The question I get more than any other, in one form or another, is whether a better thumbnail will get someone more views, and the honest answer is a bit longer than yes or no, right, because a thumbnail does not create views directly, it sits at one link of a chain, and whether better packaging brings better results depends on what happens at the links before and after it. I think if more creators and business owners understood that chain, they would stop being disappointed by thumbnails that were never going to fix their problem, and they would get much more out of the thumbnails that could. I run a video editing agency in Dubai, and our group publishes on channels of its own, each of them above 100,000 subscribers, so this chain is not theory for me, it is the thing we argue about every week on our own uploads, long before a client's video is on the table.
How do thumbnails increase views, mechanically?
Thumbnails increase views by turning more of the impressions YouTube already gives you into clicks, and when those clicks turn into satisfied watching, YouTube tends to show the video to more people, which creates more impressions to convert.
That is the whole loop, basically. YouTube shows your thumbnail to someone, which counts as an impression, and its help page explains exactly what counts as one and how click-through rate is worked out from it (impressions and click-through rate). Some of those people click. Of the people who click, some keep watching and some leave quickly. And YouTube's own explanation of its recommendations, on How YouTube Works, says it pays attention to signals like what people click, how long they watch and whether they seem satisfied, so a video that earns clicks and then holds people is the kind it has reason to keep showing. The thumbnail and title are what turn an impression into a click, right, so they are the first lever in the loop, and the loop is why a small improvement there can be worth more than it looks.
Where exactly does packaging sit in the chain?
Packaging sits between the impression and the click, and it also sets up the first seconds of watching, because the promise on the thumbnail is what the viewer checks the opening against.
That second part is the one people miss, right, because they think of the thumbnail as a door, and once the viewer walks through, its job is done, but it is really a promise, and the first few seconds of the video are where the viewer checks whether the promise was true. A thumbnail that shows the finished build, and a video that opens on the finished build before going back to the start, works as one piece, while a thumbnail that shows the finished build on a video that opens with two minutes of chat is making a promise the video keeps far too late. So when a video comes to ClickTheory, the title and the image get worked out as one promise, with plenty of title directions and a few rough sketches on the table before anything is finished, and somebody actually watches the opening, because if the first ten seconds do not back that promise up, we would rather tell you than design around it. The funnel is broken down further in the YouTube impressions funnel.
Why can a higher click-through rate still lose you views?
Because a thumbnail that overpromises can pull in clicks from people who leave quickly, and that weak watching can hurt the video more than the extra clicks helped it.
This is the trap, right, and I have watched it play out more times than I can count, a channel tries a more dramatic thumbnail, the click-through rate goes up, everyone is happy for a day, and then the views flatten, because the people who clicked were expecting something the video does not deliver, so they leave, and what YouTube sees is a video people click on and then abandon. The fix is not to make the thumbnail boring, it is to make the promise specific and true, so that the people who click are the people the video was made for. This is also why I like YouTube's own A/B test so much, right, because when it runs up to three titles and thumbnails against each other on one video, the version it ends up showing everyone is the one that earned the most watch time, not the most clicks, so the test itself punishes the overpromise (YouTube spells out how it works on YouTube's A/B testing help page), and we prepare that second design whenever a video is worth testing. The version that wins on watch time is usually the one that made the most honest promise, and I go into this in more detail in when a high CTR thumbnail is quietly hurting you and click through vs watch time.
A thumbnail is a promise, and the click is just the viewer agreeing to check whether you meant it.
What results can better packaging actually produce?
Better packaging can raise the share of impressions that become clicks, bring in viewers who are a better match for the video, and give old videos with good content a second chance, but none of those are guaranteed numbers.
To be very honest, I will never promise a specific view count or click-through rate, from us or from anyone, because the result depends on the video, the topic, the audience and what else is in the feed that week. What I can say, right, is what packaging is designed to do, so that is one, it turns more of the impressions you already have into clicks, which matters because you are not paying anything extra for those impressions, they are already being shown. Secondly, it can change who clicks, which on a business channel especially is often more valuable than how many. And the third thing is the back catalogue, because a strong video from two years ago with weak packaging can be repackaged and tested again, and that is the cheapest growth most channels have, which is why we do it on old videos worth reviving as well as new uploads. If you want to know what that looks like over a realistic timeframe, I wrote what better results from packaging look like in the first 90 days, and the practical side of improving click-through is on how to increase YouTube CTR.
What can packaging not fix?
Packaging cannot fix a video people do not want to finish, a topic nobody is looking for, or a channel YouTube is not showing to anyone yet, because it can only convert the attention that already exists.
That map is my judgment, not data, but it shows the shape of the thing. Packaging has the most leverage on click-through rate and on who clicks, it has some influence on the first seconds because of the promise it sets up, and it has much less influence on how long people watch in total or whether they subscribe, which depend on the video itself. Video quality is basically off the map, right, no thumbnail can make a video better than it is. And if a channel is very new and getting very few impressions, there is only so much the click rate can do, because a great conversion of almost nothing is still almost nothing, so for those channels I usually say the priority is the videos and the topics, with solid packaging alongside. The wider point is made in CTR is not the only number that matters.
How we tell whether packaging worked
We look at click-through rate on comparable impressions, the first thirty seconds of retention, and average view duration together, and we judge a change over enough time to mean something rather than after one afternoon.
The mistake is to look at one number in isolation, right, because click-through rate on its own can mislead you, since it usually falls as a video gets shown to wider audiences who know you less, so we compare like with like, the same traffic source, a similar stage in the video's life, against the channel's own past videos rather than some industry average. Then we look at whether the people who clicked stayed, because the point was never the click on its own. Every video's result, winners and losers both, goes into the channel's packaging playbook, and that playbook is where results really come from over time, right, because each test teaches you something about your specific audience, and the next video starts from a better guess. Do you see why I care more about the playbook than about any single thumbnail?
Why old videos are where the chain restarts
An old video with good watch time and weak packaging is the easiest place for better packaging to show results, because the second half of the chain is already proven and only the click was missing.
If people who find an old video tend to watch most of it, right, the video has already shown it can hold an audience, so a new title and thumbnail that earns more clicks feeds straight into a loop that works. That is why I nearly always look at a channel's back catalogue before its next upload, and why redesigning old thumbnails is its own service with us, described on redesigning old YouTube thumbnails. The opposite is also true, an old video that people abandon quickly is not a packaging problem, and repackaging it will just send more people to something they leave.
Where ClickTheory fits in the chain
We work on the link we can actually move, right, the click and the promise behind it, and we are honest about the links that belong to the video, the topic and the editing.
In practice that means two people on each video, a strategist thinking about the promise and a designer making it readable, with the title and the image decided as a pair, the final checked at the size a phone will actually show it, and a test version prepared when the video deserves one. For most channels the starting point is a pack of eight thumbnails at 250 dollars, the extras and the turnaround are spelled out on our services page, and you can stop after any pack without a conversation about contracts. If the weak link in your chain is the video itself, or the posting and distribution, that is work for editing and distribution teams, which our sister agency Pixel Samy Studio does, and I would tell you that on the 15 minute call rather than sell you thumbnails you do not need. At the end of the day, better packaging brings better results when it is the link that was broken, and knowing which link is broken is half of the job. So yeah. That's my way of saying it.