1 series a year, and 11 months of everything else
A few weeks on air, nightly, with a live broadcast producing whatever everyone will be talking about at 9 the next morning. Then silence until the same weeks come round again.
- ~5,250videos
- 1.45Msubscribers
- ~832Mlifetime views
- ~158,000views per video, lifetime
Public counts for the channel, as of Sep 2026.
-
A season
Love Island's publishing is spread across a long summer run and extended by winter series. There is always a next week to come back for.
-
A spike
Here, all the urgency lands in a few weeks. The other 11 months have to be made rather than scheduled.
Put the catalogues side by side and they look like the same channel: around 5,250 videos here against Love Island's 5,500 (as of Sep 2026), both reality, both attached to a nightly transmission. They are not the same channel at all.
Work on either shape alone and you will conclude that publishing is the job. Running both is what taught me it is not.
The 3 weeks are the part everyone can see. The other 11 months decide what the 3 weeks are worth.
The decisions that carried the 11 quiet months
-
The problem
It is 3 weeks of nightly transmission, then 11 months with no show to cut from. A channel that waits for the next series goes quiet for most of the year.
-
The decision
Keep publishing anyway. Compilations out of the series that had just finished, original pieces filmed with the celebrities, behind the scenes and unseen footage that never made transmission. It is the one stretch of the year with no deadline attached to it, which is exactly why it is the part that gets skipped.
-
What it was for
Coming back each year to an audience rather than to a standing start, and a channel that kept publishing through the months when nothing was on air.
The off-season problem
For most of the year there is no show to cut from, and the channel still has to publish.
You cannot manufacture a broadcast moment out of season. The obvious answer - go quiet until the show comes back - costs the most, because a channel that disappears re-earns its audience from a standing start every year.
So the off-season is its own production job. Some of it is catalogue work: finding the things in previous series people are still quietly searching for months later, and making them findable. The rest is new video that had to be made for the gap - compilations, original shoots with the celebrities, behind the scenes and footage that never went out.
Cut to the moment, not the segment. Titled the way a person would ask for it out loud. Sitting in a playlist that makes sense to somebody who has landed on one video and never seen the channel.
None of that is glamorous, and it is the part of the year that decides how the next series opens.
-
The I'm A Celebrity… Get Me Out Of Here! channel went from around 136,000 subscribers to over 244,000 while I was running it.
A returning national format with a broadcaster promoting it does most of the work in a number like that. What I did was the editing, the titles and packaging, the publishing and the catalogue. I ran the channel in partnership with ITV, and both figures are numbers YouTube displayed on the channel's own page, archived by a third party neither I nor anyone else involved controls.
- 00:00Assume no context. This is met by a stranger months after transmission.
- 00:05Name the person and the situation inside the first sentence of audio.
- 00:12The thing they searched for. Early, because they arrived looking for exactly it.
- 00:19One reason to stay past it, or they leave having got what they came for.
- 00:25Handover. Point at the next clip while the interest is still warm.
A schematic of how a clip on this channel opens, not a reading off any particular video.
Why the off-season decides the series
Views per video is the whole comparison: roughly 158,000 here. Do the same arithmetic on the Love Island channel and you get roughly 355,000 - 2 channels within a few per cent of the same amount of published video, one earning more than twice as much per upload.
Neither number is mine to claim: both are lifetime channel averages. But the gap is real and the reason is structural. Love Island's audience had a reason to come back next week because there was a next week. Here, for most of the year, there is not.
The priorities invert. On a channel publishing nightly all year, speed is everything and metadata is a chore you do at the end. Here, metadata is the product.
On a seasonal channel the back catalogue is not an archive. It is raw material for the 11 months when there is no show to cut from.
The channel
I'm A Celebrity… Get Me Out Of Here! on YouTube - the official ITV channel, and one of the ones I ran. It is still publishing; one of the channel's own recent uploads is Sir Mo shares his story of being trafficked | I'm A Celebrity… South Africa 2026.
Linked as context, not as a portfolio. That upload is the channel's, not mine - I am not claiming any individual video on this channel as my cut.
What transferred to creator work
The off-season discipline, almost entirely.
A creator's channel is a back catalogue with one new video on top of it. Most people treat the new video as the whole business because it is the part that feels like work. On any channel older than a year, the majority of what gets watched this month was published before this month.
The practical version: assume every video will mostly be found by somebody who has never seen your channel, arriving months late, with no context and no loyalty. That changes the first 15 seconds of an edit more than any hook framework does.
I came out of it suspicious of consistency advice. A channel that only works while the habit is intact is a fragile channel, and the seasonal ones prove it: they go dark for months and come back to an audience, or they do not, and the difference was decided by what was in the catalogue the whole time.
I cut long-form now as Senior Video Editor @ Simon Squibb, and I still open the same way: for the stranger, not the subscriber. The Love Island channel page is the other half of this comparison, and the about page has the 10-year version.