YouTube Most Replayed Clip Finder (Heatmap to Clips) avatar

YouTube Most Replayed Clip Finder (Heatmap to Clips)

Pricing

from $20.00 / 1,000 video with clips

Go to Apify Store
YouTube Most Replayed Clip Finder (Heatmap to Clips)

YouTube Most Replayed Clip Finder (Heatmap to Clips)

Turns YouTube's 'most replayed' heatmap into ranked clip segments (start, end, score, suggested title) for Shorts, Reels and highlight reels. Also returns the raw 100-point heatmap. No API key.

Pricing

from $20.00 / 1,000 video with clips

Rating

0.0

(0)

Developer

Craig Armstrong

Craig Armstrong

Maintained by Community

Actor stats

0

Bookmarked

2

Total users

1

Monthly active users

2 hours ago

Last modified

Features

Limited permissions

Share

YouTube Most Replayed Clip Finder

Give it YouTube video URLs and it returns the moments viewers replay most, already cut into clip segments: start and end times, a replay score, a suggested title and a timestamped link, ranked best first. It also returns the raw 100-point "most replayed" heatmap so you can draw your own curve. No API key, no login.

Built for people who turn long videos into Shorts, Reels and TikToks, for highlight reels, for chaptering, and for AI agents that need to know where the good part of a video is before transcribing or cutting it.

Output

One row per video:

{
"videoId": "Y8efWZ2M1y8",
"title": "How To Trade TRENDLINES (Full Guide)",
"channel": "Tori Trades",
"lengthSeconds": 2725,
"viewCount": 561411,
"hasHeatmap": true,
"peakSeconds": 2180,
"clipCount": 5,
"clips": [
{ "rank": 1, "start": 2137.3, "end": 2199.7, "durationSeconds": 62.4, "startLabel": "35:39", "endLabel": "36:41",
"peakIntensity": 1.0, "meanIntensity": 0.828, "score": 0.828,
"suggestedTitle": "How To Trade TRENDLINES (Full Guide) — most replayed #1 (35:39–36:41)",
"clipUrl": "https://www.youtube.com/watch?v=Y8efWZ2M1y8&t=2137s" }
],
"heatmap": [ { "start": 0, "duration": 27.3, "intensity": 0.31 }, "… 100 points" ]
}

intensity is YouTube's own normalised replay intensity (1.0 = the single most replayed moment of the video).

Input

FieldDefaultWhat it does
videos–URLs (watch?v=, youtu.be/, Shorts, embeds) or bare 11-character IDs
maxClipsPerVideo8Clips returned per video, best first
minIntensity0.5A heatmap point joins a clip when its intensity is at least this. Lower it (0.3) for flat videos, raise it (0.7) for just the spikes
minClipSeconds / maxClipSeconds8 / 60Short peaks are stretched to the minimum; long peaks are split into chunks of the maximum (so clips fit a Short)
padSeconds2Padding before and after each clip so cuts don't land mid-word
includeHeatmaptrueInclude the raw curve
proxyoffApify Proxy, if YouTube blocks direct requests from the platform

When there is no heatmap

YouTube only shows a "most replayed" graph for videos with enough views (roughly 50k+ and not every video even then). Those rows come back with hasHeatmap: false, clips: [] and a note, so a batch never fails because of one quiet video.

How clips are scored

Consecutive heatmap points above minIntensity form a peak. Each peak (or chunk of a long peak) is scored by peakIntensity × meanIntensity, so a tall, sustained peak outranks a brief spike. suggestedTitle is the video title plus the clip's rank and time range, a placeholder for your own.

Limits

  • Reads the public watch page. If YouTube changes its page structure the Actor is updated; if it rate-limits the platform, switch proxy on.
  • Heatmap data is YouTube's aggregate of viewer behaviour; it says where people rewatch, not why.

Cost

Pay per event, no subscription. You pay only for the videos a run checks.

EventPrice
Run start, once per run$0.001
Video with a heatmap, clips returned$0.02
Video without a heatmap$0.002

Ten videos that have heatmaps cost about 20 cents. A video with no heatmap costs a fifth of a cent.

Run it from code

curl -X POST "https://api.apify.com/v2/acts/cryptoretirement~youtube-most-replayed-clip-finder/run-sync-get-dataset-items" \
-H "Authorization: Bearer $APIFY_TOKEN" \
-H "Content-Type: application/json" \
-d '{"videos":["https://www.youtube.com/watch?v=dQw4w9WgXcQ"],"maxClipsPerVideo":5}'

The response is the dataset as a JSON array. The API menu on this page has ready-made snippets for JavaScript, Python and MCP clients.

More Actors from this developer