Technical SEO: a step-by-step method and prompt

Technical SEO is the work that makes sure search engines can crawl, render and index the pages you care about. It covers robots rules, sitemaps, status codes, redirects, canonical tags, rendering, speed and structured data. If this layer is broken, no amount of good content will rank, because the pages never make it into the index.

What it is

Think of technical SEO as plumbing. A crawler arrives, reads robots.txt, follows links and sitemap entries, fetches each URL, renders it, decides which version is canonical, and stores it. Every step can fail quietly. A stray noindex tag, a canonical pointing to the wrong URL, or a redirect loop can remove a page from search without any visible error on the site itself.

Google Search Central documents how this pipeline works. The goal of a technical review is to walk each stage and confirm nothing leaks.

When to use it

• Before and after a site migration, redesign or domain change.
• When Search Console reports indexing errors or a sudden drop in indexed pages.
• When you switch to a JavaScript framework or change rendering mode.
• When Core Web Vitals in Search Console show "Poor" URLs.
• As part of a quarterly SEO audit.

Step-by-step method

1. Check robots.txt. Fetch /robots.txt and confirm it does not disallow important paths or assets like CSS and JS. Make sure it references your sitemap. Our robots.txt generator gives you a clean starting file.
2. Validate the sitemap. Confirm it only lists canonical, indexable URLs that return 200. Remove redirected, noindexed and parameter URLs. Resubmit it in Search Console.
3. Audit status codes and redirects. Crawl the site. Fix internal links that point to 3xx or 4xx URLs so they point straight at the final 200 URL. Collapse redirect chains to a single hop.
4. Review canonical and indexing directives. Each indexable page should have a self-referencing canonical and no noindex. Use the URL Inspection tool in Search Console on a sample of key pages to see what Google selected as canonical.
5. Test rendering. In URL Inspection, view the rendered HTML for a key page. Confirm the main content, links and meta tags appear without user interaction. If they are missing, move that content to server rendering.
6. Measure speed per template. Run home, article and product templates through PageSpeed Insights. Fix the largest contributors first: oversized hero images, render-blocking scripts and layout shifts from ads or embeds.
7. Add and validate structured data. Add Organization, Article, Product or FAQ schema where it fits. Validate with the Rich Results Test.
8. Monitor. Recheck the Search Console Pages and Core Web Vitals reports two to four weeks after fixes, since reprocessing takes time.

Ready-to-use prompt

You are a technical SEO engineer reviewing [DOMAIN], built on [FRAMEWORK OR CMS].

Here is the data:
- robots.txt: [PASTE]
- Sitemap sample (first 50 URLs): [PASTE]
- Crawl export with url, status, redirect_target, canonical, meta_robots: [PASTE]
- Rendered HTML of one key page (optional): [PASTE]

Tasks:
1. List robots.txt rules that block important paths or assets.
2. List sitemap URLs that are not 200, not canonical, or noindexed.
3. List redirect chains longer than one hop and internal links that point to redirects.
4. List pages whose canonical points somewhere other than themselves, and say whether that looks intended.
5. If rendered HTML is provided, say whether the main content and links are present without JavaScript interaction.

For each issue, give the exact fix as a code or config change where possible. Output as a markdown table: Issue, Affected URLs (max 5 examples), Fix, Who (dev or content). Do not guess beyond the data.

Checklist

• robots.txt allows key paths and assets and links the sitemap
• Sitemap contains only 200, canonical, indexable URLs
• No internal links pointing to redirects or 404s
• Redirect chains reduced to one hop
• Self-referencing canonicals on indexable pages
• No accidental noindex on important templates
• Main content present in rendered HTML
• LCP, INP and CLS checked per template
• Structured data validated
• Recheck scheduled for two to four weeks later

How Thig automates it

The Coding agent turns technical findings into pull requests on your GitHub repo: meta tags, canonical fixes, sitemap and robots changes, schema and image optimizations. You review the diff and merge. The SEO agent supplies the findings from its crawl and your Search Console data.