A 2026 technical SEO audit covers eight areas: crawlability, site speed, JavaScript rendering, mobile, internal linking, international setup, AEO readiness, and how you prioritize the findings. The audits that actually move rankings ship a focused 20-point output prioritized by impact and effort — not a 100-point laundry list nobody acts on. Here’s the working checklist we use at KDGS, with the tools and priority framework that go with it.
This is the public-facing framework. The deeper diagnostic work — log file analysis, ecommerce-specific patterns, migration risk, framework-level rendering diagnostics — is where serious audit value sits, and I name what’s beyond this checklist in a section near the end.
Key takeaways
- A complete 2026 technical SEO audit covers eight areas. AEO readiness, JavaScript rendering for AI crawlers, and structured data alignment are the new-for-2026 additions.
- The most valuable audit output is a prioritized 20-point list, not a 100-point checklist. Impact × effort, ranked.
- Crawlability fundamentals haven’t changed — the additions are about whether AI crawlers (GPTBot, PerplexityBot, ClaudeBot, Google-Extended) can access content you want surfaced in AI search.
- Core Web Vitals in 2026 means INP (which replaced FID in 2024), LCP, and CLS. The thresholds shifted slightly; the priority hasn’t.
- AEO readiness adds answer-first formatting, structured data alignment, AI crawler access, and entity consistency to the audit scope.
The 2026 technical SEO audit checklist
A complete technical SEO audit covers eight areas:
- Crawlability and indexability — can search engines and AI crawlers find and access every page that matters?
- Site speed and Core Web Vitals — INP, LCP, CLS, and where each one breaks.
- JavaScript rendering — does your content appear in the rendered DOM for Google, Bing, and AI crawlers?
- Mobile and responsive design — mobile-first indexing is still the default.
- Internal linking and site architecture — does your URL hierarchy reflect topical organization?
- International and multilingual setups — hreflang, regional targeting, canonical signals.
- AEO readiness (new for 2026) — answer-first formatting, structured data alignment, AI crawler access, entity consistency.
- Findings prioritization — impact × effort, not a 100-point laundry list.
Each of these earns a section below, with the specific checks and tools we use.
Crawlability and indexability
The foundation of any audit. Three layers worth checking.
Search-engine crawl access. Pull your sitemap.xml and robots.txt. Confirm the sitemap is registered in Google Search Console and Bing Webmaster Tools, confirm robots.txt isn’t blocking pages you want indexed, and check for noindex tags on pages that should be public. Screaming Frog or Sitebulb gives you the full picture for free under 500 URLs. For larger sites, Botify or Lumar are the standard agency tools.
Indexation status. Search Console’s Index Coverage report tells you which pages are indexed, which aren’t, and why. The patterns that matter: crawled — not indexed (often duplicate content or thin quality), discovered — currently not indexed (typically crawl budget issues), and excluded by noindex tag (intentional or not — verify which).
AI crawler access (new for 2026). ChatGPT’s GPTBot, Perplexity’s PerplexityBot, Anthropic’s ClaudeBot, and Google-Extended all have separate crawler identities, and many sites are blocking them in robots.txt without realizing it. If you want to be discoverable in AI search, those crawlers need access. If you don’t, block them deliberately. Either decision is fine; an accidental block is not.
For a deeper look at which technical signals actually move rankings (versus which signals look important but don’t), see our piece on what actually moves rankings in a technical SEO audit.
Tools: Google Search Console Index Coverage (free), Screaming Frog (free under 500 URLs), Bing Webmaster Tools (free).
Site speed and Core Web Vitals
Three Core Web Vitals matter in 2026: INP (Interaction to Next Paint, replaced FID in 2024), LCP (Largest Contentful Paint), and CLS (Cumulative Layout Shift). Google reports them through Search Console’s Core Web Vitals report and through PageSpeed Insights for individual pages.
Targets:
- INP under 200ms (under 100ms is “good”)
- LCP under 2.5 seconds (under 1.8s is “good”)
- CLS under 0.1
The common failures we see:
- LCP killed by large hero images. Compress, use modern formats (WebP, AVIF), implement responsive <picture> elements with appropriate srcsets.
- INP killed by JavaScript blocking the main thread. Audit third-party scripts. Defer or remove anything non-critical.
- CLS killed by web fonts loading asynchronously without size reservation, or ad slots loading after the visible content.
Test in real-world conditions, not just synthetic lab tests. GSC reports field data from actual users via the Chrome User Experience Report — that’s the data Google uses for ranking signals. Lighthouse and PageSpeed Insights give you both field and lab data.
Tools: GSC Core Web Vitals report (free), PageSpeed Insights (free), WebPageTest (free, deeper analysis), Chrome DevTools (free).
JavaScript rendering
Increasingly important because of AI crawlers. Two questions every 2026 audit should answer.
Does Google see your content? Use the URL Inspection tool in Search Console to view the rendered HTML and screenshot. If critical content is missing in the rendered view but present in the live page, your site has a JavaScript rendering problem. Single-page applications and JS-heavy frameworks (React, Vue, Angular without SSR) are the common offenders.
Do AI crawlers see your content? (new for 2026) AI crawlers vary in their JavaScript execution. GPTBot executes JavaScript inconsistently. PerplexityBot is more limited. ClaudeBot’s rendering capability is undocumented. Server-side rendering or static rendering is the safest path for content you want both Google and AI engines to read reliably.
Server-side rendering (SSR), static site generation (SSG), or pre-rendering are the three solutions. Pick based on your stack and content update frequency. As a fast diagnostic: open your page in Chrome, disable JavaScript in DevTools, reload. If critical content disappears, that’s what an AI crawler with limited JS execution experiences.
Tools: GSC URL Inspection (free), Chrome DevTools (free), Screaming Frog with JavaScript rendering enabled (paid feature).
Mobile and responsive design
Mobile-first indexing has been Google’s default since 2019. The audit checks haven’t changed much, but skipping them is still common. Run GSC’s Page Experience report (Mobile Usability is now folded into this) and confirm responsive design across breakpoints. Standard test points: 320px, 375px, 768px, 1024px, 1440px.
Specifically check:
- Tap targets sized at least 48×48 CSS pixels (Google’s recommendation)
- Text legible without zooming — 16px minimum body text on mobile
- No horizontal scrolling on standard mobile viewports
- Forms work without zooming or repositioning
- Above-the-fold content loads in under three seconds on mobile
Tools: GSC Page Experience report (free), Chrome DevTools device toolbar (free), BrowserStack or LambdaTest for cross-device testing (paid).
Internal linking and site architecture
Two layers. URL hierarchy and link structure.
URL hierarchy. Do your URLs reflect a logical topical organization? Manufacturer sites should follow category → product line → product → SKU. Content sites should follow topic → cluster → article. Service businesses should follow service → sub-service → location. The structure tells search engines and AI crawlers how to relate pages to each other.
Link structure. Are your most important pages linked-to from many internal sources? Pages buried four clicks deep in navigation, or only linked from the footer, signal low importance. Pages linked from the homepage, primary navigation, and many topical clusters signal high importance.
What to audit:
- Click depth — every important page should be within three clicks of the homepage
- Orphaned pages — indexed pages with zero internal links
- Internal link distribution — are your money pages and pillar pages receiving the link equity they deserve?
- Anchor text — does the anchor describe the destination? “Click here” anchors are wasted signal
Topical authority lives in this section. If you’ve built pillar-cluster content, the internal linking structure has to reflect that organization, or the cluster doesn’t work as designed.
Tools: Screaming Frog (internal link graph), Sitebulb (visual site architecture), GSC Links report (free).
International and multilingual setups
Most KDGS readers serve US-focused or single-language audiences. If that’s you, skim this section and move on. If you operate across regions or languages, here’s what the audit covers:
- hreflang tags on every regional and language variant, pointing back to every other variant — including itself. Return links matter.
- Canonical strategy — one canonical version of each piece of content, consistent across regional variants
- Geotargeted Search Console properties for each region you serve, with appropriate country targeting
- Server location and CDN configuration matching expected user geography for speed
Common mistakes: hreflang implemented on some pages but not all, missing return links, conflicting canonical tags between regional versions, country-targeted properties that don’t match server location. Each of these breaks the international setup quietly — the site keeps working, but signals get crossed.
Tools: Screaming Frog (hreflang audit), GSC International Targeting report (free).
AEO readiness as part of the audit (new for 2026)
This is the section that didn’t exist three years ago. Five checks.
Answer-first formatting. Do your content pages lead with a direct answer in the first 60 words, or do they bury the answer below introductory throat-clearing? AI engines extract from the first 60 words of a section or don’t extract at all. This applies to every H2 section, not just the opening.
Scannable hierarchy. Are H2s phrased as questions where natural? Are claim-style sentences present that survive being lifted from context? Bullet lists and numbered steps wherever the content naturally fits them?
Structured data alignment. We audit schema for two things: presence (Organization, Article, Person, FAQPage where appropriate) and accuracy (schema describes content that’s actually on the page, dateModified reflects actual updates). The May 2026 Ahrefs study found that schema does not appear to directly drive AI citations for pages already cited by AI — but schema remains foundational infrastructure for traditional SEO and entity disambiguation. See our answer engine optimization coverage for the deeper context.
AI crawler access. Already covered in the crawlability section above. Worth re-verifying here because the decision needs to be deliberate — accidental blocks in robots.txt are a common audit finding.
Entity consistency. Does your Organization schema reference the same entity across the site? Are author Person schemas consistent across bylines? Do your sameAs links point to authoritative profiles (LinkedIn, Wikidata, Crunchbase)? Inconsistent entity signals confuse AI systems trying to resolve who you are.
Tools: Rich Results Test (free), Schema Markup Validator at validator.schema.org (free), GSC Index Coverage (for crawler access verification).
How audit scope changes by site type
The eight areas above are the universal framework. What each area looks like in practice — and which ones dominate the audit’s output — depends heavily on the site.
Service businesses. Local SEO signals carry more weight: Google Business Profile integration, NAP consistency, LocalBusiness schema, service-area pages, geographic targeting. Internal linking from service to location to sub-service is where most ranking issues live. Site speed matters, but mobile responsiveness on the booking or contact flow is where conversion impact concentrates.
Ecommerce. The audit expands significantly. Faceted navigation handling, parameter rules, Product schema implementation at scale, inventory state signaling, canonical strategy across color and size variants, pagination patterns, and search result page indexing decisions. Product Listing Pages and Product Detail Pages each have their own audit pattern. Sites running over 10,000 SKUs typically need log file analysis to surface crawl budget waste — Googlebot spending time on parameter URLs is a common, expensive finding.
Content publishers. Topical authority structure, internal linking between cluster pieces, freshness signal accuracy (dateModified across thousands of articles), legacy AMP cleanup, ad-related Core Web Vitals issues, and infinite scroll handling. Author entity consistency becomes a major signal as E-E-A-T weight increases.
SaaS. JavaScript rendering is usually the headline issue — many SaaS sites are built on React, Vue, or Next.js without consistent server-side rendering across all routes. Documentation hub indexing, blog versus product page hierarchy, and trial signup flow conversion paths all become audit territory. International setups are common and add their own layer.
Multi-region or multi-language sites. Hreflang implementation at scale, regional canonical strategy, server location and CDN configuration, country-targeted Search Console properties. These sites typically need their own audit framework rather than a universal checklist — the international setup section above understates the work involved.
A site’s complexity scales the audit, not just in length but in which signals matter most. A 50-page service business site does not need the same audit as a 250,000-page ecommerce catalog. The framework adapts. The priorities shift.
How to prioritize findings
A 100-point audit nobody acts on is worse than a 20-point audit that gets executed. The prioritization framework we use:
Impact × Effort. Plot every finding on a 2×2 grid.
- High impact, low effort → Fix this week. Common examples: missing canonical tags, blocked critical resources in robots.txt, missing alt text on important images, accidental noindex on important pages.
- High impact, high effort → Plan and resource. Common examples: SSR implementation, site restructure, Core Web Vitals overhaul, content consolidation across hundreds of URLs.
- Low impact, low effort → Knock out when convenient. Common examples: minor schema additions on secondary pages, image alt text on low-traffic pages.
- Low impact, high effort → Defer or skip. Common examples: hreflang setup for regions you don’t actually serve, schema for pages with no organic traffic.
For impact, ask: how many pages does this affect, what’s their traffic and conversion value, and how much will fixing it change them? For effort, ask: hours of development work, plus deployment risk.
Ship the audit as a prioritized list with effort estimates and impact rationale. Don’t ship a checkbox sheet. The deliverable is decisions, not data.
When this checklist isn’t enough
The framework above covers what a competent technical audit looks like. The work that separates an agency audit from a confident DIY run lives in five areas not detailed in this article.
Log file analysis. Server log analysis surfaces what Googlebot and AI crawlers actually do on the site — which pages they fetch, which they skip, where crawl budget is wasted on parameter URLs or noindex pages. For sites over 10,000 URLs, this is often where the highest-impact findings live. It cannot be done with Screaming Frog alone.
Ecommerce facets, schema at scale, and inventory signaling. Product schema implementation across thousands of SKUs, faceted navigation rules that don’t bleed parameter URLs into the index, and how out-of-stock or discontinued product signals get handled — each one becomes its own diagnostic exercise that a checklist can’t substitute for.
Migration risk assessment. Pre-migration audits, redirect mapping at scale, traffic protection planning, post-migration monitoring. The cost of a botched migration runs into months of lost organic revenue. Doing this without a checklist is reckless; doing it with only the checklist above is insufficient.
JavaScript framework–level diagnostics. React, Vue, Next.js, and similar frameworks each have their own rendering edge cases. Identifying whether content is missing in the rendered DOM is the easy part. Identifying which framework configuration is producing the gap — and how to fix it without breaking other things — requires deeper expertise.
Historical trend analysis and competitive benchmarking. Year-over-year trend analysis, competitor visibility comparisons, share-of-voice tracking on topic clusters. These provide the context that turns a list of findings into a strategy.
These categories are where ongoing audit work earns its keep at agency depth. If your site is genuinely simple, the framework above is sufficient. If your site is complex enough that a botched audit costs you real revenue, the framework above is the start, not the finish.
Frequently asked questions
How often should I run a technical SEO audit?
Full audits twice a year for most sites. Quarterly for high-traffic ecommerce or content sites where small issues compound fast. Add a focused audit any time you ship a significant site change — CMS migration, redesign, framework swap, structured data overhaul, major navigation change.
What tools do I need for a technical SEO audit?
The free stack covers most of it: Google Search Console, Bing Webmaster Tools, PageSpeed Insights, Chrome DevTools, Schema Markup Validator, Rich Results Test. For larger sites or recurring audits, Screaming Frog (~$259/year), Sitebulb, or Ahrefs Site Audit are the standard agency tools. You don’t need an enterprise platform like Botify or Lumar until you’re over 100,000 URLs.
Should I do my own audit or hire an agency?
If you have a technically capable SEO or developer on staff and the time to spend 20+ hours on it, do it yourself. The free tools and Google’s documentation are good enough for a first-pass audit. Hire an agency when you don’t have in-house capacity, when the site is complex enough that mistakes are expensive (ecommerce, multi-region, JavaScript-heavy), or when you want an outside perspective on what’s actually broken versus what looks broken.
What’s the difference between an SEO audit and a website audit?
Scope. A technical SEO audit covers crawlability, indexation, speed, structure, and AEO readiness specifically. A full website audit also covers content, UX, conversion, accessibility, security, and analytics. The two overlap, but if you’re hiring for one, ask what’s actually in scope — the words are used loosely in the industry, and the deliverable can vary wildly between agencies.
Kim Schmutzler is the founder and director of SEO at KDGS Works, a content and SEO agency in Overland Park, Kansas. Book a technical SEO audit with our team at kdgsworks.com.
