ⓘ Note: Batch scores use a different methodology than the single-site analyzer — expect small variances

The Site Analyzer, Mega Analyzer, and Batch Analyzer share the same 2026-updated signal catalog (llms.txt / llms-full.txt / .well-known/llms.txt / humans.txt / sitemap-image.xml / dateModified / max-snippet / <cite> markup / geo meta / security headers / BreadcrumbList / Article schema / rel=author / srcset / JSON Feed / manifest). The three tools intentionally use different scoring methodologies for their different workflows. A site will typically score within a few points on all three, but the numbers will rarely match exactly. This is by design:

Dimension	Site Analyzer	Mega Analyzer	Batch Analyzer (this tool)
Purpose	Deep single-page audit with copy-paste fixes	Comprehensive single-URL audit with 6 bucket grades + AI prompt	Rank-order comparison across up to 20 sites
Individual checks	~270 checks across Launch / Agency / Network / Hosting	~80 checks across SEO / Schema / E-E-A-T / Voice / Mobile / Perf	~110 weighted features across 9 buckets
Bucket grouping	3 top-level books ($97 Launch, $20 Agency, $100 Network) + Hosting	6 buckets (SEO, Schema, E-E-A-T, Voice, Mobile parity, Perf + AI)	9 buckets (SEO, GEO, AEO, Schema, E-E-A-T, Links, Social, A11y, Hosting)
Within-bucket scoring	Pass/warn/fail counted uniformly	Critical = 2pts, warn = 1pt; percent-of-max	Features weighted by importance (sum of points)
Async deep checks	Yes (CSS fetch, IndexNow chain, feed self-href, canonical HEAD, hreflang return-tags, OG image HEAD, sitemap-index children)	Partial — fetches robots/sitemap/llms/ai/humans/sitemap-image/manifest; no link HEAD checks	No — single synchronous pass for speed across many sites
New 2026 signals	All — checks every signal the catalog tracks	All — checks every signal the catalog tracks	All — scoring functions weight each new signal (llms variants, humans.txt, sitemap-image.xml, dateModified, max-snippet, cite, geo meta, security headers, srcset, BreadcrumbList, Article schema)
E-E-A-T weighting	Wikidata/ORCID are informational — don't drag the score	Wikidata/ORCID weighted moderately; author link and dateModified carry points	Wikidata/ORCID carry hard weight (20 pts combined) — absence is scored
Best for	"Is this page ready to ship?"	"Give me one comprehensive report + a paste-ready AI fix prompt."	"Which of these sites is ahead of the others, and where?"

How to read all three together:

Use the Site Analyzer when you want to ship fixes on one page — its 270 individual checks show every gap with copy-paste code.
Use the Mega Analyzer when you want a single comprehensive grade on one URL plus a ready-to-paste AI fix prompt for Claude/ChatGPT.
Use this Batch Analyzer to rank a set of competitors or your own properties relative to each other across all 9 weighted buckets.
Expect small numeric gaps (commonly 3–10 points) between the three when auditing the same URL. The gap is weighting philosophy, not a bug — all three tools check the same underlying signals.
If scores diverge by more than 15 points, run the Site Analyzer against the same URL and compare individual checks — that usually surfaces a real issue one tool caught and the other weighted differently.

Batch Analyzer & Competitor Compare

Enter URLs to Compare

Context and background

Analysis Progress

Snapshots

Comparison Results

Score Comparison Chart

Detailed Interpretations

New features — learn how to use them

Learn the full methodology