Data governance best practices

So you convinced your business leadership that investing in a data governance program is in the best interest of the company. Now what? While embarking on a data governance program is an exciting time for any enterprise data management team, it can also be a big undertaking. With little example to follow, those beginning a data governance program, and even those who have implemented one recently, would be wise to follow industry best practices and learn from some of the cautionary tales out there. Knowing what works, and what doesn’t will help you get you program stood up faster and with greater efficiency. With that, here’s your guide to understanding data governance best practice.

1. Start with your culture Data governance can be a tricky thing to get right because so much of it depends on creating policies and procedures that every person at the organization can abide by (or at least by those who interact with the data). However, anytime we’re asking employees to follow a set of processes, we should expect the opportunity for human error to be present. It’s just a reality that employees interpret rules differently, and, despite their best intentions, they can cherry-pick the rules that they will follow.

At the end of the day, your governance program is only as effective as the employees who abide by it. That’s why data governance should be embedded into the culture of a company. You can start by answering the following questions publicly and consistently: Why do these policies exist? Why should employees follow them? What are the consequences (to the business and to the employee) if they do not? Once the purpose of the program is made clear, further solidify understanding by making the program something every employee can understand. To many in the business, the term “data governance” can seem bureaucratic and harken back to failed attempts to launch a program, so consider using an alternative term that doesn’t evoke negative sentiments. Many in the industry have coined the phrase “data asset management” as a way to make governance a bit more relatable to the business.

2. Create a cross-departmental committee The most effective data governance programs are not run in silos; they’re run by a collaborative committee made up of department-level stakeholders from across the business. This not only includes stakeholders from IT, but also sales, marketing, finance, customer service, and so on. The benefit of this approach is that it ensures data definitions, rules, and processes are relevant to all areas of the business. This committee should meet regularly to identify data governance projects and work systematically and collaboratively to accomplish them.

You will want to avoid creating a committee that is too large to get anything accomplished, however. The leadership team at retail giant Amazon has the two-pizza rule, which essentially says that an effective team is made up of no more members than can be fed by two pizzas. It’s a good rule to thumb to follow for creating effective teams, and it also applies to creating an effective data governance committee.