1. Data Collection
All data is collected through AuditMe's crawl engine (v2.4), which performs breadth-first crawling of websites. The crawler:
- Respects robots.txt directives
- Follows canonical tags and redirects
- Parses HTML5 semantic elements
- Validates JSON-LD structured data
- Measures Core Web Vitals via CrUX API
- Checks mobile viewport and responsive design
2. Sample Selection
Research samples are drawn from websites that have been audited through AuditMe. To reduce selection bias:
- Stratified random sampling across industries
- Minimum 5 pages required for inclusion
- Geographic distribution across US, EU, and global sites
- Site size distribution: small (<100 pages), medium (100-1000), large (1000+)
3. Analysis Framework
Each website is evaluated across 13 audit categories. Within each category, individual checks produce pass/warn/fail results:
category_score = (passed_checks / total_checks) × 100
Overall scores use weighted averaging across categories. Weights are based on observed impact on search visibility from our analysis.
4. Severity Classification
Blocks indexing or ranking. Immediate action required.
Significant SEO impact. Should be fixed within 1 week.
Moderate improvement possible. Fix within 1 month.
Best practice, minimal direct impact. Fix when convenient.
5. Known Limitations
- Sample may skew toward SEO-conscious website owners
- CrUX data requires minimum traffic thresholds
- JavaScript-rendered content may not be fully captured
- Dynamic content (A/B tests, personalization) may vary between crawls
- Server-side rendering configurations can affect results
6. Reproducibility
We publish methodology documentation, raw data (CSV), and audit engine version numbers so that other researchers can verify or extend our findings. All data files include timestamps and engine version identifiers.