Your channel is leaking clients. Not your videos.

The leak is next to the video

We often get asked which video someone should make to finally get clients out of YouTube. The honest answer is usually: none, for now. Because when we open YouTube Studio, the problem is not what was made but what never happens afterwards.

A channel we audited this month is a good example. A year of hard work: 338 videos published, 466,557 views. And in that same year exactly 46 clicks on end screens and 56 on cards. That is everything YouTube itself sent towards a next step for that channel.

The descriptions under the videos had nine exits: community, intake, subscribe, e-book, academy, homepage, three socials and an email address. The link that mattered most sat on line three. And the e-book link in the three biggest videos had returned a 404 for months.

This is not an exception. It is the default. The video gets all the attention, the rest was set up once in the year the channel started and never touched again.

Eight things almost every channel leaves undone

1. One link in the description, at the top. Not nine. The first line is what a viewer sees above the fold. That is where your only next step goes, with a reason to click.
2. End screens and cards to the same destination. On every video that still gets views. The last twenty seconds and a card around three quarters in. Not to 'subscribe', but to the video or page your clients come from.
3. A pinned comment. The same link, one substantive line with it. The top comment gets read more often than the description.
4. Playlists by topic. Not by episode number or year. Someone who watches one video about pricing should see the other three about pricing right next to it.
5. Chapters. From the real transcript, with the search terms in them. They make long videos searchable and keep people longer.
6. Titles without episode numbers. 'Episode 247' is unfindable and tells a new viewer they have missed 246 things. The title says what the video solves, in plain language.
7. Channel description and links. The channel page is the first place someone looks after they liked a video. It says who you help and where to go, not your life story.
8. Hygiene. Dead links, scam comments, a word filter that is off, subtitles and language setting. Not exciting, but the first things a viewer bounces on.

None of these eight needs a recording. They take an afternoon, and they work on every video you already have and every video still to come.

Why this comes before new videos

A new video inherits every leak on the channel. You can make the best video of the year, and underneath it sit the same nine links, no end screen, in a playlist with a number on it. The viewer you just convinced disappears in exactly the same place as last year's viewers.

It works the other way too. A channel with a hundred videos has a hundred pages that already get traffic. Putting one link, an end screen and a pinned comment under them is the cheapest traffic you will ever send to your calendar. It is already there. It just did not know where to go.

That is why we always do this first, before any new video gets planned. It also tells you whether new videos make sense at all: if the twenty most-watched videos still produce no enquiries after a month, the problem is not the content but the offer. You want to know that before you record twelve new videos.

How to do it yourself, in this order

1. Pick one destination. The page a viewer should go to when they want more. A call, a check, a sign-up. One.
2. Sort your videos by views over the last twelve months. Start with the top ten. That is where the traffic is, so every improvement counts most there.
3. Per video: first line of the description, end screen, card, pinned comment, chapters, playlist. A quarter of an hour per video once you have the rhythm.
4. Then the channel: description, links, featured video, playlists by topic.
5. Finally the hygiene: dead links, scam comments, word filter, language.
6. Measure one number: clicks on your destination per week. Not views, not subscribers. If that number has not moved after four weeks, you know where your next problem is.

Leave running title tests alone. Replacing a test before the result is in throws away a week of measurement.

What to expect

No explosion of views. This is not a growth hack, it is maintenance that was never done. What you do see: a click number that goes from zero to something, viewers watching a second video because it sat next to the first, and a channel page that finally says what you do.

For the channel in the example we worked out that 102 clicks a year, with the same views, could be ten times that. Not by posting more, but by putting the exit on every video that was already there and never had one.

And only then: the next video. For the first time it inherits a channel that sends people somewhere.