In-House vs Agency for Meta Ads: The Real Cost-and-Capability Breakdown at $50k-$500k/Month

"We tried an agency, it didn't work, so we built in-house." A founder said that to me on a call last month, almost as a throwaway line, and I've been chewing on it ever since.

Because the way he said it made it sound like a clean verdict. Agency bad, in-house good, lesson learned. But when we actually walked through his numbers, the real story was messier and a lot more useful. The agency hadn't failed because agencies are worse than in-house. It had failed because he'd hired the wrong agency, given them almost nothing to work with, and then judged them on a metric that was never theirs to control.

So let me do the thing nobody selling you either option will do honestly. Lay out the four real ways to run Meta ads, what each actually costs once you count everything, and the spend level where my own answer flips from "hire an agency" to "don't".

I'm writing this for brands spending between ~A$50k and ~A$500k a month on Meta. Below that the maths is different. Above it, you're a small media company and you already have a team.

What you're actually buying, before we talk price

Quick reframe, because it changes how you read every number below. You are not buying "someone to run the ads". The button-pushing inside Meta is maybe a fifth of the job now. What actually moves an account at this spend level is creative volume and creative quality. The research, the angles, the briefs, the editing, the rate at which you can put fresh, well-aimed shots on goal every week.

So the real question for every option isn't "can they manage the account". It's "how many good creatives a week can this setup reliably produce, and at what all-in cost". That's the lens that makes the comparison make sense.

I measure that in concepts per week. A genuinely productive setup at this spend level lands somewhere around 5 to 10 fresh concepts a week, done with real intent rather than churned out to hit a number. Keep that benchmark handy.

Option 1: The senior in-house hire (US or AU)

The dream version. One excellent person, in your time zone, fully yours.

The honest cost is higher than the salary line. A strong senior media buyer or creative strategist in the US or Australia is realistically ~A$120k to ~A$160k a year base. Load on payroll tax, super, tools, and the rest and you're closer to ~A$13k to ~A$16k a month for one head. And that one head does not produce 5 to 10 concepts a week alone. They can strategise and buy, but they can't also shoot, edit, and design at volume. So in practice you're soon hiring a second person.

What you get is genuinely good: full context, instant feedback loops, someone who lives inside your brand and cares. That depth is real and it's the reason in-house wins at the top end.

The catch nobody mentions: the best people already have jobs. The person you want isn't sitting in your applicant pile, because they're not applying. You have to go and find them, usually by poaching from an agency or a brand that's clearly doing good work, and convince them your vision is worth the move. That's weeks of your time before anyone starts. And there's churn risk baked in. The person you headhunted can be headhunted right back, and the day they leave, a big chunk of your account knowledge walks out the door with them.

My take: brilliant at scale, slow and fragile as your only line of defence early on.

Option 2: The offshore hire

The tempting one, because the price tag looks incredible. And it's where I see the most expensive mistakes made, which is the opposite of what people expect.

The cost is real and low. A skilled video editor or buyer in the Philippines or Latin America might run ~A$1.5k to ~A$4k a month. To put the gap in perspective: a role you'd pay ~A$60k for locally can sometimes be filled for closer to ~A$10k a year offshore. That's not a rounding error, it's a different category of spend, and it's why so many founders reach for it first.

Here's the thing though. Cheap labour is only a bargain if it produces. The pattern I see again and again is a founder hiring the cheapest editor off a job board, getting work that's technically fine and strategically dead, and quietly concluding that "outsourcing doesn't work". It's not that offshore doesn't work. It's that the bottom of the market doesn't, the same way the bottom of any market doesn't.

There's also a structural shift worth naming. Global talent is getting better and cheaper fast, and the gap on the pure skill side is closing. A capable offshore creative strategist today is a genuinely strong hire, not a compromise. But "capable" and "cheapest on the board" are not the same person, and the price difference between them is the whole ballgame.

My take: a fantastic force multiplier on execution once you know exactly what good looks like and can brief it tightly. A trap if you're using it to avoid figuring that out, because you'll get exactly what you pay for and blame the wrong thing.

Option 3: The hybrid pod

The one most founders never deliberately design, and the one that quietly works best for a lot of brands in the middle of this range.

The shape is one strong, in-time-zone brain on top of cheaper execution underneath. So a senior strategist or buyer (your Option 1 head) directing a couple of offshore editors and a designer (your Option 2 hires). The senior sets the angles and holds the standard; the pod produces the volume.

The cost lands in the middle, as you'd expect: call it ~A$16k to ~A$22k a month all-in once you've got the senior plus two or three execution seats. For that you can realistically hit the 5 to 10 concepts a week mark, because you've finally got the strategy and the production capacity in the same system instead of one person trying to be both.

The reason it's not the default is that it's the hardest to build. You have to make two or three good hires that fit together, set up the feedback loops, and run the thing. Most founders don't have the time or the appetite to assemble it from scratch, which is half the reason agencies exist at all.

My take: the best long-term answer for a brand committed to owning this, with someone (often the founder, early on) willing to do the management work to stand it up.

Option 4: The agency

The fast one. You're renting a whole pod that already exists and already works together.

The cost is more honest than it looks. A real agency at this level might have a floor around ~A$10k a month, and here's a thing founders rarely hear from the other side: at that floor, the agency is often barely breaking even on you. Run a team of twenty across five clients and payroll alone is well north of ~A$50k a month, so a A$10k client only makes sense if they can scale you up. Which tells you what a good agency is actually incentivised to do: grow your account, not just maintain it.

What you get for that, on day one, is a whole stack you'd otherwise spend three months assembling: a creative strategist, editors, a designer, a buyer, often a CRO person, all already working as a unit. The speed-to-capability is the real product. You skip the hiring, the team-building, and the churn risk in one move.

The honest downsides are just as real. Diluted focus, if they're running thirty brands with junior strategists, your account gets a fraction of a tired person's attention. Thin research, the failure mode I'd watch hardest: a lot of agencies fill out the onboarding questionnaire and start cranking generic "three reasons why" scripts, instead of going six levels deep into Reddit threads and competitor reviews to find the line that makes your customer go "wait, that's me". And concentration risk: if the agency is the only thing scaling you and they sell or lose your strategist, you can be in genuine trouble inside a 30-day notice period.

My take: the right first move for most brands in this range who don't yet have an in-house engine, on one condition. You have to treat the agency like an internal hire, not a vending machine. Give them the data, the access, and the learnings you'd give an employee, and call problems out early enough that they can fix them. The hands-off "I'm paying you, go" relationship fails every single time, and when it does, the founder remembers it as "the agency didn't work".

The hidden costs nobody puts in the spreadsheet

Here's where most of these comparisons quietly lie to you. The visible cost is the monthly fee or salary. The real cost has three more lines, and they're the ones that decide the actual winner.

Your management time. Every option except a well-run agency leans on your hours. The in-house team needs leading, the offshore hire needs briefing and checking, the pod needs running. If your time is worth ~A$200 an hour to the business and a setup eats eight of your hours a week, that's an invisible ~A$6.4k a month on top of the sticker price. Founders almost never count this, and it's often the single biggest line.

Creative volume, priced properly. A cheap option that produces two usable concepts a week and an expensive one that produces eight are not the same purchase at different prices. They're different products. Always divide the all-in cost by concepts-per-week before you compare anything, or you're comparing a bicycle to a car on sticker price alone.

Churn and key-person risk. A single senior hire carries the risk that your whole account lives in one person's head. The day they leave, you don't just lose a salary, you lose the institutional memory and you eat weeks of re-hiring and re-ramping. A pod or an agency spreads that risk across several heads. It's worth a real premium that never shows up as a number.

Put those three back into the maths and the option that looked cheapest on the fee line often isn't cheapest at all. That's the whole reason "we tried the cheap version and it didn't work" is such a common story.

Where I'll admit an agency stops being the answer

Now the contrarian bit, because I run an agency and I'd rather be straight with you than sell you something that's wrong for your stage.

There's a spend level where the agency model stops being your best option, and it's lower than most agencies will admit. Somewhere north of ~A$300k to ~A$400k a month on Meta, creative is so plainly the entire business that owning it in-house becomes the obvious move. At that point your account can comfortably carry a senior strategist plus a real production team, and the depth, speed, and control of an in-house engine beats renting one. The brands I see scaling cleanly past that line almost always have a strong in-house team at the core, with agencies bolted on for extra volume or a specific gap, not running the whole show.

So the honest progression I'd actually recommend looks like this. Early in this range, an agency to get a real engine running fast without three months of hiring. In the middle, either keep the agency on tight management or start building your hybrid pod. Toward the top, bring the core in-house and use agencies for what they're genuinely best at: filling gaps and adding capacity, not being your only line.

The mistake is thinking it's a permanent either/or. It's a sequence, and the right answer changes as you scale.

And one more thing, because it cuts both ways. An agency that swings and misses on your brand isn't automatically a bad agency, the same way one bad in-house hire doesn't mean in-house is broken. Sometimes it's timing, sometimes it's the wrong person on your account that week. The brands that win are the ones that diagnose why a thing failed instead of writing off the whole category.

Where to from here

So before you copy whatever the last founder you spoke to did, run your own numbers. Take your real Meta spend, write down all four options, and cost each one properly: the fee or salary, plus your management hours, divided by the concepts per week it can actually produce. Then ask which tier you're honestly in right now, not which one you'd like to be in.

I reckon a fair few of you will be surprised. The setup you assumed was cheapest usually isn't once your time and creative volume are in the column.

If you want to sense-check your own breakdown, reply and tell me your spend and your current setup, and I'll tell you straight which tier I'd put you in and why. Which line in that spreadsheet do you think you've been quietly ignoring?