Technical SEO: the ten fixes that actually move rankings
By Wendy FunnellMay 2026 operator update
Current read: technical SEO has become the operating layer for both search crawlers and AI retrieval systems. Clean indexation, structured content, canonical signals, accessibility, and performance are no longer separate checklists; they decide whether a page can be found, understood, cited, and converted.- What changed: AI Mode and AI Overviews now make source clarity visible in the search experience.
- What to fix now: crawl waste, sitemap gaps, canonical conflicts, slow templates, inaccessible UI, and thin duplicate archives.
- Current sources: Google AI Search link update, May 2026; WCAG 2.2 redesign guidance; Ahrefs AI Overview visibility research.
1. HTTPS / HTTP redirect handling
If bothhttps://yourdomain.com and http://yourdomain.com resolve, you have duplicate-content fragmentation. Same for www. vs no-www. versions. Pick one canonical, redirect the other three to it with a 301.
This is the most common “site that should rank doesn’t” cause we see. The fix is server-level — a single .htaccess rule on Apache or one Nginx directive — and instantly resolves the fragmentation.
2. Page speed: LCP, INP, and CLS
Core Web Vitals are real ranking signals, but more importantly, they correlate with conversion rate. The three metrics:- LCP (Largest Contentful Paint) — when the largest visible element loads. Target: under 2.5s.
- INP (Interaction to Next Paint) — how quickly the page responds to user input. Target: under 200ms. (Replaced FID in 2024.)
- CLS (Cumulative Layout Shift) — how much elements jump around during load. Target: under 0.1.
3. Mobile usability
Google has been mobile-first indexing since 2019. The mobile version of your site is the version that ranks. Two binary checks:- Mobile-Friendly Test: does the site pass? If not, the framework or theme has a problem worth fixing.
- Tap target spacing: can a user with average thumb size tap any link without accidentally tapping a neighbor? If not, fix the spacing.
4. Schema markup on appropriate page types
Schema is a labeling system that tells Google what each page is. Done correctly, it makes pages eligible for rich results in the SERP — FAQ, Article, LocalBusiness, Product, Review. The schema that matters per page type: – Service pages:Service + LocalBusiness (if local)
– Blog posts: Article with author, datePublished, image
– Product pages: Product with Offer and AggregateRating if you have reviews
– FAQ pages: FAQPage schema; eligible for FAQ rich results
– Site-wide: Organization schema in the footer
The mistake is adding schema that doesn’t apply or whose required fields aren’t filled. Google’s Rich Results Test tells you what’s working; use it.
5. XML sitemap and robots.txt
The XML sitemap should list every page you want indexed. The robots.txt should disallow every page you don’t. Common problems: – Sitemap auto-generated by the CMS includes 404’d pages, admin URLs, and thin tag pages – robots.txt blocks pages that should be crawled (a common WordPress accident) – Sitemap submitted to Search Console but never updated after content additions Run a clean sitemap check once a month. Make sure the only URLs in the sitemap are the ones you want to rank.6. Internal linking depth
The strongest pages on your site should be no more than 3 clicks from the homepage. Pages buried deeper than that get less crawl budget and rank lower for the same content. The fix is two-fold: – Make sure top commercial pages link to and from related content. Service pages should link to relevant blog posts; blog posts should link to relevant service pages. – Use breadcrumbs on every page. They create implicit internal links and improve user navigation. The “orphan pages” report in Screaming Frog or Sitebulb tells you which pages have no internal links pointing to them. These are the pages that need to be wired into the site.7. Image optimization
Three components: – File size: WebP format, properly sized for the maximum display dimensions, compressed – Alt text: descriptive of what the image shows, naturally including target terms; not stuffed – Lazy loading: images below the fold load only as the user scrolls A site with 50 unoptimized 4MB images loads slowly enough to lose rankings. The fix is mechanical — most images can be 80–90% smaller with no visible quality loss.8. Title tags and meta descriptions
Title tag rules: – Under 60 characters – Primary keyword near the front – Brand name at the end (if there’s space) – Unique per page (no duplicates across the site) Meta description rules: – 150–160 characters – Genuine summary of the page (not keyword-stuffed) – Includes a clear action or value statement This is the lowest-effort high-impact technical SEO work. Most sites have 30–50 pages with thin or duplicate title tags. Fixing them is an afternoon and produces visible CTR improvements within a few weeks.9. Hreflang for multi-language sites
If you publish content in multiple languages or for multiple regions, hreflang tags tell Google which version to show which user. Without hreflang, your Spanish and English versions can be treated as duplicates of each other, suppressing both. Hreflang is harder to get right than it looks. Every page must list every alternate version; every code must be valid; URLs must reciprocate. If you don’t have a developer who can implement it carefully, don’t implement it half-correctly — Google reverts to treating each version as duplicate.10. Crawl error monitoring
Search Console’s Coverage / Pages report lists every URL Google has issues with. Read it monthly. Specific statuses to act on: – Server error (5xx): server has a problem at the time Google crawls; investigate – Soft 404: page returns 200 but Google thinks it’s an error; usually thin content – Crawled, not indexed: Google saw it and decided it wasn’t worth indexing; improve the page or noindex it deliberately – Discovered, not indexed: Google found a link but hasn’t crawled; usually crawl budget or internal linking issue Acting on these reports is what separates sites that compound their technical health from sites that quietly accumulate cruft.What’s not on this list (and why)
A lot of technical SEO best practices didn’t make this list because they don’t actually move rankings:- AMP — Google deprecated the prioritization
- Rel=”next”/Rel=”prev” pagination — Google deprecated these as ranking signals in 2019
- Excessive structured data — only the relevant schema types matter; the others are noise
- Keyword density in URLs — long ago stopped being a meaningful signal
- W3C validation — clean HTML is nice, doesn’t move rankings
Operator summary
- Technical SEO matters when it removes crawl, indexing, rendering, schema, speed, or internal-link barriers.
- Fix issues that change how Google can discover, interpret, or trust the page before chasing cosmetic scores.
- AI/search signal: valid schema, clean canonicals, and entity clarity improve rich result and generated answer eligibility.
Related ZINC guides
- Google Search Console guide
- Duplicate content and canonical signals
- Shopify SEO problems
- ZINC SEO services
ZINC Digital builds organic search programs for service businesses, mid-market e-commerce, and local operators in Miami and Panama City. We start every engagement with an audit, then move into a monthly retainer with weekly working sessions and monthly performance reviews — tied to revenue, not sessions.
