Most sites don’t have a content problem. They have a plumbing problem. The pages exist, the products are live, the articles are published — and search engines still crawl the place like a maze with half the doors welded shut. A technical SEO audit is how you find those welded doors and open them.
This checklist covers the full stack in the order that matters: crawlability, indexation, site architecture, page speed, mobile experience, redirects, duplicate URLs, structured data, JavaScript rendering, and the monitoring habits that stop everything from quietly breaking again in three months. It’s built for people who want to run the audit themselves, or at least walk into a conversation with a specialist and know exactly what they should be checking.
The following sections break the audit into manageable blocks — crawl access first, then what’s actually indexed, then the structural and performance issues that hold good pages back. Work through them in sequence and you’ll have a clear, prioritized action list instead of a 60-tab spreadsheet of panic.
- Crawl access and what blocks it
- Indexation checks and index bloat
- Internal linking and architecture depth
- Core Web Vitals and page speed
- Mobile-first rendering
- Redirects, status codes, and security
- Duplicate URLs and parameter handling
- Structured data and rich results
- JavaScript rendering issues
- Baselines, alerts, and repeat audits
What a Technical SEO Audit Actually Is
A technical audit isn’t a keyword strategy and it isn’t a content review. It’s a diagnostic of whether search engines can find, fetch, understand, and trust your pages — and whether the experience is good enough once someone lands on them.
That means two things matter equally: what the crawler sees and what the human gets. A page that loads in four seconds with a broken canonical tag fails on both counts.
1. Crawlability: Can Search Engines Get In?
Start here. Everything downstream is meaningless if crawlers are blocked from the pages you want ranked.
Check your robots directives
Pull the root robots file and confirm it isn’t disallowing key directories. Then sweep for pages carrying a noindex directive — those are the classic silent killers, often added during a staging launch and never removed.
Verify your XML sitemap
The sitemap should contain only canonical, indexable, 200-status URLs. If it lists redirects, 404s, or pages tagged noindex, you’re sending mixed signals and wasting crawl attention. Check the last-modified dates too; a sitemap that never updates looks stale.
Confirm internal links as crawl paths
Crawlers follow links, not wishes. If important pages are only reachable through a form, a JavaScript menu, or a single buried anchor, they’re effectively hidden. Every page you care about should be a few clicks from the homepage via plain HTML anchors.
2. Indexation: What’s Actually Showing Up
Compare three numbers: URLs discovered, URLs crawled, URLs indexed. Big gaps tell a story.
- Too few indexed: look at canonical tags, noindex directives, thin content, or slow server responses.
- Too many indexed: look at filter parameters, paginated archives, search result pages, and printable or session-based duplicates.
- Wrong pages indexed: something is stealing the signal — usually a duplicate variant outranking the canonical original.
Also check for index bloat: hundreds of low-value URLs that dilute crawl attention. Pruning them is often the fastest ranking win available.
3. Site Architecture and Internal Linking Depth
Flat beats deep. If a product page takes six clicks from the homepage, it’s getting less crawl attention and less internal link authority than a page three levels closer.
Map click depth across your key templates. Flag orphan pages with no internal links pointing at them — they’re invisible until something external links or a sitemap saves them. Then review anchor text: generic phrases like “click here” waste an opportunity, while descriptive anchors reinforce what the target page is about.
Breadcrumbs, related-content modules, and hub pages all flatten architecture cheaply. Look for dead ends where a visitor (or crawler) can’t go anywhere useful.
4. Page Speed and Core Web Vitals
Speed is measured two ways: lab data from a synthetic test and field data from real visitors. Trust field data more — it reflects actual devices and networks.
The three numbers that matter
- Largest Contentful Paint (LCP): how fast the main content appears. Aim under 2.5 seconds.
- Interaction to Next Paint (INP): how quickly the page responds to taps and clicks. Aim under 200 milliseconds.
- Cumulative Layout Shift (CLS): how much the layout jumps around. Aim under 0.1.
Common culprits are oversized unoptimized images, render-blocking scripts in the head, no caching layer, and web fonts loading without reserved space. Fix those four and most sites move substantially.
5. Mobile-First Reality Check
Search engines index the mobile version of your pages. So test on a mid-range phone on a throttled connection, not a desktop window resized to 400 pixels.
Look for content hidden on mobile that exists on desktop, tap targets too close together, intrusive interstitials blocking the main content, and horizontal scrolling. If mobile and desktop serve different markup, confirm the important content and structured data match.
6. Redirects, Status Codes, and Security
Crawl a sample of URLs and sort by status code.
- 301 chains: collapse A → B → C into A → C. Each hop costs time and dilutes signals.
- 302 loops: these stall crawlers and confuse users.
- Soft 404s: pages returning a 200 but displaying “not found” content. Fix the status code.
- Mixed content: any insecure asset on an HTTPS page triggers browser warnings and erodes trust.
Confirm the HTTPS version is the canonical one, that HTTP redirects cleanly to it, and that internal links point to the secure URL directly rather than bouncing through a redirect.
7. Duplicate Content and URL Bloat
The same page reachable at multiple URLs splits its own ranking signal. Audit for:
- Trailing slash inconsistencies
- Uppercase versus lowercase paths
- Parameter variations from filters, sorting, and tracking
- WWW versus non-WWW duplicates
- Paginated sequences that repeat the same content
Pick one canonical format, enforce it with redirects, and use canonical tags where redirects aren’t practical.
8. Structured Data and Rich Results
Validate your markup with the official testing tools and watch for required properties that have gone missing after a template change. Common issues: review markup without visible reviews, article markup with the wrong date format, and FAQ blocks duplicated across dozens of pages.
Structured data doesn’t directly boost rankings, but it changes how your result appears — which changes click-through rate. Worth the ten minutes.
9. JavaScript Rendering
If your site is built on a client-side framework, compare the raw HTML response with the fully rendered DOM. If key content, links, or meta tags only appear after script execution, you’re relying on the crawler to render — and rendering is slower, costlier, and less reliable.
The fix is usually server-side rendering or pre-rendering for critical templates. At minimum, make sure titles, meta descriptions, canonical tags, and primary content are in the initial HTML.
10. Baseline, Alert, Repeat
An audit is a snapshot; sites drift. Record your baseline: indexed URL count, average response time, Core Web Vitals scores, and error counts. Set alerts for spikes in 5xx errors, drops in indexed pages, and sudden redirect changes.
Then re-run the checklist quarterly, or immediately after any migration, template redesign, or platform change.
The Order to Fix Things In
Don’t chase a perfect score. Work in this sequence:
- Unblock crawling and fix accidental noindex tags
- Clean up redirects, status codes, and duplicate URLs
- Add internal links to orphaned and deep pages
- Compress images, cache aggressively, defer non-critical scripts
- Fix mobile layout and tap-target issues
- Repair broken structured data
- Address JavaScript rendering gaps
- Prune low-value pages and tighten architecture
Items one through three typically deliver more movement in a week than items four through eight do in a month.
Where to Go From Here
Technical SEO isn’t glamorous. It’s logs, status codes, and a lot of clicking through crawl reports. But it’s also the highest-leverage work on most sites — because a page that can’t be crawled, can’t be indexed, and loads slowly is a page that never gets a fair shot, no matter how good the writing is.
Run this checklist once, fix the top three blocks, and you’ll have a cleaner site than most competitors. Run it on a schedule and you’ll keep it that way.
There’s plenty more where this came from — keep exploring the site for deeper dives on performance tuning, analytics setup, and the tools that make all of this faster to manage.