Every few weeks someone asks me the same question, usually in the second or third email. "Can we do 20 thumbnails instead of 8?" And every time I say no, and every time there is a pause on the other end where I can feel them deciding whether I am lazy, badly run, or hiding something.
None of the three. The cap is the product. ClickTheory sells one pack, $250 for 8 custom thumbnails, one per video, unlimited revisions, a dedicated designer, and a title consultation baked in. That works out to about $31 a thumbnail, which is not the cheapest number on the internet and is not meant to be. What people are really asking when they ask for 20 is whether the number is a shelf we can just restock. It is not. It is a cap, and I put it there on purpose, and I have watched what happens on both sides of it.
The number came from watching designers, not spreadsheets
I did not arrive at 8 by modelling margins. I arrived at it by watching how long a designer stays sharp on one channel's visual problem before the work goes flat.
A thumbnail takes us 48 to 72 hours. Not because it takes 48 hours of labour, obviously, but because a good one needs a first pass, a night away from it, and a second look with fresh eyes. Somewhere around the sixth or seventh thumbnail for the same channel, the designer stops solving and starts recalling. They reach for the crop that worked in number three. They reuse the eyebrow. The face goes to the same third of the frame because that is where it went last time and last time was fine.
Fine is the whole problem. Fine does not move click-through rate.
By eight, a designer has learned a channel's face, its lighting, its audience, and its bad habits, and they are right at the top of their usefulness. Push to twelve and you get four thumbnails that look like echoes of the first eight. Push to twenty and you have quietly built a template, which is the one thing I promised we would never do. Every thumbnail here is custom to the video. The moment volume forces a house style onto a channel, we are selling the thing we tell people to avoid.
Eight is a testing window, not a quota
Here is the part that I think gets missed. Eight thumbnails is not eight deliverables. It is one experiment with eight data points.
We plan a pack as a set. Two or three of the eight are usually deliberate swings: a different expression, a much tighter crop, text where we normally use none. The rest are variations on whatever has been working. Then we wait and read the retention and click-through numbers together, because a thumbnail that lifts CTR while tanking the first 30 seconds is not a win, it is a bill coming due later.
Eight is roughly the smallest set where you can tell a pattern from a fluke. Under five and you are reading noise. Over twelve and the channel itself has changed underneath you, because the creator got better at hooks or the algorithm shifted the audience mix, and now you cannot tell what the thumbnails did.
A cap is not a limit on what we will make. It is a limit on how much we will guess before we look at the numbers.
On one finance channel around 80k subs, we ran a pack where three thumbnails dropped the face entirely in favour of a single object on a flat colour. Two underperformed the channel's own average. One beat everything they had run that quarter. That single result reshaped the next pack. If we had shipped twenty in one go, that lesson would have arrived after nineteen thumbnails had already gone out carrying the old assumption. I wrote more about that specific set in the thumbnail test that surprised me, and the short version is that the surprise only existed because the batch was small enough to notice it.
What the cap does to a client relationship
Uncapped volume changes how people treat the work, and not for the better.
When a client has a bottomless queue, thumbnails become a chore they hand off. Briefs get shorter. "Same as last one but different colour." Feedback arrives as a shrug. Nobody looks at the numbers because there is always another one coming. The designer becomes a vending machine and the client becomes a person pressing buttons.
Eight forces a conversation. Which eight videos? Why those? What is the title doing, since the title and the thumbnail are one unit and always have been. That is why the title consultation is in the pack instead of being sold separately as an upsell. If I hand you a beautiful image sitting under a title that fights it, I have taken your money and given you nothing.
That conversation is also where the pack earns the unlimited revisions. Unlimited revisions on eight thumbnails is a real promise. Unlimited revisions on eighty is a lie with a nice font, because at that volume the studio has to start rationing attention quietly, and the client finds out by feel rather than by being told. If you want the fuller picture of how the pricing logic hangs together, I laid it out in paying per thumbnail versus per pack.
The hiring problem nobody mentions
This is the business half of the argument, and it is the half I get pushback on.
A dedicated designer means the same person owns your channel. Not a pool. Not a ticket queue. That structure is the reason the work holds up, and it is also the reason the cap exists, because a designer can hold maybe a handful of channels in their head at once with real fidelity. Give them fifteen and they are context switching all day, and context switching is where quality goes to die in a design studio.
So the honest constraint is this: I can only take on as many packs as I have designers who are actually good, and good thumbnail designers are not a commodity you can buy in bulk on a Tuesday. Hiring one takes weeks. Training one into a channel's voice takes a pack or two. If I sold 20-thumbnail packs, I would need either more designers than I can vet properly or fewer clients than I want to serve, and the first option is how boutique studios turn into content mills.
Growing teams hit exactly this wall internally too. If you are deciding whether to build the function in house or buy it, I went through the real tradeoffs in in-house versus outsourced design, and the answer genuinely depends on your upload cadence more than your budget. Either way, the cap logic applies the same to an internal designer as it does to us.
What we do when 8 is genuinely not enough
I am not pretending the cap fits everyone. Some channels ship four videos a week. For them, eight is two weeks of coverage.
The answer is not a bigger pack. It is more packs, run in sequence, with a read of the numbers between them. That sounds like a sales answer and I understand why, so let me be specific about why sequence beats size: the second pack is always better than the first, because the second pack knows something. It knows which crop held. It knows whether that channel's audience responds to faces or to objects. Buy sixteen at once and both halves are equally ignorant.
The add-ons exist for the edges rather than for volume. An A/B variant is $15 when you want to run two versions of the same idea against each other properly. A title pack is $25 when the titles need more attention than a single consultation gives. Rush is $20 when a video is going out tomorrow and waiting 48 hours is not on the table. None of those are ways to get more thumbnails. They are ways to get more certainty out of the eight you have.
We also test styles on our own channels before they touch a client's. That is the other reason the cap holds: the experimental risk sits with us, not with your upload. Plenty of ideas that looked clever in a mockup died on our channels first, which spared everyone. A few of those are catalogued in thumbnail trends that aged badly, and the pattern is nearly always the same, a style spreads because it worked once for someone with an audience that would have clicked anything.
The offer this makes, and the offer it refuses
A cap is a promise about what you will refuse. Ours refuses volume as a selling point.
If what you need is a hundred images a month at the lowest possible unit cost, we are the wrong studio and I would rather say so in the first email than in the third pack. There are places that do that well and cheaply and I have no argument with them. If what you need is eight thumbnails that are actually about your video, made by someone who knows your channel, with the title thought through and the numbers read afterwards, that is the whole business. You can see the range of what that looks like in our work, and it is the same standard whether the channel is local to us or not, which is why the pages for a YouTube thumbnail designer in Bangalore and one working with creators in Sydney describe an identical process. We are Dubai based and work async worldwide, so the time zone matters less than the cadence.
One more thing, because it comes up. Creators who cap their thumbnail spend often have a bigger leak on the video side, especially with long-form podcast content where the edit is eating the week. That is a different service and a different team, and if the editing is the real bottleneck, the podcast editing side of things starts at $199 with clips at $25.
What I would tell you if you were about to ask for 20
Ask for eight. Pick the eight videos that matter most in the next month, not the eight that happen to be next in the calendar. Give us the real brief, argue with us about the titles, and then actually look at what happened before ordering again.
If after that you still want twenty at once, tell me and I will tell you honestly whether I think it will help you, because most of the time it will not. The number is small on purpose. The cap is not us protecting our capacity. It is us protecting the part of the work that makes it worth $31 instead of $5.
Email me at samy@pixelsamy.com if you want to talk through whether a pack fits your cadence. I would rather have that conversation than sell you a bigger one.