No clear look.
The thumbnails looked like they came from six different channels. Nothing said "this is a series," so viewers built no habit and the algorithm found no pattern.
UNIBRIGHT SOLUTIONS · AI SOCIAL MEDIA · CASE STUDY
72% More Watch Time, 103% More Reach
Peri Toons TV had a full library of videos and almost no watch time. Reach was flat. Each new upload picked up a few views, then stopped.
+72%WATCH TIME
+103%CHANNEL REACH
5PLATFORMS
AutoNOW RUNS ITSELF
01 / THE PROBLEM
The Problem
Peri Toons TV had a full library of videos and almost no watch time. Reach was flat. Each new upload picked up a few views, then stopped.
This kind of problem is hard to spot, because nothing looks broken. The videos were good. The channel kept posting. The numbers just would not move.
YouTube’s rules for kids’ content made it harder. On a Made for Kids channel, comments, end screens, community posts and personalised ads are all switched off. Most of the usual growth tactics were simply not available. Any plan had to work through format, search and viewing habits alone.
REVIEW
What Our Review Found
We checked the whole channel before we suggested anything. We found six problems. Here they are, worst first.
Nothing to watch next.
No playlists, no autoplay. Every view ended where it started. On any channel, that is the biggest single source of lost watch time.
Weak titles and descriptions.
Descriptions were short and generic. Most videos had no hashtags. A large part of the library could not be found by search at all.
No set format.
The videos read as unrelated uploads rather than one channel with a clear idea. That caps repeat viewing, no matter how good each video is.
Titles written for the wrong reader.
They read well in a list. They did not match what parents actually type into search.
No short videos.
With the usual discovery routes closed off, Shorts was the only way left to reach new viewers. It was not being used.
BUILT
What We Built
One face on everything.
The same character on every thumbnail — front-facing and large. Being consistent is what turns a pile of videos into a brand.
Playlists built to keep people watching.
Themed playlists, titled for search, set up so one video rolls into the next. With end screens switched off, this was the only real route to longer viewing.
New titles, descriptions and tags across the library.
We went through the existing videos and rewrote them for search. Videos nobody could find became videos people could find.
One format, applied every time.
A set structure, length and look, so every new video adds to the same idea instead of pulling away from it.
Short videos added.
Shorts cut from footage the channel already had, to open the one route to new viewers that was still available.
Five platforms instead of one.
The same content programme extended across five platforms, so the channel stopped depending on a single algorithm.
IMPACT
Where AI Helped — and Where It Did Not
AI did the review. It read the whole library, the titles, the retention data and the competing channels in a fraction of the time a manual audit would have taken. That speed is real, and it is available to anyone. The tools are not the advantage.
A social media strategist made the decisions. Which problem to fix first. Whether to spend the time on thumbnails or on playlists. What the format should be. Which findings were real causes and which were just symptoms. AI finds patterns. It does not know which pattern matters for a kids’ channel with comments switched off and a library nobody is watching.
We staff every social account this way on purpose: AI for the work that eats time, experienced people for the calls that decide whether the work was worth doing.
05 RESULT
The Results
RESULTDETAIL
Watch time per video up 72%The main target of the plan — people watch each video for longer
Channel reach up 103%Total reach across the channel, more than doubled
Posting on 5 platformsYouTube, Facebook, Instagram, TikTok and X, on an automatic schedule
Monetization reachedYouTube Partner Program, after the strategy work
Runs on its own nowScheduling, posting, tags and reporting all automated
MEANS
What This Means
The old videos were an asset, not dead weight.
Most of the watch time gain came from making existing videos easier to find and easier to keep watching — not from making more of them.
Watch time responds to structure.
Playlists, thumbnails and descriptions are dull work. They are also where the results came from.
The plan worked inside the rules.
All of this happened with comments, end screens and personalised ads switched off the whole time.
We automated last, on purpose.
Automating a plan that has not been proven just produces failure faster. We proved it first, then took out the manual work.
It keeps working without us in it daily.
That is the real deliverable: a system, not a retainer that stops working the moment attention moves elsewhere.
WORKS
Why This Works Elsewhere
Any channel or brand page with a back catalogue and flat numbers has a version of this problem. The instinct is always to make more content. Usually the content is fine and the structure around it is missing nothing to watch next, no clear look, nothing search can find, no short-form.
So we review first and produce second. Not the other way round.
NEXT STEP
The Next Step
Give us access to your analytics and your existing content. You get back a list of what is broken in priority order, a plan for the top three, and the number each fix is meant to move.
Services Provided:
AI social media strategy, content review, channel look and format design, playlists and search structure, platform SEO, short-form strategy, multi-platform posting, posting and reporting automation