Before He Says Why

By Noam Shemla ·

By Noam Shemla · Speeches That Lasted

TL;DR We measured Simon Sinek’s 2009 talk “How Great Leaders Inspire Action” with the same instrument our members record into, this time using the recording itself as well as its transcript. 166 words a minute scores 76 out of 100 on pace: still on the fast side of the band, the only entry in this series that is, though less dramatically fast than a rougher transcript-only estimate first suggested. The real find is in the pitch data. His voice is measurably highest in the opening ninety seconds (the framing questions, before the Golden Circle is even named) and settles lower through the “why” refrain his own theory says should be the emotional peak.

This is the first entry in the series built from a real recording end to end: audio, decoded and analysed the same way a member’s Practice Studio clip is. That changes two things. The pace figure gets more precise, and three axes that were absent before (pauses, pitch, volume) are real numbers now.

A talk on TED’s licence

“How Great Leaders Inspire Action” was recorded at TEDxPugetSound in September 2009 and is, by TED’s own count, one of the most-watched talks the organisation has ever published. It remains under TED’s copyright, licensed for non-commercial sharing with attribution. That licence does not stretch to republishing the transcript in full, so, as with the other three entries in this batch, we work from structure, timing and short attributed fragments.

Unlike the King entry above, we were able to confirm a legitimate source for a video figure here: TEDx Talks’ own YouTube channel, verified before use via YouTube’s oEmbed endpoint against a channel-mismatched reupload we found and rejected.

The first entry in this series that’s too fast

The reason isn’t mysterious once you compare this talk’s conditions to the rest of the series. A TED talk is rehearsed to a strict time limit, delivered to a seated audience that has been asked not to applaud mid-talk, with none of Selma’s thirty ovations or the March on Washington’s call-and-response. Sinek isn’t pausing for a crowd, and he isn’t narrating physical actions the way a product demo does. He talks, close to continuously, for eighteen minutes. Take away the interruptions that dragged every earlier entry’s elapsed-time pace down, and what’s left is a fast talker, measured as one.

Where his voice lifts

Forty-eight real pauses over 0.5 seconds, timed from word-level gaps instead of inferred from sentence punctuation. The single longest, 8.3 seconds, falls at 2 minutes 3 seconds in, during the opening framing (“Why is it that Martin Luther King led the Civil Rights Movement?”), far from the repeated “why” line his own thesis centres on. Pace swings hard across the talk too, from 66 words a minute in the first ten seconds up to 228 in a single window past the ten-minute mark, and the timeline above shows that directly instead of flattening it into one average.

  • 149 Hz mean pitch. Highest in the first 30 seconds (175 Hz); settles between 130 and 150 Hz for most of the talk.
  • ±60 Hz pitch variation. Not monotone by our threshold. The variation is real, but it sits away from where the thesis would predict.
  • 0.25 avg volume. Range 0 to 1 across the full recording.
  • 48 real pauses. 0.5s or longer, from word-timing gaps. A floor, not a total.

The acoustic axes this entry didn’t have before, measured from the recording.

One finding didn’t match the hypothesis going in. The talk’s own argument is that inspiring communication is limbic, emotional, felt before it’s reasoned through, so the plan for this entry was to check whether Sinek’s own voice gets more animated exactly where he delivers the repeated “why” line his theory centres on. It doesn’t, cleanly. The highest-pitch stretch of the whole talk is its first thirty seconds (average 175 Hz, against a talk-wide mean of 149), while he’s still asking the opening questions (“Why is Apple so innovative?”), before the Golden Circle has a name. By the time the refrain itself recurs, in the passages around 6 and 13 minutes in, pitch has settled into a narrower, lower band. If anything, the data says the opening hook is where he is most animated, and the repeated thesis line is delivered more levelly, closer to how he delivers everything else. A “speak with more energy on your key line” coaching tip would predict something different, and we take the result at face value instead of reading in the one we expected to find.

Punchy lines around long stories

The range is wide by design. This is a talk built on short, punchy assertions (“Why him?” “Why? How? What?”) interrupted by longer story-telling passages, like the 55-word sentence recounting Samuel Pierpont Langley’s failed race to build the first airplane. Short-long-short is a deliberate rhythm here.

A refrain that tests his own argument

That fourth lane is still the interesting one. Sinek’s talk argues, explicitly, that people remember and act on repetition of a small, consistent message rather than a wide vocabulary of features and facts, and the talk’s own vocabulary diversity, at 27.8 percent, is near the low end of this series. We don’t mean that as a criticism of the talk. By his own stated theory, it may be the point: a talk built to be remembered and repeated by its audience is, structurally, going to reuse its own key phrases rather than vary them.

Caveats the recording leaves in place

Three more recordings, and the flattest line we have measured

We later measured three more Sinek recordings across nine years: a second TED talk, an interview and a keynote. Placed against the other four speakers, what stands out is how narrow his band is.

Read one way that is the most enviable result in the corpus: he has the highest floor of the five and never produces a weak recording. Read the other way the same chart shows the trade, because the two speakers with the widest bands are also the two who reach the highest single scores measured anywhere here. Whether a repeatable delivery is the goal or the thing to escape is not a question this instrument can settle. It can only show that in this corpus, the speakers who never drop are also the speakers who never spike.

Check which side of the band you are on

“Slow down” is not the universal fix a pace score implies. Every other entry in this series suggested a speaker running long and dragging under our band’s floor, where the natural reading is “tighten it up.” This entry is the opposite case: a talk that’s tight, rehearsed and fast, scoring below full marks for running too quickly rather than too slowly. If your own dashboard shows a pace score under 100, don’t assume the question is “am I too slow”. Check which side of the band you’re on. Separately, “put more energy into your key line” is intuitive advice this data doesn’t support here: his most animated stretch, by pitch, is the opening hook, ahead of the repeated thesis. Where the energy goes first may matter more than where the emphasis lands.

The same audio pipeline our members record into: word-timed pace and pauses from timed captions, pitch and volume from the decoded recording.

Measured

  • Word count and pace against real elapsed time
  • Real pauses (0.5s+) from gaps between timed words
  • Pitch: mean 149 Hz, ±60 Hz, not monotone
  • Volume: average 0.25 across a 0 to 1 range
  • Filler and hedging counts and rates
  • Vocabulary diversity: 27.8 percent of words distinct
  • Positions of every sentence-opening phrase that repeats three times or more

Inferred, not measured

  • Word timings are spread evenly inside each caption cue, so pauses inside a single cue can’t be seen
  • Sentence boundaries follow the punctuation of the timed captions

Sources