How to Answer a Question You Do Not Know the Answer To
By Dara Halloran · Under Pressure
Someone asks a question you cannot answer. The room is waiting. This is one of the most common high-pressure moments in professional life and almost nobody has thought about it in advance.
The instinct is to produce something: a plausible-sounding paragraph assembled in real time. It is the worst available option and it is what many people do.
Why bluffing fails
It is tempting to justify a bluff by assuming nobody will notice. On the narrow question of spotting lies, that assumption is not unreasonable. Charles Bond and Bella DePaulo pooled the results of about two hundred research reports on deception detection and found that people judging whether statements were truths or lies were right only about 54% of the time, barely better than a coin toss.1
But a bluff in a meeting is not a lie judged in isolation by a stranger. It is a claim about your field, made in front of people who often know part of the answer, and it can be checked. A single follow-up question, "which study was that?" or "what was the number last quarter?", is usually enough to expose an answer that was assembled on the spot. The listener does not need to read your face. They only need to ask one more thing.
Even before that, an improvised answer tends to sound different. When you are generating filler, specifics thin out and abstraction rises: "there are a lot of factors", "it depends on the context", "broadly speaking". Listeners may not consciously label it as bluffing, but the answer stops being useful, and they notice that.
The cost is asymmetric and this is the important part. Saying "I do not know" costs you a small amount of credibility on one question. Being caught bluffing costs you credibility on everything you said before it, because the room now has to re-evaluate whether any of it was constructed.
That trade is never worth it.
The four moves
1. Say you do not know, then say what you would do.
"I do not know. What I would do is pull the retention numbers for those two cohorts and see whether the drop-off is at week one or week six. That would tell us which of the two explanations is right."
This is the strongest option and it is badly underused. It converts a gap in knowledge into a demonstration of how you think, and thinking is generally what is being assessed.
2. Answer the part you can.
"I cannot speak to the legal position. On the operational side, what I can tell you is ____."
Be explicit about the boundary. Silently narrowing the question and answering the easy part is a form of bluffing, and it is often noticed.
3. Ask a clarifying question, if it is real.
"When you say 'scale', do you mean more users on the current model, or a different model entirely? Those have very different answers."
This buys time and often reveals the question was narrower than it sounded. Do not use it as a stalling device on a question that is already clear; that is transparent.
4. Commit to following up, specifically.
"I do not have that number. I will send it to you by Thursday."
A name, a thing, a date. "I will look into it" is a way of ending the conversation, and everyone knows it. Then actually send it, because the follow-up is what turns "I do not know" into evidence that you are reliable.
Choosing between them
The four moves are not interchangeable. A rough guide:
- If the question is about facts you could look up, use move four, perhaps with move one.
- If it is about judgement or approach, use move one. Saying how you would find out is itself the answer.
- If it crosses into someone else's area, use move two and name who would know.
- If it is ambiguous, use move three, once.
You can combine them. "I do not know the exact figure. My understanding is that it is under ten per cent, but I will confirm by Thursday" is moves two and four together, and it is honest about which part is solid.
The delivery matters more than the words
All four collapse if delivered apologetically. With hunched shoulders, trailing volume and a nervous laugh, "I do not know" reads as being caught out.
Delivered at full volume, level, without apology, the same three words read as candour. Same information, opposite effect. Hedging around the admission does not soften it; it only makes it longer. There is more on how small qualifiers weaken a speaker in powerless speech.
It also helps to know that the moment feels larger from the inside than it looks. The pause before your answer and the flush of embarrassment are far less visible to the room than they feel, as the research on the illusion of transparency shows.
Practise saying "I do not know" out loud until it sounds like a statement rather than a confession. It is a surprisingly hard drill and it is worth the time.
The preparation you can do
Before any high-stakes Q&A, write the five questions you least want to be asked. For each, decide which of the four moves you will use, and draft the first sentence.
The point is to pre-decide your response category rather than memorise answers, so that under pressure you are executing a choice you already made calmly rather than making one badly on the spot.
If some of those questions are likely to arrive with an edge, the companion piece is the hostile question is usually a gift. The same principle runs through both: the room is judging how you handle the gap at least as much as whether you have one.
References
- Bond, C. F., & DePaulo, B. M. (2006). Accuracy of Deception Judgments. Personality and Social Psychology Review, 10(3), 214–234. https://doi.org/10.1207/s15327957pspr1003_2