The First Fifteen Seconds Decide the Next Fifteen Minutes
By Rita Okonkwo · The Craft
Most people write their talk from the beginning and rehearse it from the beginning, which means the opening is the part they have said out loud most often and thought about least.
That is backwards. The opening is doing a job no other part of the talk can do: it is answering a question the room is asking silently, and will only ask once.
The question the room is asking
The question is not "is this person qualified." It is closer to "is this going to be worth my attention."
Everyone in the room is making a quick, mostly unconscious estimate. They are reading your posture, your first sentence, how quickly you get to something that matters. And they are deciding how much of their attention to lend you. Attention is finite, and they have somewhere else to be, even if that somewhere else is their phone.
You do not get to skip this. You only get to answer it well or badly.
What people do instead
Three openings come up again and again, and all three spend the fifteen seconds on something other than the audience.
The apology. "Sorry, I did not have much time to prepare." "I am not really a speaker." You have just told the room that you do not think this is worth their time. They will believe you.
The throat-clear. "So, um, today I want to talk about..." followed by an agenda. That is a table of contents, and nobody has ever leaned forward for one.
The credential. Twelve years of experience, three degrees, two companies. This answers a question nobody asked. Credentials buy you patience, not attention, and only when someone already doubts you.
What works instead
Open with the thing that would make a friend say "wait, what?"
That can be:
- A specific moment. "Last March I stood up in a board meeting and completely lost the thread. Forty seconds of silence." A moment is concrete, it has a time and a place, and it puts the audience in a room with you.
- A number that does not fit. A true figure that contradicts what the room assumes. It has to be one you can source if someone asks, because an opening that turns out to be wrong costs you the rest of the talk.
- A question you intend to answer. Not rhetorical. A real one, that the talk then earns the right to answer.
The unifying property is that all three create a small gap, something the audience now wants closed. That gap is what carries them to your second minute.
A worked example
Say you are giving a five-minute update to leadership on a customer support backlog. The draft opening is:
"Thanks for having me. I'm going to walk you through the support numbers for the quarter, then some causes, then a proposal."
That is a throat-clear followed by an agenda. Here is the same opening built around a moment:
"On Tuesday morning, a customer who pays us more than any other waited six days for a reply to a billing question. That is not an outlier this quarter. I want to show you why it happens and what one change would fix it."
Assume the example is true; it has to be. It names a time, a person and a cost, and it ends by promising something specific. The agenda is still there, folded into one sentence at the end, where it is now an answer to a question the room has started asking.
A test you can run
Say your first two sentences out loud to someone who has no context. Then stop, and ask them: what do you think this is about, and do you want to know more?
If they can answer the first part but not the second, what you have written is an introduction. An opening is a different thing, and only an opening buys you the rest of the talk.
Why the opening is also where nerves peak
There is a second reason to give the opening extra attention, and it has been studied.
Ralph Behnke and Chris Sawyer, at Texas Christian University, tracked how anxious students felt at several points around a required speech.1 Anxiety was not highest during preparation, or even when the assignment was announced. It was highest just before speaking. In a later study they looked inside the talk itself.2 Two moments at the start, how sharply anxiety rose as speaking began and how quickly it fell in the first minute, accounted for most of how anxious the students felt while speaking overall.
So the first fifteen seconds are where the room decides about you and also where your own nerves are loudest. Both point to the same preparation. The longer account of that research is in the study write-up on exposure and speaking anxiety.
Common mistakes
A hook that is not connected to the talk. A startling opening that has nothing to do with your point gets attention and then spends it. The gap you open has to be the one the talk closes.
A joke you are not sure of. If it does not land, you have spent your first fifteen seconds recovering.
Saying "imagine" with nothing specific behind it. "Imagine a world where..." asks the audience to do the work. A real moment does it for them.
Rushing. Nerves speed people up at exactly the point where the room is deciding. Slow the first two sentences down on purpose.
Practise the opening on its own
The practical part: do not rehearse your talk from the top ten times. Rehearse the first fifteen seconds ten times, on its own, until it is automatic.
If the opening is so well rehearsed that you could deliver it while rattled, you get through the spike on autopilot, and the rest of the talk arrives on the other side of it with a calmer speaker attached.
A five-minute drill
Write three different openings for the same talk: one moment, one true number, one real question. Record yourself saying each one. Play them back to someone who has not heard the talk and ask which one made them want the second sentence. Keep that one, and rehearse only it until you can say it without thinking.
References
- Behnke, R. R., & Sawyer, C. R. (1999). Milestones of anticipatory public speaking anxiety. Communication Education, 48(2), 165–172. https://doi.org/10.1080/03634529909379164
- Behnke, R. R., & Sawyer, C. R. (2004). Public speaking anxiety as a function of sensitization and habituation processes. Communication Education, 53(2), 164–173. https://doi.org/10.1080/03634520410001682429