There is a version of this conversation that happens in every small business that runs Meta ads, and it always happens on a Tuesday.
Somebody has read that the account needs more creative. Not better creative, more of it, and more different. So the question goes around the room: who is making these? And the room goes quiet, because there is no answer. There is a budget for the ads. There is nobody whose job it is to make them.
That silence is the most expensive thing in the account, and almost nothing written about Meta ads addresses it.
Go and look at what the people who actually run large Meta accounts are saying right now. They agree with each other to an unusual degree. Meta's delivery system rewards a spread of genuinely different ads over one perfect one, so you need volume and you need variety. Every serious practitioner arrives at that same instruction.
Then they stop. Not one of them tells you who makes it, how many you actually need, or what it costs.
There is a reason for the gap. The people giving the advice are running eight-figure direct-to-consumer brands with five outside creative agencies on retainer and four creative strategists in-house. When they say "more creative," they are describing an assembly line that already exists. You are hearing an instruction to build one.
So here is the supply side of that instruction, worked out for a business spending hundreds or low thousands a month rather than hundreds of thousands. It is a shorter answer than you would think, and it starts by throwing out the number.
The Number You Can Test Is Also a Production Order
We have already made the case that your budget, not your ambition, decides how many ads you can test at once: a handful of genuinely different ideas, each funded enough to give you an honest answer, beats thirty starved ones. Take that as settled.
What nobody does with that number is the interesting part. It is not only a testing limit. It is a production order, and it is the only one you are ever going to get.
Zach Stuck co-founded Mars Men and Hollow, and through his agency Homestead has audited five or six hundred direct-to-consumer brands over about nine years. Asked on a recent Motion panel how he sets creative output targets, he described the only sane method anyone has put on record, because he does not prescribe a number. He derives one.
He starts with the revenue target for the month. That gives him the spend. He splits the spend between testing and scaling. He knows the minimum that has to sit behind a single creative test before the result means anything. Divide one into the other, and the number of new concepts he can launch this month falls out the other end.
"It always starts with the financial model," he said, "and works its way back all the way down to how many new creatives can I launch this month."
Read that again as an instruction to yourself rather than a description of a large brand. At the end of that arithmetic there is a small integer. For most businesses reading this it is two or three. That integer is not a ceiling you are trying to reach. It is a purchase order you have to fill, and if nobody fills it, the month simply does not happen.
Which is why the number matters more to the person making the ads than to the person running them. Two concepts a month is a manageable brief for a busy owner. Thirty is a fantasy that produces nothing, and the reason so many accounts stall is that the owner heard thirty, understood it was impossible, and made none.
The man at the top of the scale agrees, incidentally. Asked whether he would rather launch a thousand ads or fewer better ones, Stuck was direct: "I would rather launch better ads with better concepts in a way that I can validate, versus saying, hey, I need a thousand new ads." He treats the ceiling as a feature. "It gives yourself a constraint of saying, okay, I can only spend so much into this to start. I better make sure that these are good ads."
One distinction makes the order fillable. A concept is not an ad. A concept is one reason somebody should buy from you, and it can hold several ads inside it. If your concept is that customers come to you because the last company left the job half finished, that is one idea, and it can be a photo, a thirty-second phone video, and a text-led static. Those three are one test, not three. You give the idea more than one body so a weak execution does not bury a strong thought.
So the real production order for a small account reads something like: two ideas, three executions each, six files. That is a specific enough request to hand to a person. "More creative" is not.
Three Places Creative Comes From, and What Each One Costs You
Chase Chappell, who says he has run ads behind thousands of brands and more than $600 million in spend, describes the problem in the clearest language anyone has used for it: "you're not supplying your ad account with what it needs." His model splits creative into three sources, and while his version of it involves shipping product to thousands of creators, the underlying split holds all the way down to a two-person business.
You and your staff. Roughly half an hour to make a decent static image, once you have decided what it should say. It costs nothing except the half hour, which is the catch, because the half hour comes out of a day that is already full. This is the most trustworthy source you have and the one that silently stops the moment you get busy.
Real people. Chappell's version is an army of creators. Yours is your actual customers, and the person who does the work. A customer saying the true thing about you, filmed on a phone, is the most persuasive asset in this entire category, and it is the slowest to get. He measures it in weeks, and he is being efficient about it. You should expect one to four weeks between asking and having something usable, because the person you are asking has a life.
AI. Minutes, and near enough free. It is genuinely useful, and the boundary is sharp: use it to produce, not to invent. It will lay out a static, resize an asset for four placements, and write you six versions of a line. What it will not do is know why your customers actually call you, and creative built on a guess about that is fluent, plausible and forgettable.
Nobody is telling you to run all three. The point is to know which one you are relying on, because the businesses that run out of creative are almost always the ones relying entirely on the first, in a month that got busy.
Creative Is a Budget Line. Start Treating It Like One.
Here is the number that reframes this, and it comes from the same panel.
Stuck keeps outside agencies on retainer to make his ads. His test for whether one stays: "if we can get an agency's retainer, we can spend 10x that. So if their retainer is 10k and I can spend 100k behind their ads, green light, keep them around."
Turn that around and it becomes a planning number for anybody. The cost of making the ads runs to something like a tenth of what you put behind them. At the scale most businesses reading this operate at, that is a modest figure. On $1,500 a month of ad spend it is a couple of hundred dollars, and a couple of hundred dollars buys a real amount: a customer filmed properly, a batch of statics from a freelance designer, a morning of somebody's time who is not you.
The failure is not that the money is unavailable. The failure is that it never appears on the plan. Ad spend gets budgeted because Meta sends an invoice for it. Creative does not get budgeted, because the assumption is that somebody will just make something, and "somebody will just make something" is not a plan, it is a hope with a media budget attached.
If you take one operational thing from this article: put a line for creative next to the line for spend, at roughly a tenth of it, and hold it there.
The Brief Is the Job
Ask Stuck what the first hire should be for a brand serious about paid social, and the answer is not a media buyer. "I would lean more first hire creative strategist," he said, "copywriter first and foremost." Running the ads is the smaller skill. "Writing really really good briefs is probably the most crucial part of the work to actually getting good ads that can convert."
That should land oddly, because most owners think of the brief as the small administrative bit before the real work. It is the other way round. Whoever writes the brief decides what the ad says. Everything downstream is execution.
And most briefs fail the same way. Sarah Levinger, a creative strategist who works on the psychology side of this, names it exactly: the brief that says "we need a cold audience ad" has told the person making it nothing. It describes where the ad sits in your account. It says nothing about the human being who will see it.
Her replacement is to brief the moment instead. Her example: "we need creative that surfaces a trigger moment for someone who hasn't even identified they have a problem yet." Same budget, same maker, same afternoon. "That brief produces completely different creative, different hooks, different messages, different emotions."
A usable small-business brief fits on one page and answers five things. Who exactly is this for, described as a person rather than a demographic. What has just happened in their life that makes them open to hearing from you. What one thing this ad is claiming. What proof it shows for that claim. And what you want them to do next. If you cannot answer the second one, you do not have an ad yet. You have a product photo.
The Research Is Already in Your Building
The agencies pay real money to learn things you already know.
On the D2C Diaries podcast, the host put the ratio like this: "for every one minute you spend in production, you should spend five in research." Time in competitor ad libraries gets capped in the same conversation at "10% of your time rather than 50," on the grounds that copying somebody's execution without knowing why it worked just moves their guess into your account.
What replaces it is talking to people. His guest, a senior creative strategist a year into the role, had built his research around customer calls, and his sharpest instinct was to deliberately call the ones who left rather than only the happy ones, because the happy ones will tell you pleasant things and the churned ones will tell you true ones. Sitting down with the founder for an hour, he reckoned, "could be some of the best research you could do."
You can see where this is going. That is a large agency spending days to obtain, secondhand, the thing you get for free every day of your working life. You already know what people ask before they book. You know the objection that comes up on every third call. You know the sentence a relieved customer said last month that you have repeated to three people since.
There is a version of this for the visual side too. Stuck spent years running creator videos for Hollow, his alpaca sock brand, all shot the same way, everyone talking to camera from the chest up. Watching organic feeds rather than ad libraries, he noticed people filming full-frame from high angles, tried it, and the ad that finally broke through was one where you could see the socks going on. "If we didn't show the whole body, it would have never happened."
The best format ideas are in the feed you already scroll. The ad library mostly tells you what your competitors were guessing about six months ago.
Keep Ads on the Bench
One last piece of operational hygiene, and it comes from a source I would otherwise take with a pinch of salt.
Jeremy Haynes works at a scale and a volume that has nothing to do with a local business, but he has one habit worth stealing. He always keeps unused ads on the shelf. When something fatigues, or the audience wears out, or the winner stops winning, he is not starting from an empty folder. "You would always want to have ads on the side just chilling, waiting for launches."
He is also honest about where the bottleneck sits with his own clients, which is the most useful thing in the whole interview: "we'll be on their ass about filming and we'll be on their ass about getting us those videos." The agency is not the constraint. The footage is.
Every ad you run has a shelf life, and the ones that work best tend to burn fastest because they get shown the most. If the only creative you own is currently live in the account, then the day it stops working is the day you start filming, and you will lose two or three weeks of performance to a production schedule. Film two things you do not need yet. That is the whole tactic.
What This Actually Looks Like Once a Month
Strip everything above down and the rhythm is small. One hour deciding what to say, drawn from things customers actually told you. A one-page brief for each of the two ideas you can afford to judge this month. Three executions of each, from whichever of the three sources is realistic in the week you are actually having. A couple of hundred dollars set aside so the answer to "who makes this" is never silence. And two things filmed for the shelf.
That is not an assembly line. It is a Tuesday afternoon, repeated, and it is the thing standing between the advice everybody gives and an account that can actually act on it.
If you would rather that Tuesday afternoon belonged to somebody else, that is a large part of what we do when we run Meta ads for a business: deciding what the ads should say, and making sure there is always something ready to run. And if you would rather build the habit in-house, everything above is the whole method. There is no part of it we have left out.
