Choosing a speaking rate
Prepared delivery to an audience commonly falls between 120 and 150 words per minute. That is the range most presentation coaching converges on, and it is slower than people expect, because conversation runs faster. Ordinary talk between people who know each other sits somewhere around 150 to 190, and fast speakers exceed that comfortably without sounding rushed to a listener who already knows the context.
The gap matters because your audience does not have the context. New material needs processing time, and processing time is bought with pace and pauses. Broadcast news is written at around 160 wpm, but news copy is engineered for that speed with short sentences and no subordinate clauses. A technical talk delivered at 160 loses people. If the content is unfamiliar, or has numbers in it, drop toward 120.
These are averages over populations, and a single speaker varies widely by mood, by room and by how many times they have given the talk. Time yourself once against a page of your own script and use your own figure; that is worth more than any published range.
What the pause allowance is buying
The word count times the talking. It does not time everything else the talk is made of.
| Format | Typical allowance | What it covers |
|---|---|---|
| Voiceover or straight narration | 5% | Breath, sentence boundaries |
| Slide presentation | 10–15% | Slide changes, the beat after a point, glancing at notes |
| Teaching or demonstration | 20%+ | Showing the thing, writing on a board, waiting for people to catch up |
| Anything with audience interaction | 25%+ | Laughter, interruptions, the pause where you check whether that landed |
Deliberate silence is the most underused tool in presenting and the one most often squeezed out by an overlong script. If you find yourself cutting the pause allowance to make the script fit, you are not making the talk shorter, you are making it worse.
Building to a hard limit
Where the limit is enforced, as in a conference slot or a competition, aim to finish at roughly ninety percent of it. The margin absorbs a slow start, a technical problem, a question you decide to take, and the fact that most people talk faster when nervous than they did in rehearsal. Finishing two minutes early is invisible. Being cut off mid-point is not, and the part you lose is your conclusion.
Where questions come out of your slot rather than after it, put that figure in the reserve field and the comparison will work against what is left.
Reading the estimate honestly
This calculation assumes an even pace across the whole script, which nobody delivers. Openings tend to be slower and more careful; the middle speeds up; the ending slows again if the speaker is aware of it and accelerates badly if they are not. That unevenness cancels out reasonably well over a ten minute talk and not at all over a ninety second one, so short-form timings from this tool deserve more scepticism than long-form ones. The only measurement that settles it is a stopwatch and a rehearsal.
Questions people ask
How many words is a five minute speech?
At 130 words per minute with a twelve percent pause allowance, roughly 580 words. At 150 wpm with a five percent allowance it is closer to 715, and at 110 wpm with a twenty percent allowance it is about 460. The range across plausible settings is wide enough that quoting a single number for "a five minute speech" is not useful, which is why the rate and the allowance are fields here rather than constants.
Should I write the script out in full or use notes?
That is a delivery question and this tool has no opinion on it, but the timing behaves differently. A full script read aloud is highly predictable and lands close to the estimate. Speaking from bullet notes is much more variable, usually longer than the equivalent word count would suggest, because unscripted speech carries more repetition and more false starts. If you work from notes, write a full draft to get the timing, then reduce it to notes and rehearse the notes version with a stopwatch.
Why is my rehearsal faster than the actual delivery, or the other way around?
Both happen and they have different causes. Rehearsing alone tends to be faster than delivering to a room, because you skip the pauses you would naturally take in front of people. Nerves push the other way and speed people up. Which effect dominates depends on the speaker. Time two rehearsals, take the longer one, and check it against how the real thing went so you learn your own bias.
Does the estimate include time for slides?
Only through the pause allowance. If your talk has a small number of slides that you talk over continuously, the default twelve percent covers the transitions. If it has a slide that you display in silence while people read it, or a demonstration, that is dead time the word count cannot see, and you should either raise the allowance or add the seconds separately.