My live streams averaged 38 seconds: fixing low average view duration on a founder YouTube channel

Sunny Ray · Feb 13, 2026 · 5 min read

A founder channel that covers everything is a channel nobody watches past the first minute. I know because I pulled up my own numbers on February 13, live, with another creator watching, and they were bad.

"How many watch hours do you have?"

I was on a call with a Bitcoin creator friend who runs a focused channel. Early on he asked a question I had been avoiding:

How many watch hours do you have on your channel?

He made me open YouTube analytics on screen. Watch time: 85 hours in the last 28 days. For a channel with about 21,000 subscribers, that is thin.

Then he had me look at average view duration on my live streams. 38 seconds.

Thirty-eight seconds. People were clicking in, seeing what was on, and leaving before I finished my first thought. I had been streaming regularly and planning two or three streams a day. Volume was never the problem. Almost nobody stayed.

The comparison that stung

His channel had about 42,000 subscribers. His average view duration was 6 minutes and 2 seconds. His videos consistently pulled somewhere between 4,000 and 20,000 views.

Same broad topic area. Twice my subscribers, an average view measured in minutes instead of seconds, and steady numbers on every upload. He put the difference down to one thing: focus. His channel is about one subject, and every title promises one clear thing inside it. Mine was Bitcoin one day, AI the next, robotics after that, interviews in between.

What 38 seconds is telling you

YouTube defines average view duration for a live stream as the estimated minutes watched per view. So a low number means a lot of views that were barely views at all.

Live averages always run below recorded videos, because people drop in and out of streams. One streaming blog suggests that an average around 5 to 10 percent of the stream's length signals a drop-off problem. That is not an official benchmark and I would not treat it as one. But at 38 seconds on streams that ran far longer than ten minutes, I did not need a benchmark to know.

The deeper diagnosis came from what he saw on my channel page.

• Generic titles. One of mine was "Inside the Black Box." It sounds intriguing to me. To a stranger it promises nothing. What box? Why should they care?
• Thumbnails built around guests' faces. A face means something if you know the guest. To everyone else, it is just a face. His thumbnails sell the information inside the video.
• Mixed topics. The algorithm cannot figure out who to show the channel to, and a viewer who came for Bitcoin finds an AI episode next and leaves.
• Small branding fixes: drop "Show" from the channel name and use a current photo, not an old one.

His titles, by contrast, put a credible voice and one clear promise up front. Something along the lines of an expert explaining one specific Bitcoin idea. You know exactly what you get before you click.

The fix: one topic per channel

His strongest recommendation was structural. Split the channel. One channel that is only Bitcoin, built the way his is. A separate one for AI and robotics. Each with a clear audience, consistent expectations, and titles that promise one thing.

That is uncomfortable for a founder. My interests cross over, and the crossover is a big part of how I think. But the audience does not subscribe to how I think. They subscribe to a promise. A channel that keeps changing the promise teaches viewers to leave.

He also fixed something I had ignored: my microphone. Moving closer to the mic made what he called a massive difference to the sound, and he recommended an arm mount that costs about $100. Viewers forgive a lot. They do not forgive bad audio.

What this means for a B2B founder's show

Here is where I disagree with myself a little, and where this matters for anyone running a show to grow a business.

If your show exists to book meetings, view duration on YouTube is not your main scoreboard. The most valuable viewer of a B2B episode is the guest, and the guest watches all of it. A show with 200 views can still be the best sales channel you have, because every episode is a long conversation with a buyer you wanted to meet.

But public reach still compounds. Every guest you want next will look at your channel before they say yes. A channel with focused titles, clear thumbnails and people who stay to watch tells them their episode will be seen. A 38-second average tells them it will not.

So the lesson carries over. Pick one audience per show. Title every episode around the one thing the viewer learns, not around who the guest is. Put the information in the thumbnail. Fix the audio before you buy anything else.

What I would change first

The order I would give any founder in my position:

• Audit the last 20 titles. Rewrite any that do not promise one specific thing.
• Redo thumbnails around the information, not the guest.
• Decide what the main channel is about, and move everything else to a second channel or stop making it.
• Check view duration weekly, by video, and study the ones that held people longest.

Numbers like mine are not a verdict on whether your content is good. They are a verdict on whether a stranger can tell, in two seconds, why they should stay.

If you want a show built around one audience from day one, see the shows we run.

Frequently asked questions