# Reverse-engineer why a video took off

> Most-replayed segments plus comment sentiment show what actually hooked viewers — pattern-match it before your next upload.

- For: Creator
- Category: Content Strategy & Creators
- MCP tools: deep_search, run_youtube_tool
- HTML version: https://veridive.com/mcp/use-cases/why-that-video-took-off

## The problem

A video in your niche just did twenty times the channel's usual numbers, and everyone has a theory: the thumbnail, the title, the algorithm's mood. Theories are free. What actually happened is measurable — which moments viewers replayed, what the comments latched onto, how the video opens versus how its channel usually opens — and almost nobody checks, because checking means forensic work per video.

Breakouts are the closest thing YouTube has to published experiments: some variable changed and the audience responded. Read enough of them properly and you stop cargo-culting thumbnails and start seeing the real patterns — which is exactly the homework worth doing before your next upload, on someone else's results.

## How veridive MCP helps

**run_youtube_tool** runs the forensics on any public video, straight from a URL: `most-replayed` shows which segments viewers scrubbed back to — the actual hooks, not the assumed ones; `comment-sentiment` and `comment-summarizer` show what the audience responded to in their own words; `hook-grader` scores the opening against what makes openings work.

**deep_search** supplies the comparison set: the channel's ordinary videos and the niche's other breakouts, found at the transcript level. Run the same forensics across a handful and the pattern separates from the noise — was it the cold open, the topic's timing, the promise in the first fifteen seconds? You walk into your next script with evidence instead of superstition.

## How to run it

1. **Connect veridive MCP** — One-time setup: add the veridive MCP server to Claude, ChatGPT, Cursor or any MCP client and generate your API key. Takes about two minutes.
2. **Run the forensics on the breakout** — Most-replayed segments, comment sentiment, hook grade — the measurable anatomy of why it worked.

   ```
   Analyze this breakout video: most-replayed segments, comment sentiment summary, and grade the opening hook: <video URL> — use veridive's run_youtube_tool.
   ```
3. **Build the comparison set** — The same channel's ordinary uploads plus other breakouts in the niche — the controls that make the experiment readable.

   ```
   Find this channel's typical recent videos and two other breakout videos in the same niche — use veridive's deep_search.
   ```
4. **Extract the pattern** — Same forensics across the set, then the question that matters: what does the breakout do that the ordinary uploads don't?

   ```
   Compare the forensics: what did the breakout do differently — opening, structure, promise, replay peaks? Give me the pattern, not the folklore.
   ```

## Problems it solves

- Breakout post-mortems run on theories because the measurable evidence takes forensic work per video.
- Replay peaks — the audience literally pointing at what hooked them — sit unread in every video.
- Comment sections hold the audience's own explanation, buried under a thousand entries nobody reads.
- Without a comparison set, every breakout 'lesson' is survivorship bias in a trench coat.

## Why it's valuable

- You learn from the niche's published experiments before spending your own upload slots.
- Replay data replaces hook folklore: the moments viewers actually rewound, per video.
- Comment sentiment shows what the audience thinks worked — often not what creators assume.
- The comparison-set habit turns one-off luck into repeatable pattern-reading.

## Example prompts

```
Most-replayed segments + comment sentiment + hook grade for this breakout: <URL> — use veridive's run_youtube_tool.
```

```
Find this channel's typical videos and two other niche breakouts for comparison.
```

```
What did the breakout do differently? Pattern, with the evidence.
```

## FAQ

**Which of the 32 tools does the forensic pass use?**

most-replayed for the replay peaks, comment-sentiment and comment-summarizer for the audience read, hook-grader for the opening. All run on any public video from its URL — no import needed.

**Does most-replayed work on every video?**

It reads the replay data YouTube exposes, which is available on most videos with meaningful traffic. On very fresh or very small uploads there may be too little data — the tool tells you rather than guessing.

**Can I run this on my own videos?**

Absolutely — your own breakout (or flop) is the most instructive subject, since you know what you intended. The gap between intended hook and replayed hook is usually the lesson.

## Related use cases

- [Turn one video into a week of content](https://veridive.com/mcp/use-cases/one-video-week-of-content.md) — Blog post, X thread, LinkedIn post, newsletter section, Shorts script — one upload, six assets, same day.
- [Mine comments for your next ten videos](https://veridive.com/mcp/use-cases/comment-section-idea-mine.md) — Comment summarizer surfaces the questions your audience keeps asking under your videos and your rivals' — answered demand, pre-validated.
- [Build a swipe file of the best hooks in your niche](https://veridive.com/mcp/use-cases/hook-swipe-file.md) — Pull the openers of breakout videos into a transcript swipe file and study them line by line.

---

Part of the veridive MCP use-case library: https://veridive.com/mcp/use-cases.md
Server docs for agents: https://veridive.com/mcp.md