The one-at-a-time audit is fine when you own one site. It falls apart the moment you have three, or a client roster of ten, or a portfolio of twenty. You open the Mega Analyzer, paste a URL, wait 20 seconds, copy the grade. Then you repeat. By URL seven you have stopped paying attention.
The Mega Batch runs the same six-bucket scoring — SEO, Schema, E-E-A-T, Voice, Mobile, Perf + AI — across up to ten URLs in parallel, then produces a single comparison matrix and a combined LLM fix prompt that covers the whole set.
What It Does and What It Skips
Mega Batch uses the same bucket logic as the Mega Analyzer but makes one deliberate tradeoff: it does a single fetch per URL instead of two. That means it scores SEO, Schema, E-E-A-T, Voice, and Perf + AI at full depth but Mobile at an approximation — it checks the viewport tag and whether zoom is allowed, not the full desktop-versus-mobile content diff. The single-URL Mega Analyzer is still the right tool for a deep mobile parity check. The batch is breadth, not depth.
What you get in return is speed. A run of ten URLs finishes in about a minute.
The Score Matrix
The matrix is the deliverable. Ten rows — one per site — sorted by overall score, with green / amber / red cells for each bucket. The rightmost column tags the weakest bucket for that site so you can scan the table and immediately see "this site's problem is schema, this site's problem is voice, this site's problem is perf."
The second tab shows a compact per-site card: hostname, title, overall score in big type, and a one-line summary of word count, schema, sameAs depth, and voice metrics. Useful for client-ready exports or for sending one row to a collaborator.
The third tab is the mega batch AI fix prompt. This is where the tool earns its keep for agency work.
The Mega Batch AI Prompt
The prompt includes:
- A full score matrix for every site scored
- Leader and laggard per bucket, with the gap size, so the LLM can use the leader's page as a rubric for the laggard's fixes
- Per-site detail: current title, meta, H1, H2 count, word count, schema types, JSON-LD errors, sameAs count, Wikidata / ORCID / rel=me status, voice metrics, aux files present
- Instructions to produce a prioritized per-site fix list — top 5 fixes per underperforming site, each with copy-paste code, file location, which bucket it moves, and verification steps
The prompt also carries two caveats. First, for any site scoring below 70 overall, it tells the LLM to recommend the user re-run the single-URL Mega Analyzer on that site for a deeper report — because the batch can't do the full mobile parity pass. Second, the deployment-layer caveat: security headers, redirects, and schema injected at build live in a CDN / SSG layer the audit can't see, so the LLM should confirm the deployment stack before writing warning-level fixes.
If your skill level is Advanced anywhere in the jwatte.com suite, the prompt drops the explanations and asks for deployable code on each fix. Beginner adds reasoning. The level is shared across tools.
Three Ways to Use It
Client-roster benchmarking. Every Monday morning, paste your client URLs and run. The matrix shows who regressed and where. Thirty seconds of scanning replaces thirty minutes of opening tabs.
Competitive analysis. Paste your URL plus nine competitors for your target query. The matrix ranks you against them and the weakest-link tags show where you're losing. Feed the prompt to Claude and ask for a prioritized catch-up plan.
Portfolio monitoring. If you run several properties under the $100 Network model — multiple sites compounding authority across a theme — the Mega Batch is the daily scoreboard for the whole network. Save a snapshot via the Content Velocity Monitor at the start of each quarter and diff at the end.
When to Prefer the Single-URL Mega Analyzer
- You only care about one URL deeply. The single-page version runs the full mobile parity fetch, the deeper Article-schema validation, and the velocity auto-snapshot.
- Any site that scores below 70 in a batch should get a follow-up single-URL run. The prompt will remind the LLM of this.
- Before a significant rewrite ships, run the single-URL Mega Analyzer on the new version — it catches mobile parity regressions the batch would miss.
How This Fits the Methodology
The Mega Batch is the automated version of the $20 Agency 90-Day Playbook (Chapter 30) measured across multiple properties. It's also the breadth-first pass the $100 Network Chapter 26 (Monitoring at Scale) builds its dashboards around — score many properties in one pass, then zoom in on the weakest. And it's the multi-site version of the competitive audit $97 Launch Chapter 35 walks through manually, parallelized with a combined LLM prompt at the end.
Honest Limits
It fetches static HTML and does not execute JavaScript. Sites that render content client-side will look thin. For those, the batch score is still useful as a "what Googlebot sees" signal.
Ten URLs per run. Increase the cap and you risk the Netlify function timing out. Batches of ten are the sweet spot for speed and coverage.
Mobile parity is approximate in the batch — viewport plus zoom-allowed only. The single-URL Mega Analyzer is where you go for full parity.
No historical tracking inside the batch itself. For trend data across snapshots, use the Content Velocity Monitor — export JSON after each batch if you want to keep a record.