A large account structure is not automatically a bad structure. Separate markets, offers, owners, schedules, or economic constraints can justify separate campaigns. The problem appears when the account contains many units whose purpose nobody can explain and whose budgets rarely produce interpretable evidence.
An audit should discover which separations protect a real requirement and which reflect historical habit. “Simpler is better” is not enough reason to merge everything.
Map the business reason for every unit
List each active campaign and ad set with its role, owner, objective, offer, destination, market, budget responsibility, and any special constraint. Record when it was created and whether it still serves the original purpose.
A campaign for a product with a separate supply constraint may deserve its own budget control. A temporary promotion may need a distinct schedule. Two units that differ only because a buyer duplicated the same setup several times deserve closer inspection.
Use the handoff checklist to recover missing context before an AI operator proposes structural changes.
Separate legitimate constraints from reporting preferences
A team may split campaigns because it wants a cleaner report, even when the delivery responsibilities are the same. Sometimes labels, naming conventions, or reporting dimensions can answer the question without creating another independently managed unit.
That does not mean every reporting distinction should be removed. The audit should ask whether the separation changes actual control or merely presentation. If it exists to keep one client's budget separate from another's, it is an operating boundary, not cosmetic complexity.
Write the requirement in plain language. “Separate budget owner and approved monthly ceiling” is more informative than “we always structure accounts this way.”
Assess the evidence available per decision
When a limited budget is divided among many units, each unit may accumulate little evidence. The team can then spend its time reacting to sparse results rather than answering a useful question.
For an illustrative account with a $3,000 test allocation and thirty independent concepts, an equal planning split would be $100 per concept before considering actual delivery. That arithmetic does not define a minimum test budget. It shows why the number of planned comparisons needs to match the decision the account can realistically support.
Use the test-budget planning guide to choose fewer, more meaningful comparisons when necessary.
Build a structure-review table
| Unit | Why it exists | Evidence or control it needs | Audit decision |
|---|---|---|---|
| Regional campaign | Different service coverage and owner | Separate budget and availability | Retain boundary |
| Promotion campaign | Time-limited offer | Defined start/end and terms | Retain while promotion is active |
| Duplicated test set | Historical copy with no current distinction | No unique requirement identified | Investigate consolidation |
| Product group | Different contribution and supply limits | Product-level economics | Retain or redesign deliberately |
The audit decision should include uncertainty. “Investigate consolidation” is not an instruction to merge immediately. The buyer must still consider current tests, naming, historical comparability, and the consequences of a structural edit.
Avoid algorithm folklore as the justification
Meta's Andromeda engineering article explains aspects of ad retrieval and ranking. It does not supply a universal campaign count for every advertiser.
Similarly, public descriptions of automation do not prove that every account should use the same structure. The business can have constraints that a generic recommendation does not know.
Use observable account needs and a testable hypothesis. A proposal such as “combine these two units because they share the same objective, offer, market, and budget owner, and neither has a distinct remaining experiment” is more reviewable than “the algorithm prefers consolidation.”
Plan the change as an experiment in operations
Before editing, record the current structure and the intended new structure. Identify which budgets, ads, audiences, reporting views, and scheduled actions will be affected. Check for automation rules that depend on names or object IDs.
Define the operational objective: easier ownership, less repeated work, a clearer test allocation, or a more coherent budget boundary. State the performance hypothesis separately. A cleaner account can be an operational improvement even when performance lift is not isolated.
Meta Blueprint's measurement course outline distinguishes reporting and experimental approaches. Do not call a structural before-and-after comparison causal proof if other conditions changed.
Preserve a readable history
A structural change can make historical comparisons harder. Keep a mapping from old units to new responsibilities and record the effective date. Explain which comparisons remain valid and which no longer have an equivalent denominator.
Do not delete useful context simply because a campaign is no longer active. The reason for a previous exclusion or failed test can remain relevant to the new structure.
After the change, verify that the intended ads, budgets, destinations, and ownership are present. Then observe delivery at the correct level. A new creative receiving little spend may require an allocation diagnosis, rather than another structural rewrite.
Close with a smaller set of justified responsibilities
The audit should leave each active unit with a reason to exist, an accountable owner, and a decision it can support. It may recommend fewer units, but it may also confirm that several separations are necessary.
The useful outcome is an account that the team can explain and operate. Campaign count is a consequence of that design, not the target itself.