cotera

TikTok Video Performance Analyzer

Stop guessing which TikTok videos work. Pull every post from an account, rank by real engagement data, and extract transcripts from the winners to understand exactly what drives performance.

Performance analysisContent optimizationTranscript analysisPattern recognition

The Challenge

TikTok's native analytics show surface-level metrics, but they never explain why one video got 500k views while the next got 2k. Was it the hook? The topic? The time you posted? Without pulling actual transcripts from your winners and comparing them to your flops, you're optimizing blind. This prompt does the full analysis and tells you exactly what patterns to replicate.

What This Prompt Does

Pull All Posts

Retrieve every recent video from an account with full performance metrics

Rank by Engagement

Sort videos by real engagement rate, not vanity view counts

Extract Transcripts

Pull transcripts from top and bottom performers to compare what works

Surface Patterns

Identify hook formulas, optimal length, and content themes that drive results

The Prompt

The Prompt

Task

Use @TikTok/Get User PostsName it "TikTok/Get User Posts" and call it with @TikTok/Get User Posts to pull all recent videos from a TikTok account. For each video, use @TikTok/Get Post DetailsName it "TikTok/Get Post Details" and call it with @TikTok/Get Post Details to get full performance metrics including plays, likes, comments, and shares. Identify the top-performing posts by engagement rate, then use @TikTok/Get TranscriptName it "TikTok/Get Transcript" and call it with @TikTok/Get Transcript to extract transcripts from the winners to understand what content patterns, hooks, and scripts are driving the best results.

Example: Pull the last 50 posts from our TikTok account, find the top 5 by engagement rate, and extract their transcripts so we can replicate what works.

Input

The user will provide:

  1. TikTok username or account URL to analyze
  2. Optional: number of posts to analyze (default: all recent, up to 50)
  3. Optional: specific metrics to focus on (views, engagement rate, shares)
  4. Optional: date range for analysis

Example: "Analyze @duolingoName it "duolingo" and call it with @duolingo on TikTok. Pull their last 40 videos, rank them by engagement rate, and extract transcripts from the top 10."

Context

Data Collection

  1. Use @TikTok/Get User PostsName it "TikTok/Get User Posts" and call it with @TikTok/Get User Posts to retrieve the account's recent video catalog
  2. Use pagination to pull the full set of requested posts
  3. For each post, record:
    • Video ID, creation date, description, hashtags
    • Play count, like count, comment count, share count
    • Video duration

Performance Analysis

  1. Use @TikTok/Get Post DetailsName it "TikTok/Get Post Details" and call it with @TikTok/Get Post Details on each video for complete metadata
  2. Calculate for each video:
    • Engagement rate: (likes + comments + shares) / plays * 100
    • Virality score: shares / likes (higher ratio = more shareable content)
    • Conversation score: comments / likes (higher = more discussion-driving)
  3. Rank all videos by engagement rate to identify clear winners and underperformers

Statistical Analysis

Across the full post set, calculate:

  • Mean and median engagement rate — the median is more useful since viral outliers skew the mean
  • Standard deviation — to identify posts that are statistically above average
  • Engagement by video duration — does shorter or longer content perform better for this account?
  • Performance by posting day/time — are there patterns in when top posts are published?
  • Hashtag correlation — which hashtags appear on above-average posts vs below-average?

Transcript Deep Dive

  1. Use @TikTok/Get TranscriptName it "TikTok/Get Transcript" and call it with @TikTok/Get Transcript on the top 10 highest-engagement videos
  2. Analyze each transcript for:
    • Opening hook (first 3 seconds): what grabs attention?
    • Content structure: how is the information organized?
    • Call to action: how does the video end?
    • Word count and pacing: how many words per second?
    • Key phrases: are there recurring phrases across top performers?

Underperformer Analysis

  1. Also pull transcripts from the bottom 3 performers
  2. Compare against winners to identify:
    • Weak hooks that fail to retain viewers
    • Content that is too long or too short for the topic
    • Missing calls to action
    • Topics that the audience does not respond to

Output

Account Performance Overview:

  • Total posts analyzed: [N]
  • Average engagement rate: [X]%
  • Median engagement rate: [Y]%
  • Best-performing post engagement rate: [Z]%
  • Total views across all analyzed posts

Performance Distribution: | Tier | Posts | Avg Engagement Rate | Avg Views | Common Traits | |------|-------|---------------------|-----------|---------------| | Top 10% | | | | | | Middle 80% | | | | | | Bottom 10% | | | | |

Top 5 Posts: For each:

  • Video link, date posted, description
  • Play count, likes, comments, shares, engagement rate
  • Transcript excerpt (hook + key section)
  • Why it worked (analysis of hook, format, topic)

Bottom 3 Posts: For each:

  • Video link and metrics
  • Transcript excerpt
  • Why it underperformed (weak hook, wrong format, topic miss)

Content Patterns That Win:

  • Hook formulas that appear in top performers
  • Optimal video duration for this account
  • Best posting days and times
  • Hashtag recommendations
  • Content themes that drive above-average engagement

Actionable Recommendations:

  • Scripts or hook templates to replicate based on transcript analysis
  • Topics to double down on
  • Topics to stop creating content about
  • Format changes to test

Example Usage

Try asking:

  • →"Analyze our TikTok account and find which videos performed best in the last 30 days"
  • →"Pull transcripts from our top 5 TikTok videos and tell me what hooks they all share"
  • →"Compare our best and worst performing TikTok posts and explain why the winners worked"