Crawl, indexation, rendering, Core Web Vitals, structured data, internal linking and server logs, examined by developers and delivered as a prioritised list of tickets with the fix already written.
Rankings have slipped after a redesign or migration. Search Console shows thousands of "crawled, not indexed" pages. The marketing team suspects the developers; the developers suspect the content. An audit replaces suspicion with evidence.
After a replatform or redesign
New URLs, new framework, new templates. Redirect gaps, rendering changes and lost markup are the usual culprits, and they are all measurable.
Large or templated sites
E-commerce, marketplaces, listings, documentation. Faceted navigation and parameter URLs waste crawl budget and dilute signals. See programmatic SEO.
Before investing in content
New pages on a site that cannot be crawled or rendered properly will not perform. Fix the foundations first; it is cheaper.
Where pages go missing
Each stage loses pages for a different reason
Pages drop out at every stage between being linked and being eligible to rank or be cited. The audit quantifies each drop from your own logs and index coverage data, then names the cause.
Scope
Seven areas, one checklist
Crawl
Full crawl with a desktop and mobile user agent. Status codes, redirect chains, robots.txt, sitemap accuracy, parameter handling, pagination and crawl depth.
Indexation
Index coverage reconciled against the crawl: what is indexed that should not be, what is not indexed that should be, and why. Canonicals, noindex, hreflang and duplicates.
Rendering
Raw HTML compared with the rendered DOM for every template. Content, links and markup that only exist after JavaScript are flagged, with a server-rendering or hydration fix.
Core Web Vitals
Field data per template from the Chrome UX Report, then lab traces to find the cause. LCP, INP and CLS with fixes in your framework. Detailed on the Core Web Vitals page.
Structured data
Every template validated. Missing types, errors, markup that does not match visible content, and opportunities such as FAQ, Product or Event. See structured data.
Internal linking
Link graph analysis: orphans, pages with one inbound link, anchor text, navigation that buries money pages, and hub pages that should exist but do not.
Log analysis
Two to four weeks of server or CDN logs. Which bots visit, what they fetch, how often, what they get back. Includes AI crawlers such as GPTBot and ClaudeBot. Reveals crawl waste you cannot see any other way.
Severity
How findings are ranked
Every issue gets a severity and an effort estimate, so you can start with the cheapest high-impact fixes. We would rather give you eight tickets you will ship than eighty you will not.
Severity
Meaning
Example
Fix within
Critical
Pages cannot be crawled, rendered or indexed
Staging noindex shipped to production
Days
High
Significant waste or dilution of signals
Faceted URLs indexed in the thousands
Two weeks
Medium
Measurable but contained impact
LCP over threshold on one template
One sprint
Low
Hygiene and future-proofing
Missing BreadcrumbList markup
Backlog
Sample issue card · #014 · High
Faceted filter URLs are crawled and indexed
Evidence: 11,400 URLs with ?colour= and ?size= parameters in the crawl; 6,800 in the index; 38% of Googlebot requests in the log sample hit these URLs.
Impact: crawl budget spent on duplicates; category pages recrawled less often; diluted ranking signals.
Fix: canonical to the unfiltered category; noindex on multi-facet combinations; block parameter patterns in robots.txt; remove facet links from HTML for bots via server-side rendering of a clean nav.
Owner: front-end team · Effort: 2–3 days · Verify: log share below 5% after four weeks.
Numbers are from an illustrative example, not a named client.
Deliverables and process
What you get, and when
Access and scoping
Search Console, analytics, logs, repo. Agree templates and priorities. Day 1–2.
Crawl and logs
Full crawl, render comparison, log parsing, field performance data pulled per template. Week 1.
Analysis
Findings reconciled across sources, severity and effort assigned, fixes written for your stack. Week 2.
Report and workshop
Ticket list in your tracker, a short summary for leadership, a two-hour walkthrough with your developers. Week 3.
Fix sprint (optional)
Our developers implement the top items as pull requests, then re-measure. Two to four weeks.
Issue tickets with evidence, impact, fix, owner, effort and verification
Template-level Core Web Vitals report with traces
Structured data validation report per template
Crawl and log data exports you keep
AI readability check: raw HTML content, crawler directives, llms.txt
What access do you need for a technical SEO audit?
Read access to Google Search Console and your analytics, a sample of server or CDN logs covering at least two weeks, and ideally read access to the code repository or CMS. For JavaScript-heavy sites we also ask for a staging environment so we can test rendering fixes.
How long does the audit take and what does it cost?
Two to three weeks for sites up to a few thousand URLs. Large e-commerce or multi-country sites take four to six. We quote a fixed price after a 30-minute scoping call; the price depends on URL count, number of templates and whether log analysis is included.
Do you audit JavaScript frameworks like React and Next.js?
Yes. We build in React, Next.js, Laravel, WordPress and Node.js, so rendering, hydration and routing issues are things we have caused and fixed ourselves. Each finding comes with a fix that matches your framework.
Is the audit useful for AI search as well as Google?
Yes. AI engines fetch and parse the same HTML. If your main content needs JavaScript to appear, or your canonical tags point the wrong way, you are invisible to both. The audit includes an AI-readability check and a review of your crawler directives and llms.txt.
Can you implement the fixes?
Yes, and we prefer to. Findings are written as tickets with acceptance criteria. Your team can take them, or ours can run a fix sprint in your repository with pull requests for review.
Ready to make AI work in production?
Tell us about your systems, your data and your constraints. We reply within one business day with a straight answer and a suggested next step.