Meta Ads Account Diagnosis: An Evidence-Led Workflow
How do I diagnose a Meta ads account problem systematically?
Rank competing hypotheses by evidence integrity and economic consequence rather than by whichever dashboard metric moved. Each hypothesis needs a specific extract that could falsify it. A falsified branch closes; an unresolved one stays open rather than being quietly replaced by a preferred explanation.
For the surrounding account decisions, compare Why are my Facebook ads not delivering? and use Meta Ads Tools: An Evidence-Led Evaluation Framework as the next diagnostic.
In short
This guide owns one decision artifact: the filled, auditable structure below. Reader-supplied thresholds stay explicit; missing evidence stays missing.
Ranked account diagnostic
| Rank / symptom | Competing hypotheses | Required extract | Falsifier | Next isolating check | Economic consequence | Do-not-act condition |
|---|---|---|---|---|---|---|
| 1. Commerce contribution falls | Price/cost/refund mix versus acquisition deterioration | Orders, recognized revenue, COGS, fulfilment, fees, refunds, spend by cohort | Contribution stable after reconciled costs | Split new/repeat, offer, product, and channel cohorts | Cash contribution at risk | Costs or order states are immature/missing |
| 2. Reported conversions move abruptly | Tracking/schema change versus real order movement | Raw events, releases, platform export, commerce orders | Commerce and valid event occurrence agree | Trace one order and compare event/order coverage | Decisions may use false numerator | Tracking integrity unresolved |
| 3. Spend/delivery changes | Budget/bid/status versus auction/audience mix | Change log, budgets, bids, status, placement/audience delivery | No relevant constraint or edit precedes onset | Stratify CPM, reach, frequency, eligibility | Lost volume or unintended cash exposure | Account/billing/policy incident active |
| 4. Creative response changes | Asset fatigue versus placement/audience shift | Asset-version impressions/clicks plus mix extracts | Within-stratum CTR remains stable | Matched incumbent/replacement test | Production spend or missed response | Asset IDs or delivery strata cannot join |
Contribution per recognized order = (recognized revenue − COGS − fulfilment − shipping − fees − refunds/returns − acquisition spend) ÷ (matured recognized orders). Paid ROAS = Meta-attributed revenue ÷ Meta spend and is diagnostic credit, not a universal primary metric.
Interpretation boundary
Rank hypotheses by evidence integrity and economic consequence, not by a preferred dashboard metric. A falsified branch is closed; an unresolved higher-ranked integrity issue prevents downstream optimization.
Can software help?
Bach.ai audits your connected Meta account, estimates the revenue impact of what it finds, and proposes specific fixes. It applies a change only after you approve it. Think of it as an automated audit layer that surfaces issues and proposed fixes for your review — not a replacement for your team’s judgment, and creative production is not its core job, though the Pro and Agency plans can generate a limited number of variants.
FAQ
Where should you start when diagnosing a Meta Ads account?
Begin with matured contribution, then test tracking, delivery, creative, audience, and economics using a falsifier and next isolating extract.
Which unresolved issues should prevent changes to the account?
Paid ROAS cannot serve as a universal primary metric, and no action proceeds while costs, states, or event integrity remain unresolved.
Can an account diagnostic identify the cause of a performance change?
It isolates competing explanations through the ranked account hypotheses. The resulting branch is a diagnostic decision, not proof that one candidate caused the observed change.
Method and sources
“Rank competing hypotheses by evidence integrity and economic consequence rather than by whichever dashboard metric moved.”
Source: Where this guide describes platform behaviour, it follows Meta’s published advertising and Marketing API documentation, which changes without notice — verify anything load-bearing against the current version before you act on it. Every threshold the guide asks you to supply is first-party, drawn from your own account exports and commerce ledger, because no external benchmark can stand in for your own margin structure.