Technical SEO checklist for 2026: 20 checks that decide whether you can rank

Content and links decide how high a page ranks. Technical SEO decides whether it can rank at all. If Google can't crawl a page, renders it without its main content, or treats it as a duplicate of another, no amount of writing will help.

These are the 20 checks we run on every site before recommending any content work. You need Google Search Console, a crawler (Screaming Frog's free version is enough for small sites) and PageSpeed Insights.

Crawling

1. robots.txt doesn't block what matters

Open yourdomain.com/robots.txt. It should block only private areas (admin, account, cart steps) and point to your sitemap. A leftover Disallow: / from a staging site is surprisingly common after launches. Remember that robots.txt stops crawling, not indexing: to keep a page out of Google, use noindex instead.

2. Important pages are reachable through links

Search engines discover pages by following links. Every page you want indexed should be linked from at least one other indexable page, ideally within three clicks of the homepage. Pages that exist only in the sitemap or behind a search form are "orphans" and tend to be crawled rarely.

3. No crawl traps

Filters, sort orders, calendars and session parameters can generate endless URL variations. Check your crawler for parameter URLs and repeated patterns, and keep them out of internal links or mark them noindex with a canonical to the main version.

4. Server responds quickly and reliably

Slow or failing responses make Google crawl less. In Search Console, open Settings → Crawl stats and check average response time and the share of server errors (5xx). Consistently over a second deserves a look at hosting and caching.

Indexing

5. Every page has the right status code

Live pages return 200, moved pages return a single 301 to their new address, deleted pages return 404 or 410. Watch for "soft 404s": empty or error pages that return 200.

6. Canonical tags point to the right URL

Each indexable page should declare itself as canonical with an absolute URL. Check that canonicals don't point to redirected, noindexed or different-language pages, and that http/https and www/non-www variants all resolve to one version.

7. noindex only where intended

Crawl the site and list every page with noindex in the meta robots tag or the X-Robots-Tag header. Thank-you pages, internal search and account pages belong there; service pages and guides don't.

8. The page indexing report is understood

In Search Console, Indexing → Pages shows why pages aren't indexed. "Crawled – currently not indexed" and "Discovered – currently not indexed" usually point to thin, duplicate or poorly linked content, not to a technical error. "Duplicate without user-selected canonical" means your canonicals need attention.

9. XML sitemap lists only indexable URLs

The sitemap should contain only canonical pages that return 200, with accurate lastmod dates. Submit it in Search Console and compare submitted versus indexed counts. A large gap is a signal to investigate.

Rendering

10. Main content is in the HTML

View the page source (not the inspector) or use the URL Inspection tool's "View crawled page". If the main text, headings and links only appear after JavaScript runs, indexing becomes slower and less reliable. Server-side rendering or pre-rendering, standard in frameworks like Next.js, solves this.

11. Resources aren't blocked

CSS and JavaScript needed to render the page must not be blocked in robots.txt. The URL Inspection tool's screenshot shows what Google actually sees.

Speed and experience

12. Core Web Vitals pass on mobile

In Search Console, Experience → Core Web Vitals shows field data from real users. Targets: LCP under 2.5 seconds, INP under 200 ms, CLS under 0.1. Fix templates rather than single pages, because a slow template affects every page that uses it.

13. Images are sized and modern

Serve images in WebP or AVIF at the size they are displayed, with width and height set to prevent layout shifts, and lazy-load those below the fold. Never lazy-load the main image at the top of the page.

14. Mobile-friendly by default

Google indexes the mobile version of your site. Text must be readable without zooming, tap targets far enough apart and no content hidden on mobile that exists on desktop.

15. HTTPS everywhere

Every page loads over HTTPS with a valid certificate, http:// redirects permanently, and no images or scripts load over plain HTTP (mixed content).

Structure and signals

16. One clear H1 and logical headings

Each page has one H1 that matches its main topic, followed by H2s and H3s in order. Headings help both readers who scan and search engines that need to understand sections.

17. Unique titles and descriptions

Every indexable page needs its own title (roughly 50–60 characters) and meta description (roughly 120–160 characters). Duplicates across many pages are a sign of template problems. Google may rewrite them, but a good one is used far more often.

18. Structured data that matches the page

Add schema.org markup that reflects what is visible: Organization on the homepage, Service on service pages, Article on guides, BreadcrumbList everywhere, FAQPage only where the questions are shown. Validate it with Google's Rich Results Test.

19. Internal links with descriptive anchors

Link related pages using words that describe the target ("technical SEO audit", not "click here"). Each service page should link to its supporting guides, and each guide back to its service. This is how you show search engines which pages matter most.

20. hreflang for every language version

If the site has several languages, each page must list all its language versions with hreflang (including itself) plus an x-default, and every version must link back. Each version needs its own canonical and the correct <html lang>. Missing return links are the most common reason Google ignores hreflang. You can declare them in the HTML head or in the XML sitemap; doing both, consistently, is safest.

Priorities

Not all issues are equal. Fix in this order:

1. Anything that blocks crawling or indexing of important pages (checks 1, 5, 6, 7, 10)
2. Duplicate and canonical problems across templates (6, 8, 17)
3. Speed and Core Web Vitals at template level (12, 13)
4. Structure, schema, internal links and hreflang (16, 18, 19, 20)

Rerun the crawl after each round and keep a record, so you can see what changed when rankings move.

Technical work goes faster when the site is built for it. Our business website checklist covers the decisions to make before a redesign. If you want an expert to run these checks and hand you a prioritised list, see our technical SEO audit or ongoing SEO service.