In narrative interviews, most of the work happens before you sit down with anyone.
It happens when you write the opening question.
That one question, the eliciting question, determines whether someone tells you a real story or gives you a polished summary. Whether they reach back into memory or stay safely on the surface. Whether the conversation goes somewhere unexpected, or confirms what you already thought you knew. Getting it right matters more than almost anything else in the process.
A good eliciting question does one thing: it points someone toward a specific experience they actually lived through.Not an opinion. Not a general impression. A moment. “What do you think about collaboration in your team?” invites a verdict.
“Can you tell me about a time when collaboration in your team really worked or really didn’t?” invites a memory. That difference is everything. Memories have texture. They have people in them, and weather and things that were said. Verdicts are already cleaned up. The “mess” has been removed. And the mess is usually where the meaning lives. The trap of the question that sounds good. Here is the uncomfortable truth: a question can sound perfect and still produce nothing.
You test it on yourself and it feels open, neutral, inviting. Then you sit across from a real person and they stare at you for three seconds and say “I’m not sure what you mean exactly.” Or they answer a completely different question than the one you thought you asked. Or everyone gives you the same kind of story, and you realise later the question was quietly steering them there all along.
This is why testing matters. Not as a formality. As genuine research.
How to test your eliciting question?
Ask it to at least five people before you use it in a real session. Not colleagues who know your project. People who are close enough to the world you are researching to give a real answer, but far enough from your thinking that they bring no assumptions with them.
After each conversation, ask yourself three things.
Did they tell a story, or did they give a summary? If they summarised, the question may be too abstract or too evaluative. It is pulling them toward reflection rather than memory. Did the story surprise you? If every answer feels predictable, the question is probably leading. Something in the wording is pointing people toward a particular kind of response without you realising it.
Did they have to think? A small pause before answering is a good sign. It means they are actually searching for something. If the answer comes instantly and fluently, they may be reaching for something rehearsed rather than something real. Adjust, then test again.
Change one thing at a time. A single word can shift the entire weight of a question.
“A time when it worked” produces different stories than “a time when it felt right.” “Collaboration” produces different stories than “working together.” “In your team” produces different stories than “with the people around you.”
None of these is automatically better. The right wording depends on the world your participants actually live in, the language they use, the way they talk about the thing you are trying to understand.
That is exactly why you cannot write the question alone at your desk. You need to hear it land. You need to watch what it opens and what it closes.
The question is already an act of listening
There is something deeper here too.
The way you frame an eliciting question tells people what you think is worth talking about. It signals what you are ready to hear. A question that is too narrow says: I already know what matters, just fill in the details. A question that is genuinely open says: I don’t know what you carry, and I want to find out.
People feel that difference. And they respond to it.
So when you invest real time in finding the right question, testing it, listening carefully, adjusting it, you are not just improving your methodology. You are practising the thing narrative work is actually about. You are learning to ask in a way that makes space for someone else’s truth. Good luck in finding the best elicitation question.