Meta Ads Account Structure in 2026: How to Build for Andromeda, Testing & Scaling

What does a Meta Ads account structure look like in 2026?

A Meta Ads account still follows the familiar hierarchy: campaigns contain ad sets, and ad sets contain ads. What has changed is how much manual segmentation advertisers need to create inside that hierarchy.

The mistake is assuming more layers automatically create more control. Every additional campaign and ad set divides budget, conversion data and your team's attention. Structure should earn its place.

What Andromeda changed about campaign structure

Meta's Andromeda system is part of a broader shift towards machine-learning-led ad retrieval and delivery. It does not create a secret new campaign structure, and it does not mean advertisers should blindly collapse every account into one campaign.

It does make artificial complexity less attractive. When Meta can use conversion signals and understand creative more effectively, the media buyer's advantage increasingly comes from providing better inputs and making better decisions rather than manufacturing dozens of narrow audience variations.

Our Meta Andromeda guide explains this shift in more detail. The practical implication for structure is simple: remove restrictions that have no clear purpose, but keep the ones that protect a real business requirement.

Should you use one campaign and one ad set?

Sometimes. But “one campaign, one ad set” is not a universal 2026 rule.

A highly consolidated structure can work well when you have one primary objective, comparable products or offers, reliable purchase data and no genuine reason to force spend into separate segments. It gives Meta a larger pool of signal and fewer artificial boundaries.

But consolidation becomes a problem when different parts of the business have materially different economics or when you need a controlled experiment. A new country, a different margin profile, a product launch or a test that needs guaranteed spend can justify separation.

The question is not “How few campaigns can I run?” It is “What does this separation allow me to decide that I could not decide otherwise?”

Let conversion volume determine how much structure you can support

Budget alone does not determine structure. Conversion volume matters because Meta needs enough outcome data to optimise delivery and because you need enough evidence to interpret performance.

If an account generates relatively few purchases, splitting those purchases across many campaigns and ad sets can create thin, volatile data. In that situation, consolidation may improve signal density and make the account easier to diagnose.

Higher-volume accounts can support more purposeful separation because each part of the structure may still receive enough spend and conversions to be useful. Even then, complexity should have a reason.

If you are restructuring mainly because campaigns show Learning Limited, read our Meta Ads Learning Phase guide before changing the account. A learning label is information, not a business KPI.

Example structures for e-commerce Meta Ads

Lower-budget account

For an e-commerce brand with one core market, one main purchase objective and limited conversion volume, a simple sales structure can often make more sense than several prospecting, interest, lookalike and retargeting campaigns competing for a modest budget.

Keep the structure compact, make sure purchase tracking is reliable and invest the available testing capacity in genuinely different creative concepts.

Growing account

As spend and conversion volume increase, you may need separation for a meaningful creative test, a product category with different economics or a market that needs its own budget. The key is that each additional campaign has a job.

Higher-spend, multi-product account

A larger advertiser may need distinct structures for markets, product groups, acquisition economics or controlled experiments. Consolidation still matters, but “simple” does not have to mean “one campaign”. The objective is the minimum structure required to make good commercial decisions.

Does prospecting still need separate audience ad sets?

Not automatically. Creating separate interest, lookalike and broad ad sets simply because that was the historical playbook can fragment spend without producing a useful decision.

Audience controls still matter where there is a real restriction or hypothesis. But Meta's move towards Advantage+ Audience and more automated delivery means advertisers should be careful about treating narrow segmentation as the default source of performance.

Our Meta Ads targeting guide for 2026 explains when broad delivery, interests, lookalikes and Advantage+ Audience controls make sense.

Should retargeting have its own campaign?

Retargeting does not automatically deserve a separate campaign either. Separate it when you need deliberate budget control, distinct messaging, a specific exclusion strategy or a clear measurement reason.

Be especially careful when interpreting retargeting ROAS. People who already visited the site or considered the product are closer to purchase, so a strong reported return does not necessarily mean the retargeting campaign created all of that demand.

Before moving budget based on reported ROAS, understand your Meta attribution settings and compare platform reporting with commerce data.

Where should creative testing live?

There are two different jobs that advertisers often mix together: maximising current performance and buying information.

If a new creative needs enough spend to answer a defined hypothesis, a controlled testing structure can be useful. If you put every new ad into a performance campaign and Meta gives it almost no delivery, you may never gather enough evidence to answer the test.

But a separate testing campaign is not mandatory. If the main campaign can give new concepts meaningful delivery and you can evaluate them against a clear hypothesis, adding another campaign may only create overhead.

Use our Meta Ads creative testing framework to define the hypothesis, evidence and decision before choosing the structure.

CBO vs ABO inside your account structure

Campaign structure and budget structure are related but not identical. Advantage+ campaign budget, commonly called CBO, lets Meta distribute a shared campaign budget across eligible ad sets. Ad-set-level budgets, commonly called ABO, let you control allocation directly.

Use shared budget when the ad sets pursue comparable outcomes and you want Meta to allocate towards the strongest available opportunities. Use controlled budgets when a segment or experiment genuinely needs a defined amount of spend.

Our CBO vs ABO guide covers the decision in depth.

How many ads should you run per ad set?

There is no universal number. The better question is whether the ads give Meta and your team genuinely different information.

Ten cosmetic variations of one product image are not necessarily ten useful creative ideas. A smaller set covering distinct motivations can create more useful diversity: product demonstration, problem and solution, customer proof, comparison, founder explanation, offer, lifestyle or objection handling.

Watch delivery as well as quantity. If spend is spread so thinly that most ads never generate meaningful evidence, adding more creative may create inventory rather than learning.

When consolidation can hurt performance

Consolidation is useful when it removes unnecessary fragmentation. It becomes harmful when it removes a control the business actually needs.

• Markets have materially different economics or currencies.
• Product groups have different margins, inventory constraints or targets.
• A new test needs protected spend to generate evidence.
• A business unit requires a fixed budget.
• Different objectives or optimisation events are genuinely required.
• Customer acquisition and retention activity need distinct commercial treatment.

The goal is not maximum consolidation. It is purposeful structure.

When should you create a separate Meta campaign?

Create another campaign when the separation changes a real decision. Good reasons can include a different objective, country, product economics, budget commitment, optimisation requirement or controlled experiment.

Weak reasons include “we have always done it this way”, duplicating a campaign to chase yesterday's result, or creating a new campaign for every minor audience idea.

How account structure affects scaling

Scaling a fragmented account can spread incremental budget across too many small pools. Scaling an over-consolidated account can make it difficult to protect an important market or experiment. Structure therefore needs to support the next stage of spend, not only today's snapshot.

Before increasing budget, know the CPA and ROAS the business can sustainably afford. Use the KARB Target CPA Calculator and Break-Even ROAS Calculator to establish the commercial boundary, then use our Meta Ads scaling framework to decide whether to scale, hold or investigate.

Common account-structure mistakes agencies inherit

• Multiple campaigns doing essentially the same job.
• Legacy interest and lookalike ad sets kept alive without a current hypothesis.
• Small budgets divided across too many ad sets.
• Retargeting isolated purely because it historically was.
• Testing structures with no defined question or decision rule.
• Different products mixed together despite materially different economics.
• Budget changes made from reported ROAS without checking attribution and margin.
• New campaigns created to react to short-term volatility rather than diagnose the cause.

For an agency, these mistakes are multiplied across every client. The operational problem is not just knowing what a clean structure looks like. It is identifying which accounts need attention first.

A practical Meta Ads account-structure audit

Start by writing the purpose of every campaign in one sentence. If two campaigns have the same objective, economics and job, ask why they are separate. Then check whether each campaign and ad set receives enough spend and conversion volume to justify its existence.

Next, check whether poor performance is really structural. Weak creative, broken signals, unrealistic targets or landing-page problems will not be fixed by rearranging campaigns.

For agencies, structure is a portfolio problem

A media buyer can manually inspect one account. An agency managing dozens of client accounts has a different problem: knowing which account is fragmented, which creative is deteriorating, where spend is being wasted and which campaign is actually ready to scale.

KARB is built around that workflow. It helps e-commerce performance agencies review Meta Ads accounts, identify what needs attention and move from data to clearer optimisation and scaling decisions without manually digging through every account.

Frequently asked questions

What is the best Meta Ads campaign structure in 2026?

There is no universal best structure. Use the minimum structure required to support your objective, conversion volume, economics and genuine control requirements. Avoid fragmentation that exists without a clear commercial or experimental purpose.

How many campaigns should I run on Meta Ads?

As many as you can justify with distinct jobs, not as many as you can create. If two campaigns pursue the same outcome with the same economics and no meaningful control requirement, consider whether they should be consolidated.

How many ad sets should I have per campaign?

There is no fixed ideal number. Each ad set should have a reason to exist and enough budget and signal to be useful. Too many small ad sets can fragment delivery and make results harder to interpret.

Does Andromeda mean I should use one campaign?

No. Andromeda strengthens the case against unnecessary manual segmentation, but it does not remove legitimate reasons to separate markets, products, budgets or experiments.

Should retargeting be a separate campaign?

Only when separate budget control, messaging, exclusions or measurement make it useful. Do not separate retargeting purely because it was the historical default.

Should I use CBO or ABO?

Use campaign-level budget when you want Meta to allocate spend across comparable opportunities. Use ad-set-level budgets when you need deliberate allocation for a business requirement or controlled test.

Should creative testing have its own campaign?

It can, when a test needs protected spend to gather evidence. If new concepts can receive meaningful delivery inside the main campaign, a separate testing campaign may not be necessary.