autopsies / record #006 / September 2024
Dead on arrivalContact lost at seven seconds.
The opening held. The setup collapsed. A record of the precise moment attention withdraws and the structural conditions that made it inevitable.
Verdict
Not killed by bad content. Killed by structure — the gap between what the opening promised and what the setup delivered. The opening created forward momentum. The setup stalled it.
By seven seconds, the audience had made a decision the work could not reverse. No recovery was observed. What remained in the back half was inertia, not engagement. The floor is not the same thing as an audience.
Where contact was lost
Attention held by segment:
The collapse at 0:07 is the event. Everything after is its consequence. The stable floor from 0:12 onward is inertia, not engagement.
Evidence collected
- Opening: direct address, no question framing. Cleared the initial test. Not the cause.
- Setup: static frame held for 4.1 seconds with no motion, no text change, no auditory event.
- Audio: no break near 0:07. Sound continued cleanly through the collapse.
- Visual: single cut at 0:03. No subsequent cut until 0:14. Eleven seconds without movement.
- Text: present in opening frame, absent from 0:03 through 0:12.
- Two written responses were substantive — the content, where seen, was relevant.
- 94% of views arrived in the first four hours. No secondary wave observed.
What held
- Audio quality was clean throughout. No collapse attributable to sound.
- The opening frame was not the failure point.
- The two written responses were substantive. Something landed for those who stayed.
- Caption structure survived. Two follows traced back to it, not to the work itself.
Probable cause
The setup violated the implicit contract the opening created. The opening produced direction — a promise that something was arriving. The setup offered stillness. Four seconds of stillness after three seconds of momentum is not a pause. It is the withdrawal of the signal the viewer was following.
When that signal goes quiet, the default decision is to leave. Not because the content was bad — most of them never reached it.
Archive references
| Ref | Type | Note |
|---|---|---|
| EXP-007 | Active experiment | Tests visual continuity hypothesis derived from this record |
| EXP-005 | Closed experiment | Contextualizes distribution behavior observed here |
| AUT-004 | Autopsy | Comparison case — the opening that held |
record #006 / dissected September 2024 / evidence appended