Writing Video Outlines That Don't Lose Viewers
Most outlines are just a list of talking points in the order they occurred to you. That's fast to write, but it treats every point as equally interesting, and the retention graph is usually the first place you find out that section three wasn't. A creator who spends real time researching a topic can write an outline that's factually thorough and still loses a third of the audience at the exact point where the outline stops earning attention and starts just delivering information.
Warning
A list of talking points tells you what you'll say. It doesn't tell you where a viewer is most likely to click away, which is the actual job of an outline.
An outline that doesn't flag its own weak section isn't finished, it's just a list.
Why talking points aren't the same as pacing
Writing down what you want to cover is the easy half of outlining. The harder half is deciding how long each part deserves and which one is most likely to drag, and a plain list of points does neither. A talking-points list treats a genuinely fascinating discovery and a necessary-but-dry piece of setup with the same visual weight, even though a viewer will experience them very differently.
Consider an eight-minute video on why a sourdough starter keeps collapsing, outlined as five roughly equal sections: introduction, the science of fermentation, feeding schedule basics, the actual mistake, and the fix. Written as a flat list, all five sections look equally important. In practice, the feeding schedule section is necessary background but not inherently interesting, and it's exactly the kind of section a viewer skips ahead through or abandons the video during, unless it's either compressed sharply or reframed around the specific mistake it's setting up.
Before: list five things to cover, film in that order, find out in editing that the third point ran long and slow.
After: give Claude the topic and target length, get back five paced sections with timestamp estimates and a flag on the section most likely to lose viewers, before you've filmed anything.
A process for outlining with pacing built in
- 1
Give Claude the topic and the target length
State the actual runtime you're aiming for, not "make it good length." An 8-minute video and a 20-minute video need a different number of sections and a different amount of time per point.
- 2
Ask for a fixed number of sections with timestamp estimates
Five sections with rough timestamps forces a real pacing decision instead of an open-ended list. You can always merge or split sections after, but starting with estimates makes the tradeoffs visible.
- 3
Ask which section is most likely to lose viewers
Ask directly, not "does this look okay." The section most likely to drag is usually the one with the most setup and the least payoff, and that's worth knowing before you film it, not after.
- 4
Cut or reorder based on the flag, not your instinct
If Claude flags the section you were least excited to film anyway, that's not a coincidence. Move it earlier while attention is highest, or cut it down to the one detail that actually matters.
I'm making an 8-minute YouTube video on "why my sourdough keeps collapsing." Outline it in five sections with a timestamp estimate for each, and flag which section is likely to lose viewers.
”For a video that has a genuinely strong payoff moment, the actual reveal of what was going wrong, it's worth asking Claude to consider whether teasing that payoff earlier would help hold attention through the slower setup section, rather than saving all the payoff for the end.
Section 4, the actual mistake, is the strongest payoff in this video. Would briefly teasing that reveal during the slower feeding-schedule section help retain viewers through it, and if so, how would you word a short tease without giving away the full answer?
”
Inside Claude Tutorial
Structuring for where attention actually drops is transferable.
Deciding how much time each part of something deserves, and flagging the weak stretch before an audience finds it, applies to more than video outlines. The app has a full lesson on it, with practice that carries over to any paced format.
Reworking a section instead of cutting it
A flagged section isn't always a cut. Sometimes it just needs to earn its place differently, especially when it's genuinely necessary information the viewer needs before the payoff makes sense. Cutting a section that's actually load-bearing for the rest of the video trades a pacing problem for a comprehension one, which is usually the worse trade.
Tip
Ask Claude for a tighter version of the flagged section specifically, not a full re-outline. A section that's dragging usually has one unnecessary detour in it, and a tighter version keeps the useful part while cutting that detour.
Section three of this outline covers my starter's feeding schedule, and you flagged it as likely to lose viewers. Give me a tighter version that keeps the one detail that actually explains the collapsing, and cuts everything else.
”It's also worth checking whether the flagged section could be reframed rather than just shortened, presented as a quick problem-solving moment instead of a straight explanation, since a section that feels like it's building toward something tends to hold attention better than one that reads as pure background information, even at the same length.
Instead of just shortening section three, could it be reframed as a quick "here's the mistake I almost missed" moment rather than a straight explanation of feeding schedules? Show me what that version would sound like.
”Continue reading
- Sharpening a Hook Before You Film It: the same "test it, don't just trust it" approach, applied to the three seconds before the outline even starts.
- Turning Comments Into Your Next Five Videos: where the topic for an outline like this one comes from in the first place.
- Repurposing One Video Into a Week of Content: once the paced version is filmed, what to do with it beyond the one platform you made it for.