Glossary

What each YouTube metric actually measures

Four ways of answering “how well did this video do”, because the four questions behind it are genuinely different. What separates them is the baseline — what each figure is divided by — so that is what this page leads with.

Two of the figures below are both a multiple of a median, both commonly written “3×”, and they are not the same measurement. One divides by the other videos a search happened to return; the other divides by the channel's own upload history. Neither is wrong. Confusing them is, and most tools never draw the line.

Everything here is computed from values the YouTube Data API reported. Nothing on this page is an estimate, and no search volume, RPM or revenue figure is inferred anywhere in the product — those numbers are not published by YouTube, so we do not print them.

Views per day

Total views divided by the number of days since the video was published — a lifetime average, not a current rate.

Measured against
Nothing. It is one video's own figure, not a comparison.
Where you see it
Keyword research, top videos, channel discovery, the watchlist feed, saved videos.
What it cannot tell you
It carries a video's launch week forever. A video that did well two years ago and nothing since still reports a healthy views-per-day, so it cannot tell you what is happening now.

Search-sample outlier factor

A video's views divided by the median views of the other videos by the same channel that turned up in the same search.

Measured against
The channel's other videos in that one result set. At least three are needed; below that no factor is reported rather than a ratio from one or two videos.
Where you see it
Keyword research, top videos, and the outlier explorer, where 3× or more is the default threshold.
What it cannot tell you
The baseline is whatever the search happened to return, not the channel's upload history. A different query can produce a different factor for the same video, and both are correct for their own sample.

Channel-history outlier factor

A video's views divided by the median views of its own channel's recent uploads, measured separately for Shorts and long-form so the two are never averaged together.

Measured against
The channel's 50 most recent uploads in the same format. A channel needs 5 uploads at least 14 days old in that format before any factor is shown, and 2× or more is flagged as an outlier.
Where you see it
The channel watchlist and your own scoreboard — the same measurement, pointed at other people's channels and at yours.
What it cannot tell you
A video with no factor has no baseline behind it, which is not the same as a low factor. Shorts are counted as 180 seconds or less; a video whose length YouTube has not reported gets no format and so no baseline.

Measured velocity

The views a video actually added between two readings of it, divided by the days between those readings.

Measured against
The same video's previous reading. It is a change over time, not a ratio against anything else.
Where you see it
Channel trends, for channels you have connected.
What it cannot tell you
It needs two readings at least a day apart, so a newly synced channel shows nothing until the sync after next. Only your own connected channels are read this way.

Why a saved figure still means something later

A saved video records which of the two outlier factors it was measured with, so a number kept today still means the same thing when you come back to it. That sounds like a detail and it is the difference between a research note and a screenshot.

Where the rest of it is written down

The formulas behind the scores — including the opportunity score, and every place a figure is labelled Measured, Derived or Estimate — are at how our data and scores work. The craft frameworks the generators run are at the methods behind each tool.

Every one of these runs on the free plan

Plans differ by volume, not by which measurements you get. The free account is a working account with a one-time credit grant, not a demo.