Audience Retention Drop-Off Diagnosis Tree
Find why viewers leave and what to change.
The Open buttons send the prompt text as currently shown below to that site only when you click. Fill in the placeholders first.
Diagnose my retention problem. Video: [VIDEO_TOPIC] of length [VIDEO_LENGTH]. Retention data: [RETENTION_DATA] (such as percent retained at 0:30, 1:00, 25%, 50%, 75% and notable dips). Script or outline: [PASTE_OUTLINE]. Use a decision tree: IF the largest drop is in the first 30 seconds THEN check: does the opening deliver the title's promise quickly? Is there a long intro, logo animation, or channel plug? Is the first visual interesting? IF there is a steady decline with no spikes THEN check: pacing, repeated information, unclear structure. IF a sharp drop happens at a specific moment THEN identify what happened at that time (sponsor read, change of topic, long explanation) from the outline. IF viewers drop near the end THEN check for a false ending or a slow close. IF there are re-watch spikes THEN find out why and make more of that. For each branch triggered by my data, give two likely causes, an edit to try in this video if it can be re-cut, and a rule to use in future videos. Compare with the benchmarks only as rough guidance. Ask what I think was happening at the largest drop.
Fill in before use
Optional: type your details and the prompt above updates. Nothing is saved or sent.
Replace each bracketed placeholder with your own details, then paste into Any AI or another assistant.