Site Audit: How to Find and Fix Every Technical Issue (2026)

A site audit is a systematic technical analysis of your entire website to identify and fix the issues preventing optimal crawling, indexation, and ranking. Unlike a narrow on-page review, a site audit looks at your entire architecture — how pages connect, how Googlebot navigates your site, and whether technical errors are silently draining your organic potential.

This guide shows you exactly how to run a site audit, what to look for, and how to fix the most impactful issues.

Tools You Need for a Site Audit

You need two types of tools for a complete site audit:

Crawlers (simulate Googlebot):

Screaming Frog SEO Spider — the most thorough and flexible desktop crawler. Crawls up to 500 URLs free, unlimited on the paid plan ($230/year).

Semrush Site Audit — cloud-based, scheduled crawls, integrates with rank tracking.

Ahrefs Site Audit — excellent error categorization and trend tracking over time.

Real Google data (what Google actually sees):

Google Search Console — Coverage report, Core Web Vitals, mobile usability.

Always combine both. Crawlers show what's technically present. GSC shows what Google actually experiences.

Step 1: Configure Your Crawl

Before running Screaming Frog:

1. Set crawl speed to 5-10 requests/second (don't hammer your server) 2. Enable "Crawl All Subdomains" if your site spans www and non-www or subdomains 3. Connect to GSC for richer data (File → Preferences → Google Analytics/GSC) 4. Set "Respect nofollow" to OFF if you want to discover all linked pages regardless of directives

For large sites (10,000+ pages), use crawl limits and start with the most important section first.

Step 2: Identify and Fix Crawl Errors

After the crawl, navigate to Response Codes:

4xx (Client Errors):

404 Not Found: The page doesn't exist. Check if any backlinks or internal links point here. If yes: redirect (301) to the most relevant live page. If no: low priority.

403 Forbidden: Googlebot can't access the page. Investigate server configuration or password protection.

3xx (Redirects):

301 chains: A → B → C. Screaming Frog flags "Redirect Chains" in the Issues tab. Fix: update the source to redirect directly to the final URL. Every hop in the chain loses ~15% link authority.

302 temporary redirects used where 301 is correct: Replace with 301.

5xx (Server Errors): Pages returning 500 or 503 are invisible to Google. Usually a server or CMS configuration issue. Fix immediately.

Step 3: Audit Internal Links

Navigate to Screaming Frog's Bulk Export → "All Inlinks" to see which pages link to each URL and what anchor text they use.

Issues to fix:

Internal links to 4xx pages: Wasting crawl budget and user experience. Update links to point to the correct live URL.

Internal links to 301 redirects: Not as critical as 4xx, but best practice is to update source links to point directly to the final URL.

Orphaned pages (zero internal links): Filter in Screaming Frog → Bulk Export → Pages → filter "Inlinks = 0". For every important orphaned page: add internal links from 3+ relevant pages using keyword-rich anchor text.

Pages with more internal links get more crawl budget and pass more authority. Your most important pages should have the most internal links pointing at them.

Step 4: Diagnose Duplicate Content

Duplicate content confuses Google about which page to rank and splits ranking signals. Types to fix:

HTTP/HTTPS duplicates: Your site should redirect all HTTP to HTTPS. Check: does http://yourdomain.com/page/ serve content (instead of redirecting)?

WWW vs non-WWW duplicates: www.yourdomain.com and yourdomain.com should not both serve content. One should 301-redirect to the other.

Trailing slash duplicates: /page/ and /page serving identical content. Choose one, redirect the other. Set canonical tags consistently.

Pagination duplicates: Page 2, 3, 4 of blog listings often have thin unique content. Use rel="canonical" pointing to page 1, or use rel="prev"/rel="next" pagination signals.

Canonical tag errors: Check Screaming Frog's "Canonicals" tab for pages with canonical tags pointing to a different URL, or pages with self-referencing canonicals that point to a 404.

Step 5: Review Meta Tags at Scale

Screaming Frog's Page Titles and Meta Descriptions tabs give you a spreadsheet view of all title and meta data.

Filter for:

Missing titles: Any page without a title tag. Add unique, keyword-targeted titles immediately.

Duplicate titles: Export and flag. Even if pages have unique content, duplicate titles confuse Google.

Over-length titles (> 60 characters): Will be truncated in SERPs. Edit to priority keyword + brand name within 60 chars.

Missing H1 tags: Filter in "H1" tab for "Missing." Add H1 to every page.

Multiple H1 tags: Only one H1 per page.

Step 6: Check Images

Navigate to Images tab in Screaming Frog:

Missing alt text: Every image should have descriptive alt text. Essential for accessibility and image SEO.

Oversized images (> 200KB): Large images slow LCP. Compress with WebP format. Aim for < 100KB for most images.

Images not lazily loaded: Below-fold images should use the loading="lazy" attribute.

Step 7: Validate Structured Data

Use Google's Rich Results Test or Semrush's schema validation in Site Audit.

Common schema errors:

Missing required fields in Article schema (datePublished, author)

FAQPage items with empty answers

Product schema missing offers node

Fix schema errors to qualify for rich result features in SERPs. See Schema Markup guide for the full implementation guide.

Prioritizing Site Audit Fixes

Run your site audit on a monthly basis at minimum. For high-traffic sites, weekly automated audits with email alerts (Semrush/Ahrefs) catch regressions before they affect rankings. For the broader SEO audit framework, see SEO Audit checklist.