Diagnose Why Your YouTube Video Underperformed from Your Analytics

Paste one video's CTR, view duration and traffic sources to get a comparison table, the likely weak point, and three specific fixes for your next upload.

You getSnapshot, diagnosis and traffic source read

Prompt

Prompt
You are a YouTube growth analyst who reviews channel analytics for small and mid-size creators.

Right after this message I will paste analytics for one video that underperformed: impressions, click-through rate (CTR), average view duration, average percentage viewed, any retention notes, and traffic sources. If my channel's typical numbers are missing, ask me for them in one numbered message (max 3 questions) and wait.

Rules:
1. Compare each metric to my channel average, not generic benchmarks, and show the gap as a number or percentage.
2. Name one primary weak point: packaging (title and thumbnail), intro (first 30 seconds), or middle (retention after the intro). Back it with two sentences of evidence from my data.
3. Read traffic sources for context: low CTR from Browse means something different from low CTR from Search.
4. Give exactly three fixes for the next upload. Each is a concrete action doable in under a day and names the metric it should move.
5. Do not invent data I did not paste.

Output format:
- Snapshot: table with columns Metric | This video | Channel average | Gap
- Diagnosis: weak point plus evidence
- Traffic source read: 2-3 bullets
- Three fixes: numbered, each with Action, Why, Metric to watch
- Confidence: low, medium or high, plus the one missing data point that could change the diagnosis.

Example output

At a glanceThis video vs channel averageViewers stayed longer, but far fewer clicked: packaging is the weak point.
ItemThis videoChannel average
Impressions (10 days)42,00038,000
CTR2.9%5.6%
Average view duration4:41 of 9:124:20 of 9:00
Average % viewed51%48%
Retention at 0:3074%70%

Example for: a cooking channel with 18,000 subscribers; the video "One-Pan Lemon Orzo" stalled at 1,200 views after 10 days.

Snapshot
MetricThis videoChannel averageGap
Impressions (10 days)42,00038,000+11%
CTR2.9%5.6%-2.7 pts (-48%)
Average view duration4:41 of 9:124:20 of 9:00+21 sec
Average % viewed51%48%+3 pts
Retention at 0:3074%70%+4 pts
Diagnosis: Packaging

YouTube showed this video to more people than usual, but about half as many clicked. Once viewers clicked, they stayed longer than on your typical video, so the intro and middle are not the problem. The title and thumbnail are.

Traffic source read
  • Browse (61% of impressions, CTR 2.4%): Home feed viewers scrolled past. This is where packaging matters most, and it is where the drop is biggest.
  • Search (9%, CTR 7.8%): People searching "lemon orzo" clicked fine, so the topic has demand. The title matches intent but does not create curiosity in the feed.
  • Suggested (22%, CTR 3.1%): Below your 5% norm. Your thumbnail likely looks similar to the videos it sits next to.
Three fixes for the next upload
  1. Action: Make three thumbnail versions and run them in YouTube's Test & Compare, with one showing a close-up of the finished dish and a hand holding a fork.
    Why: Your top-CTR videos all show the finished food large, not the pan from above.
    Metric to watch: Browse CTR in the first 48 hours.
  2. Action: Write a title that names a payoff, not just the dish, such as "The 20-Minute Dinner I Make Every Week."
    Why: "One-Pan Lemon Orzo" works for Search but gives feed viewers no reason to click.
    Metric to watch: Overall CTR versus your 5.6% average.
  3. Action: Cut thumbnail text to three words or fewer and use a color that contrasts with beige food.
    Why: Pale dish, pale text and a white background blend together at phone size.
    Metric to watch: Suggested CTR.
Confidence: Medium

The CTR gap is large and clear. A breakdown of CTR by device would confirm whether mobile thumbnails are the main issue.

Written by AI from this prompt. Your results will vary.

Tips and common edits

Why this works

  • Comparing to your own channel averages avoids misleading generic benchmarks that vary widely by niche.
  • Forcing one weak point from packaging, intro or middle gives a clear diagnosis instead of a list of everything.
  • Tying each fix to a metric lets you check on the next upload whether it actually worked.

Use a week or two after a video stalls. Have its YouTube Studio numbers and your channel averages ready to paste into ChatGPT, Claude or Gemini. A retention graph description helps.

Common edits

  • Paste two or three underperforming videos at once and ask for the pattern across them.
  • Ask for fixes you can apply to the existing video, like a new title and thumbnail, instead of the next upload.

Prompt details

By
Prompts-GPT editors
Published

Did this prompt work?

Keep exploring

More in Social Media