The clinical framework and objective standards used to compute the A-Score™. Designed for institutional diligence and forensic accountability.
This methodology governs the independent verification of ad spend integrity, derived directly from authentic Meta Marketing API data. It establishes the clinical baseline for generating an advertiser's A-Score™.
The objective of the A-Score™ is to provide a standardized, immutable metric representing the structural health, security, and operational efficiency of a given ad account prior to or during capital deployment. It removes subjective agency reporting and replaces it with cryptographically verifiable platform truth.
The A-Score is an independent 0–100 score of a business’s Meta advertising account, produced by Ad IntellX. It measures how much confidence an advertiser can place in their ad spend by evaluating the account across six domains: ownership and control, spend efficiency, optimization activity, tracking and attribution health, delivery quality, and platform reliance risk. The score is computed from read-only Meta Marketing API data, cross-validated by the Ad IntellX Data Intelligence Engine, under this published methodology (currently v1.2).
An A-Score of 70 or above is considered good. Scores of 85–100 fall in the Verified Strong band, indicating clean tracking, protected ownership, and efficient delivery. Scores of 70–84 fall in the Stable band. Scores below 70 indicate elevated risk to ad spend: 50–69 is At Risk, 30–49 is High Risk, and 0–29 is Critical.
The A-Score has five bands. Verified Strong (85–100): the account shows strong integrity across all six domains. Stable (70–84): minor issues, no structural risk. At Risk (50–69): meaningful inefficiencies or delivery anomalies are reducing spend confidence. High Risk (30–49): structural problems — such as tracking failure or loss of account control — are actively compromising spend. Critical (0–29): the account’s spend cannot be verified as protected and requires immediate attention.
The A-Score is calculated in a fixed pipeline: account data is pulled read-only from the Meta Marketing API, cross-validated against the Ad IntellX Data Intelligence Engine’s signal layers through an integration gate, scored across six weighted domains (Ownership & Control 20%, Spend Efficiency 25%, Optimization Activity 20%, Tracking & Attribution Health 15%, Delivery Quality 10%, Platform Reliance Risk 10%), and then subjected to hard caps. Hard caps set score ceilings for structural failures — for example, an account without verified tracking cannot score above 49 regardless of other domain performance. The lowest applicable cap always wins.
The A-Score Preview is free. It requires no credit card, uses read-only access through Meta’s official OAuth, and completes in about 60 seconds. The full A-Score Verification Record — the complete findings document behind the score — is a one-time paid report priced by the account’s monthly ad spend tier, starting at $99.
No. Ad IntellX is independent and is not affiliated with or endorsed by Meta Platforms, Inc. The A-Score is computed from data returned by Meta’s official Marketing API under read-only permissions: Ad IntellX cannot view private content, cannot modify campaigns, and cannot spend budget. The A-Score documents what independently computable signals show; it does not constitute a determination regarding platform conduct.
Estimated Monthly Waste Exposure is a computed dollar range estimating the portion of an account’s monthly ad spend allocated to structurally compromised or inefficient delivery — for example, spend affected by broken conversion tracking, overlapping audiences bidding against each other, or delivery anomalies. It is an engagement- and signal-derived estimate produced by the scoring engine, not a confirmed or audited loss figure, and it is always presented as a range with a stated confidence level.
The A-Score™ is computed across six distinct domains. Each domain aggregates dozens of API signals into a weighted sub-score.
The six evaluation domains are not computed through a direct API query. They are the output of a five-layer proprietary detection engine — the Ad IntellX Data Intelligence Engine — which processes API-retrieved account data through sequential signal layers before domain subscores are computed.
Layer 1 (Core Detection), Layer 2 (Compound Signals), Layer 4 (Predictive Signals), and Layer 5 (Autonomy Signals, read-only in verification mode) each contribute distinct signal classifications. An integration gate cross-validates engine outputs against API data before any scoring computation begins. This gate is the mechanism that ensures the A-Score™ is produced from verified, integrated data — not from either source in isolation.
Elevated from a standalone reporting metric to a core methodological checkpoint, Platform Signal Integrity evaluates how well the ad account leverages machine-learning optimization capabilities native to the platform.
This includes auditing Advanced Matching parameters, Conversion API (CAPI) redundancy, and server-side latency. An account that fails to pass sufficient deterministic data back to the platform suffers a severe signal integrity penalty, crippling the A-Score™ regardless of top-of-funnel engagement metrics.
Certain critical risk conditions trigger an immediate ceiling on the maximum achievable A-Score™. These overriding rules ensure that foundational flaws cannot be masked by strong performance in secondary areas.
If active campaigns are utilizing ad sets without an attached pixel or conversion event, the account exhibits critical data loss.
If an unauthorized domain or unverified secondary Business Manager holds full admin rights over the primary ad account.
Every generated A-Score™ is backed by a cryptographically signed JSON export. This allows institutional buyers, auditors, and M&A teams to verify the authenticity of the report independently of the Ad IntellX interface.
// Verification Endpoint
GET https://api.adintellx.com/v1/verify/record?id=AX-471983
Every account is first classified for data sufficiency before scoring. This gate exists so that a low-activity or newly-connected account is never scored against an empty window and mistaken for a low A-Score™.
Dormant and Insufficient records are priced and checked out identically to a standard Verification Record, but are framed distinctly at checkout as an A-Score Baseline Verification Record, and are excluded from the aggregated benchmark panel until the account has enough activity to contribute a representative data point.
Records generated in the Dormant or Insufficient state carry a one-time, free re-verification entitlement: within 30 days of the original record, the buyer may re-run the full verification once at no charge, producing a new, linked record for direct before/after comparison. The entitlement expires after 30 days or after first use, whichever comes first.
In accordance with institutional verification standards, Ad IntellX strictly bounds its claims. The A-Score™ and accompanying forensic reports do not claim to determine:
Initial publication of the Ad IntellX A-Score™ independent verification methodology.