Good questions protect the room

Our software craftsmanship group encourages new speakers to come and practise their talks.

One is coming up in August. At the start of every meetup we read aloud the code of conduct.

It can feel formal, but the repetition is the point. It sets the culture and reminds us who we want to be.

A question can be good and still land badly. The timing and the room decide.

Something happens when someone gives a talk for the first time, or steps out of their shell because public speaking can be terrifying. A lot of it comes down to the questions people ask.

The question that helps gives something back to the speaker and never leaves them feeling small or put off speaking again. A question that does the opposite is the anti-pattern, because it teaches the whole community to stay quiet, and the discussions and the respect we have for each other are the reason the group works at all.

When good questions land badly

A couple of months ago I spoke at an AI meetup. The speaker before me had built an app with AI, and the app had security flaws. Someone in the audience spotted them and kept asking, question after question, until the speaker looked uncomfortable.

The questions were good. I'm sure the speaker would have welcomed the feedback in a code review, or over coffee. Asked in that room, at that moment, nobody else put their hand up. The room went quiet. The speaker felt awful.

The Recurse Center, a programming retreat, has a useful rule for one smaller version of this: no "well-actuallys", which they define as a correction with "no bearing on the crux of the conversation". Correctness matters. The harm is that some corrections move a room from learning towards status.

The security questions at the AI meetup were more serious than a minor correction, but the same lesson holds. A valid point can land in a way that shuts the rest of the room down.

From where I was sitting, it seemed unlikely the speaker could answer those questions well in that setting, so I won't judge what the audience member should or shouldn't have said. I care about the effect it had, on the speaker, and on every other person in the room who might have been weighing up their own first talk and decided against it.

Questions that are assumptions

Some questions are assumptions in disguise. The person has reached a conclusion and wraps it in a question mark to make it land.

Edgar Schein called the opposite of this humble inquiry, "the fine art of drawing someone out, of asking questions to which you do not know the answer." He subtitled the book The Gentle Art of Asking Instead of Telling, and that's the line.

A question you know the answer to is an instruction. Even a genuine question carries a risk the asker doesn't see. It can derail the talk. The question can be valid on its own terms and pull the room away from the point the speaker came to make. Ten minutes later you're deep in an edge case that mattered to one person, and the thread the speaker spent weeks preparing has gone.

What makes a question good

A good question helps everyone in the room. It's the one where someone leans over afterwards and says, "I'm glad you asked, I was wondering the same thing." It doesn't put the speaker on the spot, it furthers their learning, and yours.

Sometimes it's the single clarification the talk needs to land, because one undefined term changes the whole meaning.

Zenger and Folkman spent years studying what great listeners do. They found that "people perceive the best listeners to be those who periodically ask questions that promote discovery and insight," and that "these questions gently challenge old assumptions, but do so in a constructive way."

A good question leaves the speaker ready to move forward.

Amy Edmondson gave this a name in 1999: psychological safety, "a shared belief held by members of a team that the team is safe for interpersonal risk taking," a confidence that the team "will not embarrass, reject, or punish someone for speaking up."

Edmondson's talk on the same idea, The Fearless Organization, is worth watching:

Standing up to talk is an interpersonal risk. So is asking a question, admitting you don't know something, or floating a half-formed idea. Protect that risk and people take it again. Punish it once in public and they stop.

Candor has to be built in

As Ed Catmull wrote in his book "Creativity, Inc.":

“A hallmark of a healthy, creative culture is that its people feel free to share ideas, opinions and criticisms… without the critical ingredient that is candor, there can be no trust. And without trust, creative collaboration is not possible.”

Pixar built a structure for that candor, the Brain Trust. It is a regular meeting where a director gets honest feedback on a film in progress. The comments are suggestions, not orders, and they are aimed at the work, not the person. It is the same thing a good question does at a meetup, made into a habit.

Draw quieter voices in

After a meetup once, a good friend pulled me aside. He said it was good that I had the confidence to contribute and ask questions, but my real aim should be to bring the people who weren't talking into the conversation. That would be the real win. It stuck with me.

Asking a good question is one skill. Drawing in someone who hasn't spoken is the harder one, and it does more for the room than any point I could make myself. It's the same move as a fielding team calling each other in, which I compared to a cricket side.

Before you ask

At the next meetup, I run through these in my head:

1. Can the speaker answer this in the room, right now?
2. Does this help the room, or only prove you noticed something?
3. Are you asking because you don't know, or because you've decided?
4. Who in the room hasn't spoken yet, and could your question invite them in?

Those keep me from asking the wrong thing. For asking the right thing, I borrow Edmondson’s own set, in two kinds. Some broaden the discussion:

• What do others think?
• What are we missing?
• What other options could we consider?
• Who has a different perspective?

Some deepen it:

• What leads you to think so?
• What is the concern you have about that?
• Can you give us an example?
• Can you explain that further?
• What do you think might happen if we did X?

Notice "who has a different perspective?" rather than "does anyone have a different perspective?". The first assumes another view is out there and worth hearing. The second sets a hurdle almost no one in a full room wants to clear.

The same forces play out inside companies, where one confident voice can take over the room and everyone pays for it. I wrote about that in The loudest voice is not the smartest group.