How Many Ad Sets Per Campaign: The Consolidation Rule

Use the smallest campaign structure that preserves distinct commercial decisions. More groups divide evidence; fewer can hide differences in service eligibility, economics or ownership. A fixed group count does not establish an optimum.

Keep the evidence calculation explicit

Hypothetical equal-allocation plan: $100/day for seven days gives $700 total. At an assumed $20 CPA, ten equally budgeted groups receive $70 each and imply 3.5 expected outcomes; three groups receive $233.33 each and imply 11.7 outcomes. These are uncertain expectations, not observed orders, sufficient samples or a promise of a learning status. A shared campaign budget need not distribute equally.

Illustrative market variation: market A contributes $35 per paid order before ads; market B contributes $12. At $20 CPA, A leaves $15 but B loses $8. Those are assumed inputs. Combining them can hide B’s loss in the aggregate; retain commercial controls or separate reporting rather than consolidate solely to increase event count.

A consolidation protocol

1. Inventory each group’s goal, offer, event, audience boundaries, creative, owner and actual mature economics.
2. Identify redundant hypotheses, then state what will be combined and which market/margin boundaries must remain.
3. Hold offer, event, creative and outcome definitions stable for the structure comparison. Record unavoidable changes.
4. The user approves incremental spend and cash-loss caps. Choose a maturity window from actual conversion/refund timing.
5. Compare mature qualified orders and contribution, plus actual allocation. For a causal effect use a supported experiment; before/after comparison alone is descriptive.
6. At a cap, tracking defect or boundary failure, stop/revert the affected change. Sparse outcomes end as inconclusive rather than a forced winner.

Turn expected events into an approved plan

In NovaStorm, prepare a brief naming each retained group’s decision and save the proposal as paused campaign and local ad-set drafts. Include the $700 example only as a planning assumption; the user confirms the actual group count, amounts, currency and risk. Check saved settings, then compare actual delivery with the approved plan after publication.

Use the budget-level guide for allocation and the learning-phase guide to keep platform status separate from commercial evidence.