In a 1998 ESPN interview, Michael Jordan said about Phil Jackson: “Sometimes he can say one word, one sentence, and shake you up, make you think.”
Jackson rarely lectured or commanded. He led with questions, the kind that made his players stop, reflect, and find their own answers. Jordan credited that questioning style for much of how the team learned to think together.
My last article made the case for listening as the starting point of good governance. Listening is where it begins, yet hearing a stakeholder is not the same as understanding one. Understanding takes questions, and asking them is the very thing most of us are afraid to do.
Data governance is one of the most opinionated fields in business. Executives, leaders, architects, and practitioners each come to the table with their own understanding, their own viewpoint, their own definition of what good looks like. Yet in most presentations and requirements meetings, silence is the loudest voice in the room.
In that environment, asking the right question is the most powerful thing a practitioner can do.
Research suggests a three year old asks between 100 and 300 questions a day.
By the time we reach our professional years, that number drops sharply. We learn to supply answers instead of raising them. We come to be rewarded for what we know, rarely for what we wonder.
Richard Feynman, the Nobel Prize winning physicist, treated curiosity as the engine behind every good question. Socrates built an entire philosophy on it, what we now call Socratic ignorance: the discipline of knowing what you do not know. He reached the deepest truths of his time through questions rather than statements.
In data governance, the same principle applies. The practitioner who asks the right question at the right moment does more for the program than the one who arrives with all the answers.
If we never ask, how do we understand what is causing pain for our stakeholders? How do we know what they actually want from a data catalog, a business glossary, or a governance program? Without question, we build on assumptions. Those assumptions, however well-intentioned, make a fragile foundation for any program that depends on people choosing to participate.
Questions uncover what is not yet visible. What does good enough look like for data to be fit for its purpose? Are reporting platforms consuming data from authoritative sources or authoritative distributors? Is this data asset truly critical to a business decision? These belong in every governance conversation, repeated until the answers become part of how the organization thinks about its data.
When governance is shaped by the questions we dared to ask rather than the assumptions we never tested, adoption follows. People engage with programs they helped shape. They trust answers they helped find.
Here is an uncomfortable truth. Research published in Harvard Business Review surveyed more than 3,000 employees and found that only 24 percent felt curious in their jobs on a regular basis. More striking still, 70 percent reported real barriers to asking questions at work. The number one reason? Fear of looking incompetent.
Leaders, almost universally, say they value inquisitive minds. Yet the environments they build quietly discourage the very thing they claim to want.
Data governance is no exception. I have sat in steering committee meetings where the catalog was live, ownership assigned, policies documented, and by every dashboard metric, the program looked like it was progressing. Then someone asked a question that had apparently never been said out loud: “Is anyone actually using this?”
For a decade, we built data governance platforms and asked business users to come to the data, to learn the catalog, understand the lineage, and speak the language of the data team. They largely did not, and the reason was rarely disinterested. The friction was too high, and the fear of asking the wrong question was too real.
Agentic AI changes that equation entirely. For the first time, a business user can walk up to their data catalog the way they walk up to Google and simply ask, posing a question in plain language and getting an answer in seconds.
The technology will do something that data governance programs have spent years trying to do. It will make asking questions effortless.
Asking questions builds something no dashboard can measure: trust. Not the procedural kind that comes from a policy document or a signed accountability matrix, but the trust that grows when stakeholders and business users feel genuinely heard.
Asking questions uncovers what we cannot see on our own: the blind spots quietly shaping decisions, the assumptions nobody thought to challenge, the friction business users stopped reporting because nobody seemed to be listening.
When people feel heard, they engage. When blind spots are named, they get fixed. When governance is shaped by what users actually want, adoption stops being a problem you manage and becomes an outcome that happens naturally.
What is the question your next governance meeting needs that no one will want to ask?
Previous in the series: Data Governance: Lead by Listening → 🔗
