Most channels I look at have between four and eleven videos doing almost all the work. They were published a year or two ago, they still get impressions every single day, and nobody has touched them since upload. That is the entire opportunity. The topic is proven, the footage exists, the watch time is banked. The only thing standing between that video and another 50,000 views is packaging that was made in twenty minutes on a deadline.
Repackaging evergreen videos is the least glamorous growth lever on YouTube and, in my experience running packaging for channels across finance, fitness, DIY and software, the most reliable one. You are not gambling on a new idea. You are fixing the storefront on a shop that already has customers walking past.
What counts as an evergreen video?
An evergreen video is one where the search demand and the recommendation demand do not decay with the calendar. Nobody stops needing to know how to fix a running toilet. Nobody stops wanting a plain explanation of index funds. The test I use is simple: pull the last 90 days of traffic and check whether videos older than nine months are still bringing impressions. If they are, they are evergreen, whatever you thought you were making at the time.
The opposite is topical content. A reaction, a news breakdown, a product launch review. Those had a window and the window closed. Repackaging them is usually wasted money, and I will tell a client that before I take the job.
The useful middle case is a video that is evergreen in subject but dated in packaging. A 2023 tutorial with a 2023 style thumbnail reads as stale even when the content is perfectly current. That video is not losing to competition. It is losing to a visual timestamp nobody meant to put there.
Why does old footage outperform new uploads so often?
Because the algorithm already knows who to show it to. A new upload starts cold. It gets a small test audience, and everything depends on how that test performs in the first few hours. An evergreen video that has been running for eighteen months has a settled audience profile, a proven average view duration, and a standing place in suggested feeds and search results.
When you change the packaging on that video, you are changing one variable in a system that is otherwise fully warmed up. That is why the response is often fast and often large. I have watched a redesigned thumbnail move a two year old tutorial from a flat baseline to double its daily views inside a fortnight, with zero change to the video itself.
This pattern is common enough that I wrote a whole piece on what to do when old videos outperform new ones, because most creators read it as a problem instead of a signal.
If a video from two years ago is still your best performer, that is not a failure of your recent work. It is the algorithm telling you exactly which shelf to restock.
Which videos should you repackage first?
Start with impressions, not views. Sort your library by impressions over the last 28 days and look for the gap between high impressions and low click-through rate. That gap is the money. YouTube is already offering the video to people and they are declining. That is a packaging problem, not a content problem, and it is the cheapest thing in your channel to fix.
My priority order looks like this:
- High impressions, weak CTR. Being shown, not chosen. Fix first.
- Strong CTR, strong retention, low impressions. The video works when people see it, so a sharper title can widen the topic it gets matched to.
- Good views, dated visual style. Still earning, but leaking. Refresh the look and the numbers usually firm up.
- Everything else. Leave it alone.
Ignore the temptation to redo your favourite video. Favourites are almost never the right pick. The right pick is usually something you find slightly boring, which is exactly why you underinvested in its packaging the first time.
If your library is mostly how-to content, the sequencing gets a bit more specific, and I go through it properly in this guide to repackaging tutorials and evergreen guides.
What actually changes in a repackage?
Three things, together. This is why the modern name for the work is YouTube packaging rather than thumbnail design. The thumbnail, the title and the first frame promise are one unit, and changing one without the others usually cancels out the gain.
The thumbnail. The job is not to look good. The job is to be legible at 210 pixels wide on a phone in daylight. That means one subject, one idea, high separation between foreground and background, and text you can read without leaning in. Most old thumbnails fail on subject count. They try to show the whole video. A repackage almost always removes elements rather than adding them.
The title. Old titles tend to be descriptive. New titles need to be specific. "Budgeting Tips" becomes "The budgeting method that finally stuck after 4 failed attempts". Same video, but now the title carries a promise and a small piece of tension. Title work also changes which search queries the video gets matched against, which is the quiet half of the gain.
The first frame promise. Whatever the thumbnail implies, the first ten seconds must confirm. If the thumbnail says the video is about one specific mistake, the opening cannot spend forty seconds on channel housekeeping. You do not need to re-edit the whole video. Trimming a cold open or re-recording twelve seconds of intro is often enough to protect the retention curve you already have.
The order matters too. I write the title first, then design to it. Designing a thumbnail and then bolting a title on is how you end up with two competing promises in one search result.
Does repackaging reset a video's performance?
No. Changing a thumbnail or title does not delete a video's history or restart its ranking from zero. YouTube keeps the watch time, the engagement history and the existing search placement. What changes is the click decision for every impression served from that moment forward, which is why results appear as a slope rather than a spike.
A few practical points I give every client. Change one video at a time if you want clean reads. Give each change at least fourteen days before judging it, because daily CTR is noisy and weekends behave differently from weekdays. And measure CTR alongside average view duration, never alone, because a thumbnail that overpromises will lift clicks and sink retention, which YouTube will notice before you do. If you want the platform's own framing on impressions and click-through rate, the YouTube creator resources cover the reporting side clearly.
One more honest caveat. If a video has genuinely weak retention, packaging will not save it. Better packaging on a video people leave after ninety seconds just tells YouTube to stop recommending it faster. Repackaging amplifies what is already there. It does not manufacture quality.
How far back should you go?
Further than feels sensible. I have had strong results on videos four and five years old, as long as the content is still accurate and the footage is not visibly ancient. The cutoff is credibility, not the upload date. A five year old software tutorial showing a dead interface is finished. A five year old explanation of compound interest is fine forever.
I have written up the deeper version of this, including how to handle dated footage and outdated on screen references, in repackaging videos from years ago. And if you make practical, hands-on content, the visual grammar shifts a fair bit.
The other thing worth doing on old winners is checking whether they have a Shorts life. A tutorial with a strong single moment often converts into vertical, and that needs its own cover treatment rather than a cropped 16:9 frame. I go into the format specifics on the page about Shorts cover design.
Should you repackage in batches or one at a time?
Batch the design work, stagger the publishing. Designing eight thumbnails in one run is far more efficient than eight separate briefs, because the designer builds one mental model of your channel and applies it consistently. But pushing all eight live on the same day makes the results unreadable. Roll them out across two or three weeks and you can actually tell which changes worked.
That batch rhythm is exactly why our packaging work is priced the way it is. $250 for 8 custom thumbnails, which is about $31 each, one per video, with unlimited revisions, a dedicated designer and title consultation included. Each thumbnail takes 48 to 72 hours. If you want to test two directions on your biggest video, an A/B variant is $15. A full title pack is $25. Rush is $20.
Everything is custom per video. We do not run templates, and the styles we use get tested on our own channels before they go anywhere near a client library. You can see the range of that work in our portfolio.
Is it worth paying someone, or should you do it yourself?
It depends entirely on whether you can be objective about your own back catalogue, and most creators cannot. You know what the video contains, so you cannot see what the thumbnail communicates to someone who does not. That blind spot is the single biggest reason self-made packaging underperforms, and it gets worse the longer you have owned the footage.
The honest version of the tradeoff is this. If you have design skills, a defined channel style and the discipline to test properly, do it yourself and spend the money on better audio. If you have a library of proven evergreen videos and no time, hiring out the packaging is one of the few YouTube expenses with a traceable return, because you can compare before and after on the exact same video. I laid out both sides in more detail in this piece on agency versus doing it yourself.
We work from Dubai, fully async, with creators in most timezones. If you have eight evergreen videos sitting on flat CTR, send me the list at samy@pixelsamy.com and I will tell you which four are worth doing first.