Advocacy-Inquiry: The Debrief Questioning Technique That Surfaces Real Reasoning

Forty minutes into our second Operation Aetherfall pilot debrief, I had a judgment burning a hole in my notepad. One player had sat on information that contradicted the commander's plan through the entire briefing scene – information the team needed and didn't get. Two questions formed in my head, and both were bad. "Why didn't you share that earlier?" is an accusation wearing a question mark. The safe alternative – "How did that scene feel for everyone?" – would've cost me nothing and surfaced nothing.

There's a third option, and it comes from healthcare simulation. Debriefing with good judgment, developed by Rudolph, Simon, Dufresne, and Raemer at Harvard's Center for Medical Simulation, pairs advocacy – a specific observation plus your honest assessment of it – with inquiry, a genuinely curious question about the reasoning behind what you saw. In practice it sounds like this: "I noticed the contradictory data didn't come up until after the plan was set. I think it might have changed the decision. Help me understand what made it hard to bring up." Observation, judgment, curiosity. All three out loud, in that order.

Advocacy-inquiry is the engine of the analysis phase in the three-phase debrief structure we run after every session, and it's the most useful facilitation technique in the 200+ studies my co-designer Adam White and I synthesized while building our scenarios. It's also learnable. The phrasing feels strange for about a week, and then it changes how you ask questions everywhere.

What is debriefing with good judgment?

Debriefing with good judgment is a method for giving feedback after a training exercise that treats the facilitator's judgment as something to share openly rather than hide or weaponize. Rudolph and colleagues introduced it in a 2006 paper whose title carries half the argument: "There's No Such Thing as 'Nonjudgmental' Debriefing." Their 2007 follow-up gave the method its working definition – combining rigorous feedback with genuine inquiry.

The theory underneath is what makes the technique work. People's actions are driven by their frames: the knowledge, assumptions, and feelings they were operating from in the moment. The action you watched is the visible output of an invisible frame. Correct the action and the frame keeps producing more like it. To change what someone does next time, you have to surface the frame and work on it directly – and the only person who can show it to you is the participant. Advocacy-inquiry is the surfacing tool. You put your observation and your judgment on the table, and you trade them for a look at their reasoning.

The same Harvard group later built DASH, an instrument that rates debriefing quality partly on exactly these conversational qualities – the stance matters enough that people measure it.

Why do "what went wrong?" and "how did that feel?" both fail?

They fail for opposite reasons: the judgmental question hides the participant's reasoning, and the feelings-only question hides yours.

"What went wrong out there?" arrives carrying a verdict. Participants hear the verdict before the question and answer accordingly – they defend, justify, minimize. You get a courtroom, and courtroom testimony tells you almost nothing about frames. There's a slower cost too. After a couple of these, participants learn the debrief is a place where wrong answers exist, and they start producing whatever answer they think you're looking for. That's the end of honest analysis.

The overcorrection is just as common. A facilitator who's been warned off judgment swallows theirs entirely and asks how everything felt. Feelings questions have a legitimate job – our debrief opens with ten full minutes of them, on purpose, because people can't analyze their own behavior while they're still activated. But as an analysis strategy they're toothless, and they're dishonest in a way participants can smell. Rudolph and colleagues' paper title is the rebuttal: there's no such thing as nonjudgmental debriefing. You have a judgment. It leaks through your tone, your word choice, which moments you decide to ask about. Withholding it doesn't make the debrief safe. It just deprives participants of the most useful data in the room – what an experienced observer actually thought – while leaving them to guess at it.

The choice was never judgment versus no judgment. It's judgment delivered as a verdict, judgment leaking around a carefully neutral question, or judgment shared next to genuine curiosity.

How do you build an advocacy-inquiry question?

Three parts: an observation specific enough to point at, an assessment you own with "I," and a question you don't already know the answer to.

The observation has to be concrete – something a camera would've caught. "I noticed the estimate went through without a challenge" works. "I noticed you were all being passive" doesn't; passive is a judgment wearing an observation's clothes.

The advocacy is your honest read, and the "I" does real work. "I was concerned that might leave the rest of the team without context" marks the judgment as yours – a hypothesis you're offering, not a ruling you're issuing.

The inquiry is where most attempts quietly fail. "Help me understand what was driving that decision" only works if you mean it. If you already know the answer you want, you've built a leading question with better manners – "don't you think you should have raised it with the whole group?" is advocacy in a trench coat. The test of genuine inquiry: you have to be ready for the answer to prove your assessment wrong.

Here's the rewrite for three moments that show up in nearly every workplace training exercise we run:

The moment	The reflex question	The advocacy-inquiry version
An estimate nobody believed went unchallenged	"Why didn't anyone push back?"	"I noticed the estimate went through without a challenge, and two of you told me at the break you disagreed with it. My worry is the silence read as agreement. What was happening for you in that moment?"
One person made the call for the whole group	"Do you think everyone felt heard?"	"I noticed the decision landed about a minute after you proposed it, and three people hadn't spoken yet. I was concerned the team traded buy-in for speed. Walk me through how it looked from your seat."
A hard conversation happened in private instead of with the team	"How did that feel?"	"I noticed you chose to address the conflict privately rather than in the group meeting. I was concerned that might leave the other team members without context. Help me understand what was driving that decision."

Read the middle column again. None of those reflex questions is monstrous, which is what makes them worth studying. The first is a verdict, the second is a leading question pretending to be curious, and the third belongs in the reactions phase doing honest work – not standing in for analysis.

What does advocacy-inquiry sound like at a real table?

Our facilitator notes for the Operation Aetherfall analysis phase include this line, written out in full so it's available under pressure: "I noticed you chose to address Korr by leading with the structural data rather than directly challenging his plan. Help me understand what was driving that decision." Korr is the commander in our scenario. He's working from a flawed plan, several players hold information that contradicts it, and how they handle that disagreement is the training core of the session – which makes their reasoning about it the debrief's most valuable cargo.

Notice that the Korr question compresses the advocacy. When your assessment is closer to admiration than concern, the judgment can ride along implicitly – "that worked, and I want to understand why" is also a judgment. The full three-part structure earns its keep when the observation stings.

The raw material comes from play. While the game runs I keep a notepad and jot down three to five specific moments: a question that changed a conversation's direction, a challenge to the commander, how a decision actually got made, someone holding back, a disagreement starting to heat up. (The craft of running the session itself is its own article.) Each note is a candidate for an advocacy-inquiry question, because generic questions produce generic answers. "What did you learn about conflict?" gets platitudes. "The moment Bramm pushed back on Korr's timeline" gets analysis.

Not every question needs the full apparatus. Most of our debrief question bank is pure inquiry – "Whose input was included in that decision? Whose wasn't? Why?" – and pure inquiry is the right tool when you genuinely don't have a judgment yet. Advocacy-inquiry is for when you're carrying one. Which, if you watched the exercise closely, is more often than most facilitators admit.

How much should the facilitator be talking?

Less than the participants, and it's worth measuring. Research on conversational balance in simulation debriefing found that sessions where the debriefer-to-participant talk ratio stayed under 1:1 – learners talking more than the facilitator – received notably higher quality ratings than sessions the facilitator dominated. Your expertise doesn't come out in explanation. It comes out in question selection.

Advocacy-inquiry fits that constraint well, and I suspect that's part of why it works. The whole move is three sentences: what I saw, what I made of it, what I'm curious about. Then the floor goes back to the participants, and your job becomes the hardest part of the craft – waiting through the silence while someone assembles an honest answer.

How do you practice the phrasing?

Out loud, before you need it, from real notes.

During Aetherfall's mid-session break I review my gameplay notes and draft one advocacy-inquiry question per moment, in writing. The syntax won't assemble itself on the fly at first. Under the mild pressure of a live debrief you'll revert to whichever question style is already grooved in, and for most of us that's the verdict. Writing the questions in advance is what practice looks like until the structure becomes the groove.

A fill-in frame helps: I noticed ___ (something a camera would've caught). I was thinking ___ or my concern was ___ (owned with "I"). Then a question you don't know the answer to. The inquiry half borrows freely from the same family as phrases that open conversations – "help me understand," "walk me through," "I'm curious about" – which is no accident. A debrief is a structured conversation about disagreements someone just lived through.

I'll be honest about my own learning curve. Adam spent 13 years as an electrical engineer before becoming a licensed counselor, and genuine curiosity is his native register. Mine is "here's what I'd do" – twenty years of product management trains you to arrive with an answer. I had to practice the inquiry half until it stopped being a formality I was performing on the way to my real point. The tell, for me: if I felt impatient during the participant's answer, I hadn't asked a real question.

Fair warning: the structure follows you home. It turns out spouses and kids also have frames.

Back at that pilot table, I went with the third option – the observation, my concern, and a question I didn't know the answer to. The player hadn't been avoiding conflict at all. They'd assumed someone else at the table already knew, and that saying it out loud would've been redundant. My judgment was wrong in a way no verdict would have discovered. That's what the inquiry is for.