We get asked about Meta account structure constantly: CBO or ABO, bid caps or cost caps, how many campaigns, how many ad sets, how many ads per ad set. Everyone wants the optimal setup, as if there's one correct answer waiting to be discovered.
There isn't. But because we sit across roughly a hundred ad accounts, a dozen of them spending over $1M a month and a larger group spending over $100k a month, we get a real vantage point into how the accounts that actually scale are structured. So instead of giving you a theoretical answer, here's the exact structure of three real accounts spending $1M to $3M a month on Meta, and the patterns that show up across all three.
Account 1: $0 to $3M a Month in Under 8 Months
This is the fastest ramp we've seen. The business launched in mid-2025 and hit $3M in a single month of spend roughly six to eight months later, without the kind of funding that would make that pace unremarkable.
They run CBO (Campaign Budget Optimization), setting one budget at the campaign level and letting Meta distribute it across ad sets. On bid strategy, they started with bid caps and rode them to about $500k a month, which is a pattern we see constantly: bid caps work well until they suddenly don't, and an account that was scaling smoothly sees daily spend collapse to almost nothing. This account hit that wall and switched to cost caps, where it's stayed stable at significant scale ever since. If you're running bid caps right now, know that almost every account we've watched do this eventually makes the same switch.
Budget management here is worth noting specifically: they set campaign budgets meaningfully higher than what actually gets spent. A $30k daily budget might only see $10-15k a day actually go out, because the cost cap is acting as the real constraint, not the budget line. That's a deliberate choice, not a mistake, it gives them a ceiling to raise if they want more aggressive delivery, without having to touch the budget field constantly.
Their ad set structure is distinct: multiple ad sets per campaign, but each ad set holds only one concept, usually with two to four variations of that same concept living inside it. It's a versioning approach, iterate within the ad set rather than spreading distinct ideas across ad sets.
And the ad count is large, over 700 ads live in the account. But only about 10% of those ads are driving roughly 80% of both spend and sales. Nine hundred ads is a lot of surface area, and almost all of the actual result is coming from a small fraction of it.
Account 2: $1M a Month With the Simplest Structure We've Seen
This account spends a consistent $1M a month, also on CBO, also on cost caps (they never ran bid caps at all). Where account one set budgets well above actual spend, this one runs on-target: whatever the budget says is roughly what goes out the door. You could argue they'd get even more efficiency by tightening the cost cap further, but at this level of consistency, it's clearly working.
The ad set structure here might be the simplest 7-figure account we've come across: one single campaign driving the entire million dollars a month, three ad sets inside it, and 10 to 30 ads per ad set. But unlike account one's approach, each of those ads is a genuinely distinct concept, different formats, different visuals, different messaging, rather than variations on a single idea packed into a dedicated ad set.
Total ad count sits around 75, noticeably fewer than account one's 700+. And the distribution is looser too: about 30% of ads are driving 70% of spend, compared to the tighter 10%/80% split in account one. Fewer total ads, and a slightly broader base of winners carrying the account.
Account 3: $1M a Month Running Highest Volume Instead of Cost Caps
The third account also spends about $1M a month on CBO, but takes a different approach to bidding entirely: Highest Volume, rather than cost caps. Instead of telling Meta "get me a $50 CAC," you're telling it "get me as many customers as you possibly can" with no target cost attached. The upside is that you reliably spend your full budget; the tradeoff is that CAC tends to run higher than a well-tuned cost cap would deliver. It's often a good starting point precisely because it forces full delivery, with cost caps introduced later once there's enough data to know what target is realistic.
Budget management is unsurprisingly on-target here too, Highest Volume is built to spend exactly what you tell it to.
Structurally, this account runs closer to ten campaigns, split out by country or product line, each with a single ad set inside it holding just five to ten ads. Like account two, these are distinct concepts rather than variations of one idea. Total ad count lands around 150, and the power law shows up again: roughly a third of the ads are carrying the bulk of the spend.
What's Consistent Across All Three
A few things hold true regardless of how different these accounts otherwise look:
CBO, not ABO, at scale. None of these three accounts run Ad Set Budget Optimization. That's not a rule, if ABO is genuinely working for your account, there's no reason to change it just because of this pattern, but it's worth noting that we're not seeing ABO show up on the seven-figure accounts we have visibility into right now.
Bid caps tend to be a phase, not a destination. The accounts that use bid caps at all treat them as a way to get initial traction, then move to cost caps or Highest Volume once they hit a ceiling. We haven't seen a mature seven-figure account running bid caps as its long-term strategy.
There's no single correct ad set architecture. One account packs variations of a single concept into dedicated ad sets. Another spreads ten to thirty genuinely distinct concepts across just three ad sets total. A third runs five to ten distinct concepts per ad set across ten separate campaigns. All three scale to seven figures a month. Structure follows what the business and creative pipeline actually support, not a universal template.
The power law is unavoidable. Every account here, regardless of ad count or ad set architecture, has a small share of ads driving the large majority of spend and sales. The tightest concentration we saw was 10% of ads driving 80% of results. The loosest was closer to 30% driving 70%. Either way, most of what gets made in a high-spend account is not what's actually moving the number.
What This Means If You're Building the Creative Behind These Accounts
This is a hits business. Most of what you make is going to underperform, and that's not a failure signal, it's the mechanism by which the handful of genuine winners get found. We've watched a single ad scale to roughly $500,000 a month in spend inside one account. That kind of outlier is what the other 80-90% of underperforming creative is quietly funding the search for.
The practical implication for account structure and creative production is the same one either way: you need enough genuinely distinct concepts flowing into the account to give the power law something real to work with. An account stuffed with variations of one or two ideas, no matter how cleanly it's structured on the media buying side, has a much smaller pool of concepts competing to be that next $500k-a-month outlier. The accounts here that spread real conceptual diversity across their ad sets, not just more of the same idea, are giving themselves more shots at the kind of win that actually moves the account.
This is the same dynamic we've written about with Meta's Andromeda system, which increasingly rewards accounts fed with genuine creative diversity rather than raw volume. Building and sustaining that diversity month after month, not just producing a lot of ads, is exactly what our creative process is designed around.
If your team is producing plenty of ads but they're mostly variations of a small handful of ideas, that's usually the bottleneck, not your account structure. Biddyco is a creative agency for Meta ads built to feed accounts like these with the kind of conceptual range that gives the power law more to work with, for brands spending $100k-$5M+/month who need volume and genuine diversity, not just more of what's already been tried.
