing to note here is that we need business stewards besides technical stewards. Technology helps a lot, but it can’t do everything, you need business expertise and commitment.
Another thing to point out is that if your company manages its data Today, then you likely have already all the people you need to manage your data. Pressing problems get solved, requiring time and effort from valuable resources. Ramping up governance will help these key resources address the problems more efficiently, and solve the root cause issues where applicable, focusing on business priorities, thus saving effort and time, while maximizing the chances of identifying and seizing the relevant opportunities.
Wrap-up
In summary, getting business stakeholders involved in data initiatives is essential for success. And data governance is crucial for getting key business stakeholders on board. Data governance’s primary objective is not about control, but about finding the optimal way to realize the value of the data. It facilitates communication and collaboration between experts and business decision-makers and helps ensure that data initiatives are aligned with business priorities and needs. Having data governance properly running can be a project in itself, but it will help people who are already investing time and effort make the difference, and bring you the full potential of data.
