Questions you can't answer: what to say

We have already covered how to structure a presentation for an executive team, and how to run a demo as a conversation rather than a monologue. This is about the moment no structure can protect you from: the question you do not have an answer to. The advice 'never say you don't know' appears in plenty of sales handbooks. It is wrong, and there is data on why.

Why admitted uncertainty persuades

In 2010 Uma Karmarkar and Zakary Tormala published a series of studies in the Journal of Consumer Research on what they call source certainty. The finding: low-expertise sources become more persuasive when they express certainty, while high-expertise sources become more persuasive when they express uncertainty. It violates expectations, stimulates involvement and increases influence. A rep who has spent twenty minutes walking through the buyer's problem in the buyer's numbers has already established themselves as the expert in the room. From that position, 'I don't know that off the top of my head' is a signal that the rest is true.

Two other results point the same way. Elizabeth Tenney and colleagues showed in Psychological Science in 2007 and in the Journal of Experimental Social Psychology in 2008 that calibration beats confidence: a source whose confidence matches how often it is right is judged more credible than one that is equally sure of everything. And Elliot Aronson showed in a 1966 experiment that a competent person who made a clumsy blunder was seen as more likeable afterwards, while the same blunder lowered the standing of an average person. It only works for someone who has already shown competence, which is the whole point: honesty works after twenty good minutes, not instead of them.

Why the guess costs more than it earns

Across 67,149 demos, Gong's analysis found not one that led to a closed deal where the rep pitched uninterrupted for more than 76 seconds, and successful demos had 21 percent more speaker switches per minute. A question you cannot answer is a speaker switch, which makes it a good sign, right up until you fill it with a monologue of guesses. Guesses about money are the most expensive of all: Gong's study of more than a million sales cycles with executives shows that mentioning ROI at all during discovery correlates with lower next-meeting rates. An invented ROI answer to a question about numbers is the worst combination there is. Gartner, writing in Harvard Business Review, reports that B2B buyers spend 17 percent of their purchase process talking with suppliers. Every minute is expensive, and a minute that later has to be corrected costs double.

Three kinds of question, three answers

What makes 'I don't know' hard to say is that it sounds the same whatever the question. It should not. Questions you cannot answer come in three kinds, and each has its own wording.

1. The fact question: 'Do you support SSO through our provider X?' You do not know. The answer: 'I don't know that off the top of my head, and I don't want to guess about your IT environment. I will check with our technical lead this afternoon and send you an exact answer tomorrow morning.' Who, when, and nothing more.
2. The comparison question: 'How do you compare with vendor Y on this?' You are not familiar with the competitor's latest version. The answer: 'I can tell you what we do, and I won't claim things about their current version that I can't back up. If you are evaluating both, I suggest you put the same three questions to them and compare the answers.' You hand the buyer the questions instead of a verdict.
3. The commitment question: 'Can you have it ready by January?' It is not your call to make. The answer: 'I want to say yes, and that is exactly why I won't until I have spoken to the person who owns the delivery plan. Can I come back with a date you can hold us to, by Friday at the latest?' A promised date beats a date you cannot guarantee.

The lines

• 'I don't know that off the top of my head. I will come back with an exact answer tomorrow.'
• 'Good question, and I would rather give you the right answer than a quick one. Can I take it away?'
• 'What I know is A. What I don't know is B, and I will check that.'
• 'I don't want to guess about your environment. Who on your side can I confirm the details with?'
• 'That one is not mine to promise. I will get you an answer from the right person by Friday.'

What is not on the list: 'I think', 'it should work', 'I'm fairly sure'. All three are guesses with a hedge word in front, and the buyer only hears the guess.

The follow-through is the trust moment

Admitting uncertainty only builds trust if the answer actually arrives. Write the question down word for word in the meeting, repeat it aloud so the buyer hears you got it right, and send the answer within 24 hours using the same words the question was asked in. An answer that lands the next morning, exact and with a source, is often the thing the buyer remembers best from the whole presentation, because it is the only thing that happened after the meeting. If you cannot get the answer within a day, send what you have and a new date. Silence is the one outcome that takes back the trust your honesty built.

The hard part is not the wording but the reflex to fill the silence, and that reflex can be trained away. In Salesprep's Presentation module you present to an AI audience that asks questions back, including ones your material has no answer to, and the scores on six metrics show whether you guessed, hedged, or said what you knew and promised the rest. Presentation is available on Team Pro, and three free calls in the Cold call module come with the account for anyone who wants to try the AI buyer first, no card required.