A creator emailed me last March asking whether he should hire a full time thumbnail designer at around $2,800 a month or keep buying packs from us. He published six videos a month. I did the math out loud on the call and told him the honest answer, which was that he should not hire anyone yet, and that if he ever got to twenty videos a month he should stop paying us and build a team. He did not love hearing the second half. He hired internally eight months later anyway, and he was right to.
That call is the reason I keep writing about this. The in-house versus outsourced argument gets treated like a values question, like you either believe in owning your craft or you believe in buying it. It is not a values question. It is a volume question with a quality floor attached, and most channels answer it wrong because they compare the wrong two numbers.
The number everyone compares is the wrong one
Almost every creator who asks me about this compares a monthly salary to a monthly invoice. Salary loses or wins, decision made, call over.
That comparison is broken in both directions. A salary is not the cost of a hire. The cost of a hire is salary plus the hours you personally spend feeding that person context, plus the six to ten weeks before their work is actually good, plus the risk that they leave in month five and you restart the whole thing during a launch. Meanwhile an invoice is not the cost of outsourcing. The cost of outsourcing is the invoice plus the friction of async, plus the ceiling on how deeply anyone outside your channel understands your audience.
Our pack is $250 for 8 custom thumbnails, roughly $31 each, one per video, unlimited revisions, dedicated designer, title consultation included, 48 to 72 hours per thumbnail. That is a real number and I am not shy about it. But $31 a thumbnail is only meaningful next to a real internal number, and the real internal number is almost never the salary alone. I wrote a longer breakdown of the market in what thumbnail design should cost, because the spread between what people pay and what they think they pay is wider than the spread between agencies.
The right question is not "which is cheaper". It is "at my publishing volume, which one buys me a designer who actually knows my audience".
Volume is the variable that decides it
Here is my rough line, and I will defend it.
Under about eight videos a month, hiring in-house is a mistake. You cannot fill a designer's week with eight thumbnails. You will either pay a full time person to do part time work, or you will start inventing jobs for them, which means channel art, community posts, and a slow drift into a generalist role where thumbnails stop being anyone's specialty. That drift is the actual failure mode. Not cost. Drift.
Between eight and about eighteen videos a month, it is genuinely close, and it depends on how much of your own time you are willing to spend managing. This is the band where most people burn a year making the wrong choice slowly.
Above roughly twenty videos a month, outsourcing starts to fight you. Turnaround windows stack, revision threads multiply, and you begin wanting someone in the room at 11pm when the edit changes and the hook moves. At that point hire. I mean that. We are not the right answer for a daily uploader, and I would rather say so than take the money and get slow.
I have written before about why we cap at 8 thumbnails per pack, and the reason connects directly to this. The cap is not a pricing trick. It is a promise about attention. Past a certain volume per client, per designer, quality goes soft in ways clients do not notice for two months and then notice all at once.
What in-house actually buys you
I want to be fair to the in-house case, because I think agencies argue against it dishonestly.
An in-house designer sits in your standups. They watch your retention graphs. They hear you say "that one felt off" and they know what you mean by off, because they have heard you say it forty times. That accumulated taste is real and it compounds. After a year, a good internal designer can produce a thumbnail that is more on-brand than anything an outside studio will make, because brand is mostly a thousand small decisions nobody wrote down.
In-house also gives you speed at the tail end. Video drops in four hours, the title changed, the face needs to be bigger. Internally that is a ten minute fix. Externally it is a message, a queue, and a wait.
And in-house means the work belongs to you in a practical sense. Files, layers, fonts, the accumulated library. That matters more than people expect, particularly for channels that produce recurring formats where consistency across a season is the whole point.
What in-house quietly costs you
Now the other side, from watching this go wrong repeatedly.
You get one taste. One designer means one aesthetic instinct, one set of blind spots, one reference library. Our studio tests styles on our own channels first before we bring them to anyone, and that only works because we are running across many niches at once. A solo internal designer has no test bed. They have your channel, and your channel is a bad place to experiment because every experiment costs you a real video.
You get a hiring problem you are not equipped for. Most creators cannot evaluate thumbnail designers. Portfolios show finished pieces, not judgment. The skill you actually need is not rendering. It is knowing which of three ideas to kill. That does not photograph.
You get management overhead you did not budget. Feedback, scheduling, cover, performance conversations. If you are the creator and also the editor-in-chief and also now a design manager, thumbnails become one more thing you are half-doing.
You get a single point of failure. Vacation, illness, a better offer. Channels that run entirely on one internal designer discover this at the worst possible moment.
The hybrid most people should actually run
The honest answer for a lot of mid-size channels is neither pure option.
Run outsourced for your core upload schedule, and keep one internal person on everything that is not thumbnails. Or run internal for your bread and butter format and outsource the formats you are still figuring out, which is where an outside eye is worth the most and where your internal person is weakest because they are as stuck in your habits as you are.
I have watched a niche education channel around 120k subs do the second version well. They kept in-house for their recurring series where consistency mattered, and sent us the experimental long form stuff. Their internal designer was relieved rather than threatened, because the experimental work was the part he dreaded. That is usually how it goes when it is framed properly.
If you are weighing a monthly commitment against buying packs when you need them, I laid out the tradeoffs in agency retainer versus per-pack thumbnails, including the case where a retainer is genuinely worse for you.
Quality is not the differentiator people think it is
Here is the part that annoys agencies.
At the top end, a strong in-house designer and a strong outside studio produce roughly comparable work. The gap is not craft. It is consistency of judgment over time versus breadth of pattern exposure. Internal wins on the first, external wins on the second.
Where the gap actually shows up is in specific formats. Close-up heavy work is a good example. We saw on a beauty channel in the 60k range that pushing the face crop tighter moved click-through in a way that surprised us, and we only trusted the result because we had seen the same shape of result elsewhere. I wrote that up in beauty thumbnails and close-ups. An internal designer working one channel would have run that test once, gotten one noisy result, and had no way to know if it generalized. Framing matters here: that was one channel, one niche, one period. Not a law.
Shorts are the other place breadth pays. The rules moved, and they moved fast, and most of what people believed about shorts thumbnails six months earlier stopped being true. Studios notice format shifts earlier for the boring reason that we see more channels.
How to decide this week
Do this instead of agonizing.
- Count your real publishing volume over the last three months. Not your plan. Your actual output.
- Multiply your monthly thumbnail count by what you would pay externally per piece. That is your true outsourced number.
- Add up the total cost of an internal hire including your own management time at whatever you value an hour.
- Ask which option gives you a designer who has seen more thumbnails fail than you have.
- Then ask which option you can reverse in thirty days if you are wrong.
That last one decides more cases than the first four combined. Outsourcing is reversible. Hiring a person is not, at least not cleanly, and treating a human being as a reversible experiment is a bad thing to do.
What I would tell you if you called me
I turn down work when the fit is wrong, and volume is one of the clearest wrong-fit signals I have. I keep a running list of what we say no to partly so people can self-select out before a call. If you publish daily, hire. If you publish six times a month and are considering a $3,000 a month salary line for thumbnails, please do not.
Some formats sit oddly in this framing. Business interview shows, for instance, generate a lot of episodes and a lot of derived clips, and the thumbnail load looks like high volume when it is really one recurring visual system repeated. That is a case where outsourcing scales fine well past my usual line, which is why I keep thumbnail design for business podcasts as its own thing. If your bottleneck is actually the editing side of that pipeline rather than the art, our sister studio edits video podcasts from $199 with clips at $25, and I would rather point you there than pretend a thumbnail fixes an editing problem.
If you want to see the range before you decide anything, our work is the fastest way to judge whether an outside studio would clear the bar you already have internally. If it does not clear that bar, keep your designer. That is a real answer and I am fine giving it.
The whole debate is smaller than it feels. Count your videos, count your hours, and pick the option you can undo. Everything else is preference dressed up as strategy.