How Scores Work
AEOrank produces a single score from 0 to 100 by measuring 36 criteria across 5 pillars and applying percentage-weighted aggregation.
How it works
Section titled “How it works”1. Criterion scoring
Section titled “1. Criterion scoring”Each of the 36 criteria is scored from 0 to 10 based on specific checks against your site.
2. Percentage weights
Section titled “2. Percentage weights”Each criterion has a percentage weight. All 36 weights sum to exactly 100%.
final_score = Σ(criterion_score / max_score × weightPct)Where weightPct is the criterion’s percentage weight (e.g., 7 for 7%).
3. Example calculation
Section titled “3. Example calculation”| Criterion | Score | Weight (%) | Contribution |
|---|---|---|---|
| Topical Authority | 0/10 | 7% | 0.0 |
| FAQ & Speakable | 6/10 | 5% | 3.0 |
| llms.txt Presence | 0/10 | 5% | 0.0 |
| Content Structure | 8/10 | 5% | 4.0 |
| E-E-A-T Signals | 7/10 | 6% | 4.2 |
| AI Crawler Access | 5/10 | 3% | 1.5 |
| (30 more criteria) | … | … | … |
Each criterion contributes (score / 10) × weightPct points to the total. The sum across all 36 criteria is your final score out of 100.
4. Score Gates
Section titled “4. Score Gates”Two gates can cap your final score regardless of weighted totals:
Coherence Gate: If topic-coherence scores below 6/10, your overall score is capped. A site without topical authority cannot reach high AEO scores even if all other criteria pass — AI engines require a coherent subject focus to trust and cite a source reliably.
Duplication Gate: If 3 or more duplicate content blocks are detected across pages, your score is capped at 35. Pervasive duplication signals low-quality content that AI engines should not cite.
5. Status thresholds
Section titled “5. Status thresholds”Each criterion also gets a pass/warn/fail status based on its percentage of max score:
| Status | Threshold | Meaning |
|---|---|---|
| ✓ Pass | ≥ 70% | Criterion is in good shape |
| ⚠ Warn | 40–69% | Room for improvement |
| ✗ Fail | < 40% | Needs attention |
Determinism
Section titled “Determinism”AEOrank scoring is deterministic. The same URL scanned twice produces the same score. This is by design — over 80% of scoring weight comes from structural, deterministic signals that don’t change between scans.
Improving your score
Section titled “Improving your score”Focus on high-weight criteria first, organized by pillar:
Answer Readiness (26%) — Start with topical authority and original data. These have the highest individual weights (7% and 5%) and directly gate your maximum possible score.
Content Structure (25%) — Fix content structure (5%), then answer-first formatting and Q&A format (4% each). Structural improvements lift multiple criteria at once.
Technical Foundation (25%) — Add llms.txt (5%) and FAQ & Speakable markup (5%). These are binary checks with large weight — zero investment for full points once deployed.
Trust & Authority (12%) — Strengthen E-E-A-T signals (6%) with author information, publication dates, and citations.
AI Discovery (12%) — Address page freshness, canonical URLs, and sitemap last. Important for discovery but lower individual weights.