Data leaders often talk about data governance’s importance, but not everyone makes the connection between governance and the data foundation that makes it possible. That’s reliability. Data must be reliable in the sense that whatever metric you define, the value is reliably correct for it.

Once you have data reliability, the next level is the actual governance. That is not about data being exact as much as it is putting everyone on the same page and sharing a common set of KPI definitions. In terms of its effect, it’s also about setting your organization up to deliver business impact and become more successful. That’s another connection more data leaders need to make. 

In the pre-SaaS world, the goal of governance used to be called a “single source of truth”. Nowadays, I think it’s more accurate to say you have a common agreement and processes in place around what can be reported or not. Whenever you’re in doubt, there should be an official company dashboard so that any term or concept is easy to locate. 

Here are three ways that good data governance leads to business impact and a more successful company.

  1. It simplifies and creates clarity.
    An example I like to use here is a company where a potential investor was visiting and asked how many customers we had. This would seem to be a basic question, but five different people offered five different answers. By drilling into this a bit, we were able to surface that:

                  - Some people defined customer by the number of different customer emails in the database
                  - Some realized that employees also were being counted and discarded them from the total       
                  - Others removed companies that had a 100% discount on our product as beta testers, and only
                    included those that had made at least one payment

    We immediately captured and documented the best definition in a data dictionary that became accessible to everyone. As data leaders, we also communicated clearly that only one definition of customer and one dashboard of customers could be official. To address the reality that people would still want to use the term “customer,” we simply worked on the nomenclature. Maybe a beta customer who wasn’t paying for our product was now a Collaborator instead.
     
  2. It gives data culture true purpose.
    Most data leaders would agree that data governance is a requirement of good business practices. Yet some still look at it in isolation and don’t see its connection to data culture as a means to build impact. 

    Yet data culture is an excellent tool to help people speak the same data language around terms and definitions. Data culture allows people to back up their intuitions and validate their ideas with objective information, based on common definitions of, say, customer, inventory, or sales-qualified lead. 

    Data culture, in turn, is not an end in itself but a way to produce business impact by delivering more value to customers who are willing to pay for it, and who ideally become the next net promoters of your company.
     
  3. It makes decisions data informed.
    Why do I prefer data informed over data driven? Because it's still very important that everybody across an organization can think creatively and produce their own hypotheses. You shouldn't wait for data to give you The Answer as if it were a crystal ball. At my company we expect our people to propose ideas, then feel compelled to validate these ideas and intuitions with hard evidence provided by data. Without agreed-upon definitions, a creative idea has no firm foundation to prove out its hypotheses. 

    You can see this phenomenon at work watching a mature data user testing out an idea. They don’t use data to prove the theory, but to try to disprove it. When you can no longer disprove your hypothesis in any way, you know that your idea is solid. Without the guardrails on your thinking that good data governance provides, confirmation bias is easier to fall into, which can in turn lead to disastrous business strategies.  

The perfect place in the hierarchy
I may be overstating data governance’s importance for effect, but I’ve seen two ways that companies fail: one is due to technology problems, and the other is governance issues. In order to scale successfully, companies need the right data technology, the right data governance process, and the right levels of accessibility for everyone.  

When data governance works well, it creates a powerful foundation. But to reap data governance’s benefits we should think in terms of Maslow’s Hierarchy of Needs. With reliable data and data governance, you have the building blocks you need to rise toward the top, which I define as predictive and prescriptive analytics and AI. I would imagine a successful data leader is progressively making themselves redundant in each of the stages of the pyramid so that they can move up to the next one.

Ultimately, then, the positive value of data governance operates at the level of business impact as well as personal impact. If you can make data governance run well, your company is far more likely to succeed, while you are able to move upward toward more interesting challenges, stronger self-esteem, and professional self-actualization.