SEO + GEO · Technical SEO

Technical SEO · Core Web Vitals, crawl, indexation and rendering so your content actually ranks

Technical SEO is the invisible foundation that makes content rank: Core Web Vitals in Google’s 75th percentile, optimized crawl budget, SSR rendering solved, validated @graph schema, coherent hreflang and canonicals. Without that foundation, neither the best content nor the best link building moves the needle.

  • Core Web Vitals in the green (LCP < 2.5s, INP < 200ms, CLS < 0.1) at Google’s real 75th percentile
  • Optimized crawl budget: prioritized crawlable URLs, parameters and facets under control
  • Time to indexation reduced to days, not weeks, on newly published pages
  • 0 critical errors in Search Console: coverage, mobile usability and schema in the green
Request a technical audit
See methodology

Technical SEO is the discipline that optimizes web infrastructure —performance (Core Web Vitals), crawlability, indexability, rendering, URL architecture, hreflang, canonicals, robots, sitemap and structured data— so search engines can discover, render, understand and index every page correctly. Unlike content SEO or link building, technical SEO is not about writing better or earning links: it’s about making sure the machine can read what you already have. It’s a precondition, not a substitute. A technically solid site is the one that ranks on Google and gets cited by ChatGPT, Perplexity, Gemini and AI Overviews.

What technical SEO covers

10 audit areas we work on every project

Complete technical SEO stack applied to B2B environments: WordPress, Next.js, Shopify, headless and complex ecommerce. Execution order: top first.

  1. 01

    Crawlability

    Bot access to every important URL: robots.txt, meta robots, X-Robots-Tag HTTP headers, blocks via authentication or firewall. A single misplaced Disallow wipes out entire sections.

  2. 02

    Indexation

    Which URLs should (and which URLs should not) be in Google’s index. Coverage review in Search Console, correct canonicals, duplicates, thin content and zombie URLs purged.

  3. 03

    Rendering

    How Google (and the LLMs) processes your JavaScript content: SSR, SSG, ISR or CSR. We verify that the rendered HTML matches the HTML the bot sees, with no partial hydration or critical post-JS content.

  4. 04

    Core Web Vitals

    LCP, INP and CLS measured in the field (CrUX), not in the lab. Goal: 75th percentile in the green on mobile and desktop. Explicit performance budget per key page.

  5. 05

    Schema · JSON-LD

    Structured data as an enriched @graph: Organization, WebPage, Article, Service, FAQPage, BreadcrumbList, Product, Review. Validated with Rich Results and Schema.org validator.

  6. 06

    Hreflang · international

    Multi-language and multi-country sites: bidirectional hreflang, x-default, sitemap with annotations and coherence with canonicals. A single mistake here sinks rankings in your main market.

  7. 07

    Canonicals

    One canonical per URL, absolute, coherent with hreflang, sitemap and internal links. No self-loops, no chains, no conflict with tracking parameters.

  8. 08

    Robots · sitemap

    Clean robots.txt with explicit access for Googlebot, Bingbot, GPTBot, PerplexityBot, Google-Extended and ClaudeBot. Segmented XML sitemap with real lastmod and no 3xx/4xx URLs.

  9. 09

    URL architecture

    Folder structure, slugs in the right language, faceted parameters, pagination, 301 redirects with no chains and no self-loops. Every URL with a clear indexation purpose.

  10. 10

    SSR · CSR · hydration

    Technical decision per template: what is SSR, what is SSG, what is CSR. Pre-JS vs post-JS HTML measurement so the bot sees the same content as the user.

Core Web Vitals

INP, LCP and CLS: the 3 metrics Google ranks on

Google has used Core Web Vitals as a real ranking signal since 2021. INP replaced FID in March 2024. The good threshold is measured at the 75th percentile of real users (CrUX field data), not in Lighthouse.

MetricWhat it measuresGood (p75)Needs improvementPoor
LCP · Largest Contentful PaintTime to paint the main visible element (hero, featured image, H1)< 2.5 s2.5 – 4.0 s> 4.0 s
INP · Interaction to Next PaintLatency of the user’s worst interaction with the page (click, tap, keyboard)< 200 ms200 – 500 ms> 500 ms
CLS · Cumulative Layout ShiftVisual stability: how much elements move during load< 0.10.1 – 0.25> 0.25
TTFB · Time to First ByteServer time to the first byte (diagnostic, not a direct ranking signal)< 800 ms800 – 1800 ms> 1800 ms

Levers that move CWV in production: preload the LCP image with fetchpriority="high", inline critical CSS, always declare width/height, defer non-critical JS, avoid heavy hydration in the hero, fonts with font-display: swap and preload, and watch third-party scripts (analytics, chat, ads) that trigger INP.

Crawl + indexation

Crawl budget, JS rendering, robots.txt and sitemap

On large sites (>10k URLs, ecommerce, media) crawl budget is a finite resource. On small sites what matters is correct rendering and canonical-sitemap-hreflang coherence.

  1. 01

    Crawl budget and logs

    Server log analysis to see which URLs Googlebot crawls, how often and what response it gets. Detects traps: infinite facets, calendars, tracking parameters, sessions in URLs.

  2. 02

    JavaScript rendering

    Googlebot processes JS but with delay (two-wave indexing). SSR or SSG solves the problem on the first pass. Pure CSR is only acceptable in private or non-indexable areas.

  3. 03

    Robots.txt and AI bots

    Explicit access for Googlebot, Bingbot and relevant AI crawlers: GPTBot, PerplexityBot, Google-Extended, ClaudeBot, OAI-SearchBot. Blocking by omission = disappearing from generative results.

  4. 04

    Segmented XML sitemap

    One sitemap per content type (posts, pages, products, categories), master index, real lastmod, no 3xx/4xx/canonicalized URLs, registered in Search Console and Bing Webmaster Tools.

  5. 05

    Coverage in Search Console

    Weekly review of the pages report: indexed, excluded, discovered not indexed, crawled not indexed, 5xx errors, redirect loops. Each anomaly with a technical ticket.

Web architecture

URLs, hreflang and canonicals coherent with each other

The URL + canonical + hreflang triangle must be consistent. An error in one of the three vertices breaks the other two and drags down entire rankings.

  1. 01

    URL structure

    Folder hierarchy that reflects the real business taxonomy. Slugs in the market language, no stop-words (the/a/an), lowercase, hyphen-separated, no numeric IDs or session parameters.

  2. 02

    Bidirectional hreflang

    Each language version declares all the others and itself. x-default for the fallback. Coherent with <link rel="canonical"> and the sitemap. Validated with dedicated tools (Merkle, Sistrix).

  3. 03

    Canonicals with no loops

    One single absolute canonical per URL. No chains, no self-loops, no conflict with hreflang or with UTM/GCLID tracking parameters. Consistent with internal linking.

  4. 04

    Governed 301 redirects

    3-tier policy: .htaccess only for infrastructure (HTTPS, domain), redirects plugin for slug migrations, functions.php only for programmatic rules with documented reasoning.

Schema · JSON-LD

Structured data Google and LLMs use as ground truth

JSON-LD schema is not an extra: it’s the syntactic map engines use to understand without ambiguity. We deploy it as a @graph with all entities linked by @id.

  • Organization

    Organization

    Company entity with sameAs to LinkedIn, Crunchbase, GMB, Wikidata. With aggregateRating if there are verifiable reviews.

  • WebPage

    WebPage + speakable

    Correct type per template (WebPage, AboutPage, ContactPage, CollectionPage, ItemPage). SpeakableSpecification pointing to H1, TL;DR and FAQ.

  • Article

    Article + Author Person

    For blog and resources. With author as Person, datePublished, dateModified, wordCount, about linked to Wikidata entities.

  • Service

    Service / Product

    For service pages or product sheets. With structured provider, areaServed, serviceType and offers.

  • FAQPage

    FAQPage

    Frequently asked questions marked up as Question + acceptedAnswer. Answers of 40-80 words citable by LLMs.

  • Breadcrumb

    BreadcrumbList

    Navigational path with itemListElement positioned 1-N. Reinforces architecture and appears in the SERP as a visual breadcrumb.

cronuts rule: schema is deployed via ACF (schema_jsonld field) or CMS equivalent, never hardcoded in the theme. That way each page can adjust its graph without touching code.

cronuts methodology

5 phases to audit and resolve technical SEO debt

Proven sequence on B2B accounts in professional services, SaaS and ecommerce. Each phase has an auditable deliverable.

  1. 01

    Full technical audit

    Crawl with Screaming Frog, log analysis, Lighthouse CI, Search Console, CrUX and rendering test. Deliverable: prioritized report with impact and effort per ticket.

  2. 02

    Core Web Vitals and performance

    Intervention on LCP, INP and CLS with explicit budget per template. Preload, critical CSS, defer third-party JS, image and font optimization. Real measurement in CrUX, not just Lighthouse.

  3. 03

    Crawl, indexation and rendering

    Cleanup of robots.txt, segmented sitemap, coherent canonicals, validated hreflang, governed 3-tier redirects and SSR verification on critical templates.

  4. 04

    Schema @graph and GEO

    Enriched JSON-LD deployment on key templates: Organization, WebPage, Article, Service, FAQPage, BreadcrumbList and speakable. Validated with Rich Results and Schema.org validator.

  5. 05

    Continuous monitoring

    Alerts on Search Console coverage, weekly CrUX, monthly crawl diff and schema regressions. Dashboard with technical KPIs the business actually understands.

Technology stack

Tools we use to audit and ship technical SEO

  • Crawl

    Screaming Frog SEO Spider

    Full site crawl: headers, status codes, canonicals, hreflang, schema, images, redirects, JS rendering with headless Chromium.

  • Performance

    Lighthouse CI + PageSpeed Insights

    Performance metrics in lab and field (CrUX). Integrated into the deploy pipeline so we don’t regress on Core Web Vitals.

  • Indexation

    Google Search Console

    Coverage, SERP performance, field Core Web Vitals, schema errors, sitemap, security, manual actions. Primary source of truth.

  • Schema

    Rich Results Test + Schema.org validator

    JSON-LD validation before deploy. Catches errors and warnings that break eligibility for rich snippets and AI Overviews.

  • Logs

    Log file analyzer

    Server log analysis to understand which URLs Googlebot crawls, how often and with what response. Critical on large sites.

  • Monitoring

    Notion · Looker dashboard

    Monthly deliverable with KPIs: CWV p75, coverage, indexed/excluded pages, valid schema, crawl errors, broken redirects.

Does your site meet Core Web Vitals on Google?

If the answer is “I don’t know,” it doesn’t. We measure it for you in an initial technical audit and hand you the 90-day plan to go green on LCP, INP and CLS, with coherent crawl, indexation and schema.

Email info@cronuts.digital
Book a diagnostic