How to Run a Technical SEO Audit That Actually Fixes Rankings

A technical SEO audit is not a one-time event — it's a diagnostic protocol that should run quarterly for any site with more than 100 indexed pages. Done right, it surfaces the technical issues blocking your organic growth before they compound into traffic losses. Here's the four-phase (now five-phase) process we run for every B2B client.

Phase 1 — Crawl Analysis

Start with a full crawl using Screaming Frog, Sitebulb or Ahrefs. You're looking for: pages that return non-200 status codes, pages blocked by robots.txt that shouldn't be, redirect chains longer than two hops, pages without canonical tags or with incorrect canonicals, and pages with duplicate or near-duplicate titles and meta descriptions.

• 4xx errors: broken pages that may have backlinks pointing to them (redirect to relevant live page)
• 5xx errors: server errors that prevent crawling (escalate to dev team immediately)
• Redirect chains: A → B → C chains leak PageRank — flatten to A → C
• Blocked pages: confirm robots.txt isn't accidentally blocking important content
• Duplicate content: find and resolve with canonical tags or 301 redirects

Phase 2 — Indexability and Canonicals

Crawlability (a bot can reach the page) and indexability (Google will include it in results) are different things. Cross-reference your crawl data with Google Search Console's Coverage report. Pages marked 'Excluded' or 'Noindexed' need investigation — some exclusions are intentional (thank-you pages, gated content), but many are accidental. Check: noindex tags, canonical mismatches, hreflang errors for multilingual sites, and orphan pages (content with no internal links pointing to it).

Phase 3 — Performance and Core Web Vitals

Use PageSpeed Insights and Chrome UX Report (CrUX) data in GSC to measure real-user Core Web Vitals. Field data from CrUX reflects how actual users experience your site — it's what Google uses for ranking. Focus on your most valuable pages first: homepage, key service pages, top blog posts. Fix LCP issues (image optimization, preloading), INP issues (reduce JavaScript blocking) and CLS issues (explicit element dimensions).

Phase 4 — Schema and Structured Data

Validate all existing schema with Google's Rich Results Test. Look for: schema that errors (invalid JSON-LD), schema that's present but not eligible for rich results (correct but incomplete), and high-value pages that have no schema at all. Priority schemas for B2B sites: Organization, WebSite, Service, FAQPage on service/product pages, Article on blog posts, and BreadcrumbList on all internal pages.

Phase 5 — Content Depth and Cannibalization

Technical SEO audits often stop at performance and schema — but content cannibalization is a technical problem with a content solution. Use GSC to find cases where two or more pages are ranking for the same primary keyword. Cannibalization dilutes click-through and sends conflicting signals to Google about which page is the authoritative one.

For each cannibalization case, decide: consolidate (redirect the weaker page to the stronger one), differentiate (make each page clearly serve a different search intent), or canonicalize (designate one as the authoritative version). This phase often reveals the most actionable quick wins in a B2B site audit.

How Long a Real Audit Takes

A surface audit (phases 1–2 only, small site < 200 pages): 4–8 hours. A comprehensive audit (phases 1–4, medium site 200–1,000 pages): 2–4 days. A full technical + content audit (all 5 phases, large site 1,000+ pages): 1–2 weeks. Anyone promising a comprehensive audit in under 2 hours is running a checklist, not an audit.

What We Actually Find, Most of the Time

After running audits for dozens of B2B sites, the issues that appear in almost every audit are: redirect chains from years of site updates (fix rate: 90%+ within a sprint), orphan pages with no internal links (creates index bloat, easy to fix), missing FAQPage schema on service and pricing pages (fast to add, high ROI), and at least one case of keyword cannibalization between a pillar page and a related blog post.

The rarer-but-expensive issues are JavaScript rendering failures that make entire page sections invisible to Googlebot, and hreflang errors on multilingual sites that result in the wrong language version serving in the wrong market.

Prioritizing Fixes After an Audit

Not every issue found deserves immediate action. Prioritize by impact × effort:

Fix immediately (high impact, low effort): Broken internal links, missing canonical tags on key pages, missing FAQPage schema on service pages, orphan pages that have backlinks.

Schedule in next sprint (high impact, higher effort): JavaScript rendering issues on content pages, redirect chain flattening, Core Web Vitals failures on high-traffic pages.

Monitor but deprioritize (low impact): Thin content on low-traffic pages, minor schema warnings, redirect chains on pages with no backlinks.

A well-run audit produces a prioritized fix list, not an exhaustive problem dump. The goal is actionable clarity, not comprehensiveness for its own sake.