Technical SEO
Technical SEO : let search engines and AI crawlers read your site without obstacles
If Google can't crawl, render and index a page, that page doesn't exist for search, however good its content. We audit and fix your site's technical foundations, and work with your developers so changes are implemented correctly and stay that way.
Why it matters
Crawling, rendering, indexing: three filters before a page can rank
To appear on Google, a page goes through three stages. First, crawling: Googlebot has to discover the URL (via links or the sitemap) and be able to access it. Next, rendering: if the content depends on JavaScript, Google has to execute it to see it. Finally, indexing: Google decides whether the page adds something of its own and stores it, or discards it as a duplicate or low value.
Technical problems are rarely visible at a glance. An overly broad robots.txt rule, a misconfigured canonical or a template that generates thousands of parameter URLs can shut important pages out or waste crawling on useless ones. AI assistants' crawlers are less forgiving still: many don't execute JavaScript, so anything missing from the initial HTML may never be seen.
A simple robots.txt for an online shop: it blocks areas with no search value and declares the sitemap. Remember that robots.txt controls crawling, not indexing: to remove a page from the index you use noindex, and the page must be crawlable for Google to see it.
What we review
The areas covered by a full technical audit
Crawling and crawl budget
Robots.txt, XML sitemaps, status codes, redirect chains and orphan URLs. On large sites, how Googlebot splits its time between useful and useless pages.
Indexing
Search Console's page indexing report, noindex, duplicate content, "crawled – currently not indexed" pages and thin pages worth merging or retiring.
Rendering and JavaScript
We compare the initial HTML with the rendered page, find content, links or metadata that only appear after JavaScript runs, and recommend SSR, prerendering or static HTML.
Core Web Vitals
LCP, INP and CLS using real-user data (CrUX) and lab data. We identify concrete causes: images, fonts, third-party scripts, server response.
Architecture and internal linking
Click depth, internal link distribution, menus, breadcrumbs and pagination. Important pages should sit close to the homepage and be well linked.
Canonicals and duplicates
Trailing slashes, www, parameters, product variants. A canonical is a hint, not a directive: it has to agree with links, sitemaps and redirects.
Hreflang and international
Reciprocal annotations, correct language and region codes, x-default and consistency with canonicals. Common sources of errors on multilingual sites.
Structured data
Organization, LocalBusiness, Product, Article, BreadcrumbList and other Schema.org types, validated and consistent with visible content.
Log file analysis
Server logs show what Googlebot actually crawls, how often and where it wastes time. They also reveal the activity of AI crawlers.
Core Web Vitals
The thresholds Google uses
Google assesses each metric at the 75th percentile of real visits, separately for mobile and desktop. INP replaced FID as the responsiveness metric in March 2024.
| Metric | What it measures | Good | Needs improvement | Poor |
|---|---|---|---|---|
| LCP (Largest Contentful Paint) | How long the main element of the page takes to appear | ≤ 2.5 s | 2.5–4 s | > 4 s |
| INP (Interaction to Next Paint) | How quickly the page responds visually to interactions | ≤ 200 ms | 200–500 ms | > 500 ms |
| CLS (Cumulative Layout Shift) | How much content shifts unexpectedly while loading | ≤ 0.1 | 0.1–0.25 | > 0.25 |
Process
From audit to changes in production
A technical report that nobody implements is worthless. So the process ends when changes are live and verified.
Full crawl and data collection
We crawl the site with tools such as Screaming Frog, simulating both a browser without JavaScript and one that executes it, and cross-reference the results with Search Console, Core Web Vitals data and, where possible, server logs.
That way we see the site as Google sees it: which URLs exist, which are indexed, which get crawled and which add nothing.
Activities
Outputs
- URL inventory with indexing status
Diagnosis and prioritisation
Each issue is ranked by impact (how many pages and which pages it affects), implementation effort and risk. A problem affecting your main categories comes before a hundred minor warnings on pages with no traffic.
We explain each point on two levels: what it means for the business and exactly what needs to change in the code or the CMS.
Activities
Outputs
- Prioritised report
- Development-ready tickets
Implementation
If we have access, we make changes directly in the CMS. If there's a development team, we work with them: clear specifications, acceptance criteria and review in a staging environment before release.
For risky changes (robots.txt rules, bulk redirects, template changes) we prepare a rollback plan.
Activities
Outputs
- Changes deployed and documented
Verification and monitoring
After each release we check the change works as intended: a fresh crawl, URL inspection in Search Console and tracking of indexing and performance over the following weeks.
On ongoing projects we set up alerts to catch regressions: a noindex slipping into production, a rise in 5xx errors or a drop in indexed pages.
Activities
Outputs
- Verification report
- Regression alerting
Special cases
Technical problems that need judgement, not just tools
JavaScript SEO
Apps built with React, Vue or Angular can rank well, but only if content and links arrive in the HTML or are rendered reliably. Google renders JavaScript in a second phase; many AI crawlers don't render it at all. We recommend server-side rendering or static generation for pages that need to rank.
Faceted navigation
Filters on a shop or portal (size, colour, price, area) can generate millions of URL combinations. We decide which combinations have search demand and deserve an indexable page, and which are handled without creating crawlable URLs or kept out of crawling.
Learn moreLarge sites and migrations
On sites with tens of thousands of URLs, crawl budget and architecture become decisive. And in a domain, CMS or structure change, a technical mistake can cost months of traffic.
Learn moreFAQ
About technical SEO
How often should I run a technical audit?
A full audit at the start, and always before a redesign or migration. After that, on active projects, continuous monitoring with periodic reviews is more useful than large audits every so often.
Are Core Web Vitals a ranking factor?
They're part of the page experience signals Google takes into account, but they don't make up for weaker content. Their direct effect on rankings is usually moderate; their effect on conversion and user experience is clear. Our Core Web Vitals guide goes into detail.
Do you need access to the code?
Not always. Many fixes can be made in the CMS. For template, server or performance changes, we work with your developer or, if you don't have one, through our web development service.
Does my WordPress or Shopify site have technical problems?
Most platforms work well out of the box, but themes, plugins and apps add common problems: duplicates, heavy scripts, tag or collection URLs with no value. The audit will show them.
What does log file analysis add that Search Console doesn't?
Search Console summarises; logs show every actual request from Googlebot and other crawlers, including AI crawlers. They're especially useful on large sites to see which sections are under-crawled or wasted.
Does technical SEO affect visibility in ChatGPT or Perplexity?
Yes. If their crawlers can't access your site, or content only appears after JavaScript runs, being cited becomes unlikely. We address this as part of generative engine optimisation.
Related
You might also be interested in
SEO migrations
SEO migrations for domain, CMS, redesign or URL structure changes: URL inventory, 301 redirect mapping, staging tests and post-launch monitoring.
Core Web Vitals explained
Core Web Vitals explained: LCP, INP and CLS thresholds, field versus lab data, the tools to measure them and how to improve each metric on your website.
Generative Engine Optimization
Generative Engine Optimization: we help ChatGPT, Gemini, Perplexity and Google's AI Overviews find, understand and cite your website as a source.
Is your site technically sound?
A technical audit tells you what is holding your visibility back and in what order to fix it.