How we structure Meta ad campaigns
Fewer campaigns, broader targeting, more creative. The account structure we use for consumer brands on Meta, and the reasoning behind it.
Meta ad accounts tend to grow complicated. A campaign for each audience, an ad set for each interest, a separate setup for every promotion. After a year, many accounts have dozens of campaigns, each with too little budget to learn anything.
Our approach goes the other way. Meta’s delivery system is very good at finding buyers when it has enough data and enough creative to work with. The job of the account structure is to give it both, and to keep testing separate from scaling.
The core principle: consolidate
Every campaign and ad set needs enough conversions to optimise properly. Splitting budget across many small segments starves each of them. So we keep the structure lean:
- One main scaling campaign, with broad targeting and the best-performing ads.
- One testing campaign, where new creative gets a fair chance with dedicated budget.
- Optional supporting campaigns, only when there’s a clear reason, such as a distinct product line, a major promotion or a specific retargeting need.
That’s often it. Simpler accounts are easier to read, faster to learn and less prone to campaigns competing against each other in the same auctions.
Broad targeting, creative-led
We rely mostly on broad targeting, with age, location and occasionally gender set where the product genuinely requires it. Interest stacks and narrow lookalikes made sense when the platform needed help finding people. Today they mostly restrict the system.
That means the creative does the targeting. An ad about a specific problem, featuring a specific kind of person, in a specific tone, will find the people who respond to it. If we want to reach a new segment, we make ads for that segment rather than building a new audience.
Advantage+ shopping campaigns
For many consumer brands, Meta’s Advantage+ shopping campaigns work well as the main scaling engine. They automate much of the targeting and placement, and they tend to perform best with:
- A large, varied set of proven ads.
- Clean conversion tracking, including the server-side Conversions API alongside the pixel.
- A sensible cap on how much budget goes to existing customers, so the campaign keeps acquiring new ones.
We still run a manual campaign alongside for testing, because it gives more control over how new ads are evaluated.
How testing works
New creative goes into the testing campaign, grouped by concept so each idea gets a fair share of spend. We give each concept enough budget and time to show a clear signal, then judge it on cost per purchase and the quality of early indicators like hold rate and click-through.
Winners move into the scaling campaign. Losers get switched off, but the learning goes into the log. Was it the hook, the format, the message or the audience the ad implied? That record shapes the next round of briefs.
Budget and patience
Two habits make a big difference:
- Change things gradually. Large, sudden budget changes or constant edits reset the system’s learning. We adjust in measured steps and give changes time to settle.
- Judge on the right window. Daily results are noisy. We look at rolling multi-day performance and blended business numbers before making big decisions.
What still matters most
Account structure is the plumbing. It needs to be clean, but it isn’t what makes an account grow. Growth comes from a steady supply of creative that gives people real reasons to buy, sent to a page and an offer that convert. Structure simply makes sure the platform can find the right people for that creative and learn quickly from what works.