Not All Questions Are Equal | Flotog BA Insights

The fastest way to ruin a discovery session is to start it with a good closed question. I mean that. Not a bad question. A good one, clear and specific and easy to answer, asked at the wrong moment. "Do you want the report daily or weekly?" is a perfectly reasonable question to ask near the end, once you understand what the report is for. Asked at the start, it quietly tells the stakeholder that the conversation is about scheduling a report, when the conversation should have been about whether a report is even the right thing to build. A question is not just a way of getting information. It is a frame. It tells the other person what kind of answer you are looking for, how much room they have, and what is and is not on the table. Before you worry about the wording of any single question, you have to understand what each type of question does to the person hearing it.

The four shapes you reach for most

There are many ways to slice question types, but four cover most of the work, and you want them available without having to think.

Open questions invite a story. "Walk me through how the team handles a return." They give the person room to take you somewhere you did not know existed. You use them early and often, whenever you are still mapping the territory.
Closed questions narrow to a fact or a yes. "Does the refund go out the same day?" They are for confirming, for pinning down a detail, for checking you understood. They are precise, and they are also a kind of cage, so you save them for when you actually want to close something, not when you are still opening it up.
Probing questions go deeper into something already on the table. "You said that step is painful. Painful how?" They are the follow-up that most people skip because it feels like prying. It is not prying. It is the difference between a surface answer and a useful one. The first answer is the headline. The probe gets you the article.
Context-free questions are the underused ones, and they are worth their own paragraph.

The questions you can ask before you know anything

Years ago, Gause and Weinberg made the point that some questions are worth asking on any project, before you understand a single thing about the domain. They called them context-free, because you can carry them from a banking project to a hospital project without changing a word. "What problem does this solve?" "Who is affected by it and how?" "What happens if we do nothing?" "How much is solving this worth to you?"

These questions feel almost too simple to ask a senior expert, which is exactly why people skip them and exactly why they keep paying off. They surface the assumptions everyone in the room is carrying silently. Half the time, the people in the room give different answers to "what problem does this solve," and the meeting you thought was about requirements turns out to be a meeting about disagreement that nobody had named yet. That is a good day. You would much rather find the disagreement now than in user acceptance testing.

The funnel: order is the real skill

Here is the part that separates someone who knows the question types from someone who can actually run a session. The order matters more than any single question. The reliable pattern is a funnel. You start wide, with open and context-free questions that let the person take you somewhere. You let them talk longer than feels comfortable. Then, once you can see the shape of what they are describing, you narrow. You probe the parts that matter, and only at the very end do you close down to the specific facts and confirmations you need to write something precise.

Most struggling sessions invert this. They start narrow because narrow feels safe and efficient, and they never widen out, so the BA leaves with a tidy set of answers to questions the stakeholder was never given the chance to challenge. Wide to narrow. Discover, then confirm. If you remember nothing else from today, remember the direction of travel.

A simple funnel you can run tomorrow

Open the territory: "Talk me through a normal week for your team." Let them ramble. The detours are where the real information hides.
Find the friction: "Which part of that do you dread?" Now you know where to dig.
Probe the friction: "Tell me about the last time that went wrong." Stories beat opinions every time.
Confirm and close: "So if I have this right, the blocker is X and not Y?" Now the closed question earns its place.

Four habits that quietly kill sessions

Even with the right types in the right order, a handful of small habits leak information out of a session before it can land.

The first is the leading question, the one with the answer baked in. "You would want approval before it goes out, right?" The stakeholder agrees because agreeing is easier than correcting you, and you have just put your own assumption into their mouth and written it down as theirs.
The second is the double-barrelled question, two questions wearing one coat. "How often does this happen and who handles it?" The person answers whichever half is easier and the other half evaporates. Ask one thing at a time. It feels slower, but it is faster, because you do not have to circle back.
The third is the question that is really a statement. "Wouldn't it make more sense to automate this?" That is not a question; it is your opinion in a question mark. If you have a view, you can offer it, but do not disguise it as curiosity. People can feel the difference, and they trust the disguise less than the honest opinion.
The fourth is the jargon question, the one that uses your vocabulary instead of theirs. The moment you ask a warehouse supervisor about the "reconciliation workflow" you have made them choose between admitting confusion and bluffing, and most people bluff. Ask in their words. Their words are the requirement. Yours are a translation that has not been checked yet.

Why this travels

A doctor asking "where does it hurt" before "is it a sharp pain or a dull one" is running the same funnel. A manager who opens a one-to-one with "how are things" before "why is the report late" gets a truer answer than one who leads with the accusation. The structure underneath good questioning is not a BA trick. It is how you find out what is true from another human being who may not have the words ready.

A junior BA can take the four types and the funnel and run a noticeably better session this week. A senior already does the funnel by instinct, and reading it laid out like this is mostly a reminder of what they stopped noticing they do. Tomorrow we go to the hardest ground of all, the things a stakeholder cannot tell you because they do not consciously know them. Question types will not be enough there. That is where the craft really begins.

Go out and be successful.

Oluwatosin Ogunkoya | Flotog BA Insights | www.flotogbainsights.com

Tomorrow: Surfacing What They Cannot Say

How to reach the knowledge stakeholders have but cannot put into words.