Appearance
Technical SEO Audit
Overview
A comprehensive review of a website's technical components to identify issues, prioritize fixes, and create actionable recommendations. Based on Gerry White's methodology from thousands of audits.
Common Mistake
Don't just list problems ("3,500 title tags too long"). Explain what they mean and prioritize by impact.
Prerequisites
- [ ] Google Search Console access
- [ ] Screaming Frog license
- [ ] Chrome DevTools familiarity
- [ ] Bing Webmaster Tools access
- [ ] Google Merchant Center access (if ecommerce)
- [ ] Cloudflare/CDN access (if applicable)
Audit Process
Procedure
Step 1: Client Discovery Chat
Before touching any tools, understand the client's context:
Questions to ask:
- What are your primary business goals?
- What can and cannot be changed? (APIs, external services)
- What's the history of this domain?
- What are your key KPIs?
- Any known tech debt areas?
Pro Tip
Find out what they care about BEFORE doing the audit. Otherwise you might spend time on areas they don't prioritize.
Use Sistrix to check domain history and technology changes over time.
Check CrUX data at crux.fyi with google.com for Core Web Vitals history.
Step 2: Manual Browse (1 Hour)
Spend approximately one hour literally browsing the website:
While browsing:
- Take extensive notes (ensure they're readable later!)
- Test on mobile device or emulator
- Have Chrome DevTools console open
Check DevTools for:
- Broken links (404 errors)
- Broken images
- Console errors
- Oversized assets
- Protocol issues (HTTP/2 vs HTTP/3)
Add extra columns in Network tab:
- Right-click column headers
- Add: Protocol, Content-Type, etc.
Use bookmarklet to extract all links:
javascript
javascript:(function(){var links=document.querySelectorAll('a');var data=[];links.forEach(function(a){data.push(a.href);});console.table(data);})();Check for hack content: Run site:example.com and look for suspicious pages (OnlyFans, pharma, etc.)
Step 3: Google Search Console Review
Focus on indexing issues, not just traffic:
Priority checks:
- Soft 404 pages - Pages returning 200 but treated as errors
- Duplicate without canonical - Common on login pages with parameters
- Crawled, not indexed - Feed URLs, thin content
- Crawl stats - Which hosts is Google hammering?
Example issues:
- "Oh no, that brand is out of stock" pages indexed (ASOS example)
- Login pages with parameters creating thousands of duplicates
Step 4: Configure Screaming Frog
Based on manual findings, configure BEFORE crawling:
Remove known issues:
- Parameters that cause duplicates
- Crawl traps (infinite pagination)
- Feed URLs, login pages
Add custom extraction:
- Page Speed Insights data
- JSON-LD/Schema
- Custom XPath for specific elements
Example XPath for specific text:
xpath
//div[contains(text(),'out of stock')]Step 5: Template Analysis
Key Insight
Most websites have only 3-4 page templates. Fix templates at scale.
For each template, analyze:
- Title tag patterns (use pivot tables to visualize distribution)
- Meta description coverage
- Canonical implementation
- Internal link structure
- Mobile rendering
- Load performance
Product pages - check states:
- In stock
- Out of stock
- No products found
- Pagination
- Discontinued
Step 6: Additional Checks
Bing Webmaster Tools:
- Sync from GSC (one click)
- Check XML sitemaps submitted
- 5-10% extra traffic opportunity
Google Merchant Center:
- Check product feed approval rates
- Red lines = investigation needed
- Huge traffic source often neglected
Cloudflare/CDN:
- Performance optimizations available
- Many checkboxes that improve speed
- Check HTTP headers for configuration
Other files to check:
manifest.json- PWA configuration/.well-known/assetlinks.json- Password sharingllms.txt- LLM configuration (emerging)
Step 7: Report Presentation
Use inverted pyramid writing:
- Most important finding first
- Supporting details
- Background/educational content at bottom
Format:
- Slides + Google Sheets (not long documents)
- Dynamic sheets that check live status
- Link to tickets for implementation
Always include:
- What you checked (even if fine)
- Prioritization with business alignment
- Screenshots of issues (but NOT raw tool exports)
Verification Checklist
- [ ] Client goals understood
- [ ] Manual browse completed
- [ ] GSC indexing issues reviewed
- [ ] Screaming Frog configured and crawled
- [ ] Templates analyzed
- [ ] Bing Webmaster Tools checked
- [ ] Merchant Center reviewed (if applicable)
- [ ] Report prioritized by impact
What to Avoid
| Don't | Do Instead |
|---|---|
| Screenshot raw tool output | Explain what it means |
| List 4,300 broken links | Identify root cause (e.g., footer link) |
| Obsess over page speed scores | Prioritize conversion-impacting issues |
| Tell devs how to code | Describe the problem, let them solve |
| Use "Site Health Score" | Focus on actual business impact |
Troubleshooting
Client doesn't have GSC access
Request access immediately. Consider using Bing as alternative data source.
Crawl taking forever
Remove unnecessary areas, check for crawl traps, reduce crawl rate if needed.
Can't identify page templates
Export unique URL patterns from Screaming Frog, group by URL structure.