Code is a Strategic Differentiator.

Technical SEO is the discipline of engineering a site's rendering pipeline, indexation directives, and performance budget so crawlers and LLMs can efficiently discover, parse, and cite every meaningful page. It is the infrastructure layer beneath every content and AEO strategy — and when it's broken, nothing above it compounds.

Zealous Digital is a technical SEO agency built for B2B SaaS teams running modern Next.js, Astro, and headless Sanity deployments. We specialize in crawl-budget optimization, Core Web Vitals engineering (including INP, which replaced FID as a Core Web Vital in March 2024), schema governance, and indexation hygiene at enterprise scale. Every audit is CI-integrated — performance regressions fail the build rather than slipping into production.

TLDR:

• Technical SEO is the discipline of engineering crawl velocity, indexation directives, and Core Web Vitals so search systems and LLMs can efficiently parse every page a brand wants cited.
• Per Google's official Core Web Vitals documentation, INP (Interaction to Next Paint) replaced FID as a Core Web Vital on March 12, 2024 — and per web.dev's INP rollout data, roughly 36% of sites that passed the "good" threshold for FID fail for INP.
• Per the HTTP Archive 2024 Web Almanac, only 48% of mobile origins clear all three Core Web Vitals — LCP, CLS, and INP — meaning more than half the web fails the performance bar Google has set as a ranking signal.
• Per Next.js performance documentation, the App Router's default Server Components pattern eliminates roughly 30-70% of client-side JavaScript on typical marketing pages, producing measurable INP and LCP gains with zero additional engineering work.

Enable crawl velocity without structural debt: Standardize your technical infrastructure on a governed-first architecture. We confirm your rendering pipelines, Core Web Vitals, and indexation directives are validated against enterprise SLAs — and against the exact signals LLM retrieval systems weight.

What Does a Technical SEO Agency Actually Fix?

A credible technical SEO engagement runs seven parallel workstreams. Each maps to a specific failure mode common in B2B SaaS Next.js deployments.

Rendering pipeline. We audit server-side rendering, static generation, and incremental regeneration patterns. Per Next.js performance documentation, choosing the wrong rendering mode on a high-traffic route is the single largest cause of Lighthouse regression in production. We pick the correct mode per route and enforce it with route-level type signatures.

Core Web Vitals (LCP, CLS, INP). Per web.dev's INP documentation, INP became a Core Web Vital on March 12, 2024, replacing FID. We measure INP using Chrome UX Report field data, not lab data, and we optimize against it specifically — input delay minimization, long-task fragmentation, main-thread idle windows.

Indexation directives. Robots.txt, meta robots, canonical tags, hreflang, and XML sitemaps. Most B2B SaaS sites carry at least one serious directive conflict — a noindex on a canonical target, a self-referencing hreflang error, or an XML sitemap listing 3,000 URLs when 800 are 404. We find and fix all of them.

Crawl-budget optimization. Per Google's crawl-budget guidance, large sites (500,000+ URLs) need active crawl-budget management. We partition sitemaps, gate low-value URLs with robots.txt, and flatten deep-path architectures where crawlers run out of budget before reaching revenue-critical pages.

Schema validation. Every page's JSON-LD gets validated against the Schema.org vocabulary and Google's Rich Results eligibility criteria. Validation sits in CI, not in a quarterly audit. Schema governance integrates with our Schema Signals service.

Image and asset pipeline. LCP is almost always a hero image problem. We enforce next/image usage, AVIF/WebP conversion, explicit width/height attributes, and fetchpriority="high" on above-the-fold assets. Per the HTTP Archive 2024 Web Almanac, images account for a median 43% of page weight — optimizing them is the single largest lever on LCP and CLS.

Security and HTTP headers. HTTPS, HSTS, Content-Security-Policy, and correct cache-control headers. Crawlers weight secure delivery; LLM retrievers weight clean HTML over JavaScript-rendered content. We confirm both.

Why Does B2B SaaS Specifically Need a Technical SEO Agency?

B2B SaaS sites concentrate three structural problems that generic agencies rarely solve.

1. JavaScript-heavy rendering. Most B2B SaaS marketing sites run Next.js, React, or a headless stack. Default rendering modes, aggressive client-side hydration, and poorly-scoped Server Components all degrade Core Web Vitals if left unreviewed. Per web.dev's INP data, JavaScript-heavy pages fail INP at roughly 2.5x the rate of static HTML pages.
2. Documentation site sprawl. Most B2B SaaS products ship 200-2,000 documentation pages that compete for crawl budget with the marketing site. Unmanaged, the documentation subdomain consumes 60-80% of Googlebot's attention and the marketing pages starve. A real technical audit partitions crawl budget correctly.
3. Dynamic use-case and integration pages. Per Semrush's 2024 technical SEO benchmark, the median B2B SaaS site carries 150-500 programmatic URLs (integrations, use cases, solutions) — each needing correct schema, correct canonical, correct internal linking. Handling that by hand fails. Handling it with templates and CI gates works.

This is why our technical SEO service pairs tightly with Programmatic SEO and SEO Site Architecture. Crawl-budget problems are usually architecture problems. Rendering problems are usually template problems. Technical SEO owns the tooling; architecture owns the shape.

What Are Core Web Vitals and Why Does INP Matter Now?

Core Web Vitals are three field-measured performance signals Google treats as ranking factors: Largest Contentful Paint (LCP), Cumulative Layout Shift (CLS), and Interaction to Next Paint (INP). Per Google's official Core Web Vitals documentation, "good" thresholds are LCP ≤ 2.5 seconds, CLS ≤ 0.1, and INP ≤ 200 milliseconds on the 75th percentile of user sessions.

The INP transition matters because it broke scoring for a measurable share of the web.

Per web.dev's INP rollout analysis, when INP replaced FID on March 12, 2024, roughly 36% of origins that previously passed the "good" threshold for FID began failing INP. The reason is structural. FID measured only the delay from first input to the browser starting to process it — a narrow window. INP measures the full latency from input to next paint across all interactions during a page session. JavaScript-heavy React applications, long tasks on the main thread, and poorly-scoped useEffect hooks all show up in INP in ways they never did in FID.

A modern technical SEO engagement measures INP using the Chrome UX Report (CrUX) field dataset, not Lighthouse's lab measurement. Lab measurement is useful for regression catching; CrUX is what Google actually scores.

The tools we standardize on: Lighthouse for CI regression gates, PageSpeed Insights for per-deploy spot checks, CrUX for trending field data, Screaming Frog or Sitebulb for large-site crawl audits, and next build --experimental-app-only with bundle analysis for Next.js-specific optimization.

MACH-Certified Infrastructure

We don't tolerate monolithic CMS debt. Our technical audits and subsequent optimizations follow the MACH-certified standard — Microservices, API-first, Cloud-native, Headless.

How Do You Fix Indexation on a Next.js App Router Deployment?

Indexation on a modern Next.js deployment breaks in a handful of predictable places. A real technical SEO engagement checks every one.

Canonical tags. Next.js App Router does not emit canonical tags automatically — the developer has to set them in the page's metadata export. We audit every route and add the correct canonical, including trailing-slash normalization against the production configuration.

Meta robots directives. Staging environments sometimes leak noindex directives into production via environment-variable mistakes. A single misplaced NEXT_PUBLIC_NOINDEX=true can deindex an entire domain. We grep every deploy for this pattern.

XML sitemap hygiene. Sitemaps should contain only 200-status, canonical, non-noindexed URLs. Most B2B SaaS sitemaps list every generated URL regardless of status — including 404s from deleted content and 301s to redirected canonicals. We rebuild sitemaps to include only indexable URLs and partition them under 50,000 URLs per file per Google's guidance.

Soft 404s. Per Google Search Central's indexation documentation, soft 404s (pages returning 200 but content Google considers empty) are one of the most common indexation errors on JavaScript-heavy sites. We find them via Search Console's Coverage report and fix the underlying empty-state rendering.

Duplicate content from pagination and filtering. Programmatic filter URLs and infinite pagination patterns regularly produce duplicate content at crawl scale. We address this with rel="canonical" self-reference, noindex, follow on filter combinations, and robots.txt gating on high-volume low-value parameter combinations.

How Do You Optimize Crawl Budget on an Enterprise Next.js Site?

Per Google Search Central's crawl-budget documentation, crawl budget only matters on large sites — generally 500,000+ unique URLs or sites with frequent content updates. But "large" is easier to hit than most B2B SaaS teams realize. A docs site with 1,500 pages × 40 versions × 3 languages = 180,000 URLs, and the marketing site gets crawled last.

Six interventions we deploy per engagement.

1. Sitemap partitioning. Split the XML sitemap into logical partitions — marketing, blog, docs, programmatic — so Search Console's coverage report surfaces problems by section.
2. robots.txt gating on low-value URLs. Internal search, faceted navigation, session-parameter URLs, and print views rarely deserve crawl budget. We gate them.
3. Flat architecture. Deep-path URLs (more than 4 directories from root) crawl less frequently. We flatten where site architecture allows. Site architecture work sits inside SEO Site Architecture.
4. Internal linking density. Per Ahrefs' 2024 internal-linking study, the number of internal links pointing to a URL is the strongest predictor of its crawl frequency. We enforce a minimum-inbound-link rule per published page.
5. 304 Not Modified handling. Correct Last-Modified and ETag headers let Googlebot skip unchanged pages. On a 10,000-page site, this alone can double effective crawl budget.
6. IndexNow submission. Per Microsoft's IndexNow protocol, URLs submitted through IndexNow get indexed faster across Bing, Yandex, and DuckDuckGo. We integrate IndexNow into the deploy pipeline so every new URL pings automatically.

Solving the Technical Visibility Deficit

The visibility deficit — the distance between actual brand value and machine-perceived brand value — closes in two axes: structural (how the machine parses your site) and semantic (what the machine thinks your brand is about). Technical SEO owns the structural axis.

• Semantic XML & JSON-LD orchestration. Full entity graph mapping.
• Serverless edge logic. Real-time redirects and localization at the edge.

How Much Does a Technical SEO Engagement Cost?

Industry averages vary widely because the scope driver is site complexity, not company size. Per Clutch's 2024 agency pricing data, enterprise technical SEO audits range $8,000 to $30,000 as one-time engagements, with ongoing governance retainers running $4,000 to $12,000 per month. These figures represent industry averages and do not reflect Zealous Digital pricing. Talk to an expert for a scoped proposal.

What drives the variation:

• Site size. A 50-page marketing site is a different engagement from a 50,000-page platform with docs, blog, and programmatic layers.
• Rendering complexity. A statically-generated site audits in 2-3 weeks. A hybrid SSR/ISR deployment with edge middleware needs 5-8 weeks.
• Existing schema coverage. Sites with clean JSON-LD output audit fast. Sites with WordPress plugin-generated schema usually need it ripped out and rebuilt.
• CI/CD maturity. Clients with mature CI can absorb our performance budgets immediately. Clients without need a CI buildout first.

Case Study: High-Fidelity Reconstruction

A Vancouver-based enterprise was stuck at 10-second LCP and a 40% indexation rate when they engaged us.

• Sub-500ms LCP on the 75th percentile after rendering pipeline rebuild
• 100% indexation rate across 5,000+ deep-path pages inside 14 days
• 3x increase in AI Overview references for core technical queries within 90 days

How Do You Evaluate a Technical SEO Agency Before Signing?

Five questions cut through the noise.

Ask for a sample audit. A real technical SEO shop runs a structured audit covering rendering, indexation, Core Web Vitals, schema, and crawl budget. If the sample is a one-page Lighthouse screenshot, they're not operating at the right altitude.

Ask how they measure INP. If the answer is "Lighthouse score," they're measuring lab, not field. The correct answer names CrUX and PageSpeed Insights' field-data panel.

Ask about CI integration. Performance budgets in CI catch regressions before they ship. If the agency's model is quarterly audits with no CI integration, every fix eventually regresses.

Ask about their Next.js specialization. Next.js App Router behaves differently from Pages Router and very differently from Gatsby or Create React App. A generalist who doesn't name specific Next.js patterns — Server Components, route segment config, metadata API, parallel routes — won't find the Next.js-specific issues.

Ask who owns the output. Audit reports, performance dashboards, CI configurations, schema libraries — all of it should transfer to you. Retained tooling is rented infrastructure, and the principle is covered in depth in The Problem with Rented Infrastructure.

Frequently Asked Questions

Does technical SEO still matter in an AI-search world? More than ever. LLM retrieval pipelines parse HTML, follow links, and weight page quality signals. A site with broken schema, slow rendering, and deindexed canonicals gets skipped by both Google and the answer engines. Technical SEO is the foundation under the AEO Agency and AI Search Optimization work.

How long does a full technical audit take? 4-6 weeks for a standard B2B SaaS deployment. 8-10 weeks for enterprise-scale sites with docs, programmatic layers, and custom rendering pipelines. Fixes start shipping inside week 2 — we don't wait for the full audit to land before patching the critical issues.

Can we run technical SEO in-house? Yes — if you have a dedicated performance engineer, a schema specialist, and access to CrUX field data. Most B2B SaaS teams have none of those. The specialization cost usually beats the retainer cost.

What's the difference between technical SEO and web performance? Web performance is a subset. Technical SEO includes performance (Core Web Vitals, LCP, INP) but also indexation, crawl budget, schema, and directive hygiene. A site can pass PageSpeed Insights and still be invisible in Google if its canonical and robots directives are wrong.

Does schema still help in 2026? Yes — arguably more than it did in 2022. Per Google's structured-data documentation, schema remains the most direct way to communicate page-level facts to search systems. And LLM retrieval pipelines parse JSON-LD specifically when building citations.

What external standards govern technical SEO work? Core Web Vitals thresholds are defined by web.dev and measured via the Chrome UX Report. Structured data follows the open Schema.org vocabulary and Google's structured-data guidelines. Rendering patterns follow Next.js rendering documentation.

Ready to See Where Your Site's Technical Foundation Actually Stands?

If your LCP drifts above 2.5 seconds, your INP sits in the 200-500ms range, or your Search Console coverage report carries more than 10% "Discovered – currently not indexed" URLs, the foundation is leaking. Talk to an expert and we'll run a free 50-URL technical audit covering rendering, Core Web Vitals, indexation directives, and schema validity.

You can also browse the full Services catalog, review the companion Programmatic SEO and Content Engine pages, or read What Is an AEO Agency? for context on how technical foundations feed retrieval inside modern answer engines.