Last week I attended DATAVERSITY’s free two-day program, Data Governance and Information Quality. It featured a dozen live online presentations, and I was happy to join thousands of data professionals to learn more about the current state of data governance. While the conference was online this year due to health and safety concerns, it was nice to feel connected to other data enthusiasts and hear from experts, no matter the format.
There was a lot of valuable information shared, and here are some things that stood out to me:
- “AI does not fix stupid”– Scott Buckles. In his talk, Unknown Threats: De-Bunking the Top Myths Behind Data Governance, Buckles address the myth that AI eliminates the need for data governance. It is important to take a step back before implementing AI to understand the importance of information architecture. AI will not be useful until there is understanding behind the information that exists and how it is organized.
- “The most important thing to do in a conflict is nothing,” –Len Silverston. This insight was surprising to me, as I always thought that immediate action during a conflict was important. However, Silverston expresses the importance of taking a step back from conflict, and taking a moment to cool down before addressing the conflict. Culture and human behavior are the key to success, so it is important to get a broader view of any situation beyond “me, me, me” thinking.
- “Both business and technology personas are being brought to the table” –Susan Wilson. The way data is being addressed in organizations is changing, with new roles such as the Chief Data Officer, and the consideration of both the business and technology sides of an organization working together.
- “We desperately need responsive and resilient data governance” –Laura Madsen. Agile data governance is as important now as ever, and it has changed the landscape of data governance. For example, businesses have moved from having one person with lots of knowledge running the show to multiple people with similar capabilities sharing responsibilities. There’s been a move from having lots of meetings to eliminating standing meetings for a more flexible environment. There has also been a move from a lack of clarity for roles to distinct functions for participating resources, as well as establishing concrete timelines for efforts.
- “A communication and marketing approach is important to engage resources” –Curtis Mischler. The marketing team can align with regular meetings and presentations, as well as establish intranet sites and other mechanisms to inform the data governance team. This can also make implementation easier, by building on existing processes instead of trying to create something brand new.
- “There is more attention being paid to data governance now than five years ago.” – Bob Seiner. Many companies are now trying to become data centric. There is an understanding that if we continue to treat data the way we have been treating it, we are not going to get a return on investment, so data governance is essential.
- “We are explicitly governing unstructured data, whereas in the past there was an emphasis on structured data.” –David Kowalski. Companies are now putting into place stricter and more defined rules concerning unstructured data, such as rules against sending documents to yourself at home. The industry is also starting to see formalized policies on definitions of what unstructured data is and how to best handle it.
- “We are in the golden age of data.” – Malcolm Chisholm. More companies are seeing the value of data governance, and new practices are being established in the field. For example, data catalogs are now driving data governance, which is something that had not been widely occurring before. We are continuing to see the value of data and make adjustments on how to best utilize it.
These were just some of my key takeaways from DGIQ2020, but there were many more insights made during the event. If you are looking for more information, the talks are available on DATAVERSITY’s website.