Technical SEO & Migrations

What I own

• Full site crawls — Screaming Frog audits covering status codes, redirect chains, duplicate/thin content, and broken internal links.
• Indexation & coverage — Search Console analysis of indexed vs. excluded pages, crawl budget waste, and sitemap accuracy. Grew one client's clean indexed footprint from 170 to 189 pages.
• Core Web Vitals — brought a client's CWV to passing across both mobile and desktop.
• Structured data — schema implementation reviewed against Google's Rich Results requirements.
• Platform migrations — a complete URL-mapping tracker pre-migration, daily Search Console monitoring post-launch, and same-day fixes for redirect issues.

Core Web Vitals: the current thresholds

Per Google's own web.dev documentation, these are the "good" thresholds — every one has to pass at the 75th percentile of real Chrome visits over a rolling 28-day window for a URL to get an overall "good" assessment:

INP replaced First Input Delay (FID) as an official Core Web Vital in 2024, and it's a stricter, more complete measure — it scores the slowest interaction across the entire page visit, not just the first one, which makes it a genuinely harder metric to pass on JavaScript-heavy pages. I read field data (CrUX, via PageSpeed Insights or Search Console) rather than relying on lab scores alone, since a page can look fast in the lab while real users on mid-range phones and slower connections experience something worse.

Crawling, indexing, and crawl budget

Google's own crawl-budget guidance is honest that this mostly matters for large sites — crawl budget is rarely a limiting factor for sites under a few thousand URLs, but it becomes a real constraint on large e-commerce catalogs and multi-site portfolios with faceted navigation, parameter-driven duplicates, or redirect chains eating into how often Googlebot revisits the pages that actually matter. I audit crawl paths against page priority, clean up the sitemap so it lists only canonical, indexable URLs, and use Search Console's Index Coverage report as the source of truth for what's actually indexed versus excluded and why — the two statuses I chase down first are "Crawled – currently not indexed" (usually a thin- or duplicate-content signal) and "Discovered – currently not indexed" (usually a crawl-budget or internal-linking problem). Google evaluates and ranks based on mobile-first indexing, so I treat the mobile render as the one that counts, not the desktop one.

Structured data

I implement schema as JSON-LD — Google's preferred format — and validate every template against the Rich Results Test before it ships. Beyond eligibility for rich results in traditional search, well-formed, accurate schema is also one of the clearest signals AI systems use to understand and cite a page's content correctly, which matters more every quarter as AI Overviews and chat-based answers pull a growing share of informational queries. Schema is only worth what it accurately describes, though — if structured data claims something the page doesn't actually show (a price, an availability status, a review count), that's a trust problem, not a ranking boost.

AI crawlers and access control

A newer part of a technical audit in 2026: deciding, deliberately, which AI crawlers (GPTBot, ClaudeBot, PerplexityBot, Google-Extended, and others) get access via robots.txt, and confirming that whatever content those crawlers are allowed to see actually renders without requiring JavaScript execution — a lot of AI crawlers don't execute client-side JS the way Googlebot does. This is a real, current-day audit item now, not a hypothetical one.

Worked example

Related reading

For the step-by-step process, see the guide: How to Do a Technical SEO Audit.