01.6 — Feature · Coming soon· 4 min read
Analytics.
Klyph pulls performance data from every connected platform and normalizes it into one language: views, completion rate, shares, saves, and retention curve.
Careful
Analytics is not live yet — this page previews the target design.
§ 1
Metrics model
Platforms disagree on definitions — Klyph maps them into a single, honest schema so you can compare a TikTok to a Reel without guesswork.
| Field | Type | Description |
|---|---|---|
| views | number | Unique plays over 1s across platforms |
| completion_rate | 0–1 | Fraction of viewers who watched to the end |
| avg_watch_s | number | Average watch time in seconds |
| shares | number | Shares / sends / retweets / reposts |
| saves | number | Bookmarks where supported |
| hook_retention | 0–1 | Fraction still watching at 3s |
§ 2
Retention curves
For every post, Klyph stores a second-by-second retention curve (where available) and overlays it on the clip. Drops map back to exact moments in the edit — so you learn what costs you viewers.
Tip
Win/loss against your own library.
Klyph flags new posts as winners or laggards vs. your rolling 14-day median within 24 hours of posting.
§ 3
Exports
CSV and Parquet exports on paid plans.