What counts as YouTube watch time?

Time viewers spend actively watching your videos. Pauses don't count, fast-forwarding doesn't count, and viewers who close the tab without watching don't count. Both logged-in and signed-out viewers contribute to watch time.

How is YouTube watch time calculated for monetization?

Total watch hours from public, non-Shorts videos over the trailing 365 days. Private and unlisted videos don't count. Shorts watch time has its own separate threshold (10M Shorts views in 90 days). The 4,000 hours can come from any combination of long-form videos.

Does my own watch time count toward my channel's watch hours?

Your own views are filtered out of public metrics and don't contribute to YPP eligibility. Watching your own video to inflate metrics doesn't work — YouTube has been detecting this for years.

Why is my YouTube watch time dropping?

Two common causes: a recent video underperformed and dragged down the rolling average; or older videos that were carrying the channel are aging out of the 28- or 365-day windows. Open Studio's Reach tab and compare last 28 days to the previous 28 to find the change.

Can YouTube remove watch time from your account?

Yes, in two cases: detected fake/bot views, and views on videos that were removed by YouTube for policy violations. Views from real human audience aren't taken away after the fact.

Blog·May 6, 2026

YouTube watch time: how it actually works and how to grow it

Watch time is the most-cited YouTube metric and one of the most misunderstood. Here's what it really measures, where it matters, and the levers that move it.

By Chapter Generator team·8 min read

"Watch time" gets used as if it were one metric. It's three different numbers wearing the same name, and they don't all matter for the same thing. Total watch hours matter for monetization eligibility. Average view duration matters for ranking. Average percentage viewed matters for retention diagnosis. Confusing them is why so much "watch time advice" misses.

The three numbers

Total watch hours

Sum of every viewer's watched seconds across all videos. Big number, useful mostly for the YPP eligibility threshold (4,000 hours from long-form, public videos in the trailing 365 days).

Once you're past the YPP threshold, total watch hours stops being a useful operational metric. It's a vanity stat from that point forward — a channel with 1M watch hours doesn't rank better than a channel with 500k.

Average view duration (AVD)

The average number of seconds the typical viewer watched per video. AVD is what the YouTube algorithm uses for ranking. Combined with click-through rate, it determines reach.

AVD scales with video length. A 5-minute video with 2-minute AVD is at 40% retention. A 20-minute video with 5-minute AVD is at 25% retention but more total seconds. Both are signals; the algorithm uses both depending on context.

Average percentage viewed (APV)

The percentage of the video the typical viewer watched. The cleanest single retention signal — independent of length, easy to compare across videos. APV above 40% is generally strong.

Where each number matters

Goal	Metric to track
YPP eligibility	Total watch hours (4,000 trailing 365 days)
Algorithmic ranking	AVD + APV
Diagnose specific videos	Retention curve + APV
Compare videos of different lengths	APV (length-normalized)
Channel-level health	Trend in AVD over time

The five highest-leverage ways to grow watch time

1. Tighten the first 30 seconds

Audience retention curves cliff in the first half-minute. Every viewer you save through that cliff multiplies into the rest of the video's watch time. The single highest-leverage edit on any video is making the first 30 seconds tighter.

2. Add chapters

Chapters lift retention by helping viewers re-find sections (which adds re-watch time) and by signaling where to skip if they were going to leave anyway (which keeps them in-video instead of out of it). Easy win — minutes of work, durable benefit. Run any URL through our chapter generator if you don't have chapters yet.

3. End with a clear next watch

End screens, suggested-watch callouts in the final seconds, and a deliberate next video on your channel that pairs with this one. Each one shifts the typical viewer from "close the tab" to "keep watching," which is the strongest single algorithmic signal you can produce.

4. Match length to topic

Don't pad to hit a target length. Don't cut short to fit a target length. The optimal length is whatever the topic deserves at the pace your viewers want. APV is the diagnostic — if your APV is below 30%, your videos are too long; above 60%, they could probably be longer.

5. Build playlists

Playlists auto-play to the next video, which adds a full new video's worth of watch time for every viewer who lets the autoplay through. Playlists also rank in search separately — a well-titled playlist drives external watch time you wouldn't otherwise get. See our playlists strategy guide.

Things that don't lift watch time meaningfully

Begging viewers to watch to the end. Viewers ignore this; the retention curve doesn't change.
Hiding key info to trick viewers into staying. The retention curve dips at the moment they realize the trick. Net negative.
Adding sponsor reads in the middle. Mid-roll sponsor segments consistently produce a small retention dip — worth it for revenue, but not for watch time.
Posting more frequently. Total watch hours rises only because of the cumulative effect; AVD on individual videos can drop if quality slips.
Buying watch hours. Detected, removed, can result in account termination.

How to read your watch-time data in 5 minutes

YouTube Studio > Analytics > Engagement.
Look at Average view duration and Average percentage viewed for the last 28 days. Compare to the previous 28.
Click into the top-performing video. Read its retention curve. Note the moments where retention spiked or held flat.
Click into the worst-performing video. Read its retention curve. Note where it cliffed.
Pick one specific edit to make on the next video based on the two curves. Just one.

Done weekly, this routine compounds. The creators with rising AVD over months and quarters are doing roughly this — not anything clever.

FAQ

What counts as YouTube watch time?: Time viewers spend actively watching your videos. Pauses don't count, fast-forwarding doesn't count, and viewers who close the tab without watching don't count. Both logged-in and signed-out viewers contribute to watch time.
How is YouTube watch time calculated for monetization?: Total watch hours from public, non-Shorts videos over the trailing 365 days. Private and unlisted videos don't count. Shorts watch time has its own separate threshold (10M Shorts views in 90 days). The 4,000 hours can come from any combination of long-form videos.
Does my own watch time count toward my channel's watch hours?: Your own views are filtered out of public metrics and don't contribute to YPP eligibility. Watching your own video to inflate metrics doesn't work — YouTube has been detecting this for years.
Why is my YouTube watch time dropping?: Two common causes: a recent video underperformed and dragged down the rolling average; or older videos that were carrying the channel are aging out of the 28- or 365-day windows. Open Studio's Reach tab and compare last 28 days to the previous 28 to find the change.
Can YouTube remove watch time from your account?: Yes, in two cases: detected fake/bot views, and views on videos that were removed by YouTube for policy violations. Views from real human audience aren't taken away after the fact.