Most agencies audit their clients once. They produce a report, present it in a kickoff call, and then move on to the work of implementing the highest-priority fixes. Six months later, the site has been through three rounds of content updates, a plugin update broke the sitemap, and a new CMS deployment introduced render-blocking scripts that dropped the INP score from 180ms to 680ms -- but nobody noticed because no one was looking.
We run a full technical SEO audit every 90 days for every client on our retainer programmes. Not a partial check. A complete pass through every item on this checklist, with scores recorded and trended over time. The reason is straightforward: technical SEO is not a one-time fix, it's an ongoing maintenance discipline. Sites regress. Google changes what it measures. Third-party scripts accumulate. The only way to catch these issues before they cost you rankings is to look, regularly and systematically.
In this article, I'm publishing our full 15-point checklist -- the exact checkpoints we run, the benchmarks we grade against, and the tools we use to run each check. Whether you're an in-house SEO manager at a Houston TX business or running an agency, this is a process you can implement immediately.
Why Technical SEO Is More Important in 2026
Technical SEO matters more now than at any point in the past five years, for two connected reasons: Google's page experience signals have become more granular and more impactful, and the rise of AI-powered answer engines means that content which can't be efficiently crawled and parsed simply doesn't get cited.
Google replaced FID (First Input Delay) with INP (Interaction to Next Paint) as a Core Web Vital in March 2024. INP measures the latency of all user interactions throughout a page session, not just the first one. It is a significantly more demanding metric than FID was, and many sites that passed FID comfortably are failing INP. Google Search Central has confirmed that page experience signals -- including all three Core Web Vitals -- are ranking factors.
The AI citation angle matters too. When Perplexity or Google's AI Overview system crawls content for citation purposes, it uses the same Googlebot infrastructure -- which means content that's poorly structured, slow to load, or blocked by misconfigured robots.txt is just as invisible to AI systems as it is to traditional search. Our guide to AEO strategy for 2026 covers the content side of AI search optimisation. This article covers the technical foundation that makes that content accessible in the first place.
For Houston TX businesses competing in local search, the technical baseline matters acutely. Local packs and Google Business Profile listings are increasingly populated based on signals that include page experience data. A Houston law firm or real estate agency with a slow, technically broken website is at a structural disadvantage relative to competitors who have invested in technical health -- even if their content and backlink profile are stronger.
The 15-Point Technical SEO Checklist
We've organised the 15 checkpoints into four groups: Performance and Core Web Vitals, Crawlability and Indexation, Structured Data, and On-Page Technical. Each checkpoint includes what we measure, what tool we use, and the benchmark we grade against.
Performance & Core Web Vitals
1. LCP -- Largest Contentful Paint. LCP measures how long it takes for the largest visible element (usually a hero image or headline) to render in the viewport. Our target is under 2.5 seconds on mobile. We test using PageSpeed Insights (which uses real-world Chrome User Experience Report data) and Ahrefs Site Audit for bulk page testing. Common causes of LCP failure: unoptimised hero images (not using WebP, not preloaded), slow TTFB from the server, render-blocking CSS, and third-party scripts that delay rendering.
2. INP -- Interaction to Next Paint. INP replaced FID in March 2024 and is the Core Web Vital most sites are currently failing. The target is under 200ms. INP measures the time from any user interaction (click, tap, keyboard input) to the next visual update from the browser. The primary culprit in most cases is heavy JavaScript execution on the main thread -- event handlers that do too much work, React or Vue components that take too long to reconcile on interaction. Fixing INP almost always involves profiling the main thread with Chrome DevTools and deferring or breaking up long tasks.
3. CLS -- Cumulative Layout Shift. CLS measures how much the page layout moves unexpectedly during load. The target is under 0.1. The most common cause we see in client audits is images without explicit width and height attributes: the browser reserves no space for the image, so when it loads, everything below shifts down. Ad slots that expand after load are the second most common culprit. CLS is one of the easier Core Web Vitals to fix once you've identified the cause -- it usually requires adding two HTML attributes or a CSS rule.
4. TTFB -- Time to First Byte. TTFB measures how long it takes for the browser to receive the first byte of the HTML document from the server. Our target is under 600ms. High TTFB is almost always a server-side issue: slow hosting, database queries on page load, no caching headers, or geographic distance between the server and the user. For Houston TX businesses with a primarily US audience, server location matters -- a site hosted in Europe will consistently have higher TTFB for US visitors than an equivalent site on US-region hosting. Fixing TTFB means either upgrading hosting, implementing server-side caching (Redis or similar), or deploying a CDN.
Crawlability & Indexation
5. robots.txt audit. The robots.txt file controls which parts of the site Googlebot is allowed to crawl. It is surprisingly common to find mission-critical pages accidentally blocked in robots.txt, particularly after site migrations or CMS upgrades. We check every client's robots.txt manually and cross-reference it with Google Search Console's URL Inspection tool for high-priority pages. A blocked homepage or service page can wipe out rankings in days -- and it's not always obvious why until you check the file.
6. XML sitemap health. The XML sitemap should list every indexable URL on the site, each returning a 200 HTTP status code. It should be submitted to Google Search Console (and Bing Webmaster Tools). We check: that the sitemap is current (some plugins cache the sitemap and it falls out of date), that all URLs return 200 (not 301, 404, or noindex), and that the sitemap itself is returning a 200. A sitemap with 404s in it tells Google that the site is poorly maintained -- it's a trust signal in the wrong direction.
7. Crawl budget and redirect chains. Google allocates a crawl budget to each site based on its size, crawl rate, and domain authority. Sites that waste crawl budget on redirect chains, duplicate content URLs, or parameter variations give Google less opportunity to crawl and index the content that actually matters. We check crawl stats in Google Search Console (Settings > Crawl Stats) and look for redirect chains longer than one hop. Every redirect chain should be collapsed to a direct redirect from the original URL to the final destination.
8. Canonical tag audit. Every indexable page should have a self-referencing canonical tag in the head. Pages with multiple URL variants (with/without trailing slash, with/without www, with UTM parameters) should have canonical tags pointing to the primary URL. We check for: missing canonicals, conflicting canonicals (canonical pointing to a different page than the page itself intends), and canonicals pointing to noindex pages (which creates a contradiction that Google handles unpredictably).
9. Noindex audit. We run a full crawl of every client site and flag every page tagged with noindex in the meta robots tag or X-Robots-Tag header. Then we cross-reference against a list of pages that should be indexed. It is not uncommon to find important service pages, blog posts, or location pages with noindex left over from a development phase or accidentally added during a CMS migration. A noindex tag on a page that should be indexed means that page does not exist in Google's index -- period.
Structured Data
10. Organization schema. Every website should have Organization schema (or LocalBusiness schema for local businesses) implemented site-wide, typically in the footer or a global template. It must include: name, url, description, address (with PostalAddress), contactPoint (with telephone), and sameAs (linking to social profiles and Wikidata). We validate using schema.org Validator -- zero errors is the only acceptable result.
11. Article / BlogPosting schema. Every blog post and article should have Article or BlogPosting schema that includes: headline, author (with @type Person, name, and jobTitle), publisher (linking back to Organization schema), datePublished, and dateModified. Keeping dateModified current is important -- AI systems and Google both use freshness signals, and a post that was updated recently but doesn't reflect the update in schema is missing an easy win.
12. FAQPage schema. FAQPage schema is the highest-leverage schema type for both AEO and GEO purposes. We implement it on all service pages, all blog posts with a FAQ section, and all landing pages. The questions and answers in the schema must exactly match the visible FAQ content on the page. Mismatches between schema and visible content are a trust violation that can result in the schema being ignored or, in persistent cases, a manual action.
13. LocalBusiness schema. For Houston TX clients and any client serving a defined geographic market, LocalBusiness schema is essential. It extends Organization schema with: geo (latitude/longitude coordinates), openingHoursSpecification, priceRange, and areaServed. Correctly implemented LocalBusiness schema contributes to Google Knowledge Panel generation and local AI citation signals -- two of the most important GEO infrastructure elements. See our GEO guide for how this feeds into AI citation performance.
14. Schema validation: zero errors. All schema across the site should pass validation at schema.org Validator with zero errors and zero warnings. Common errors we find: missing required properties (often @id or url), incorrect @type values, mismatched quotes in JSON-LD, and schema that references URLs returning 404. We run every client's schema through validation as part of every quarterly audit and treat any error as a blocking issue to be resolved before the next cycle.
On-Page Technical
15. H1 audit. Every page on the site should have exactly one H1 tag. The H1 should be the primary heading visible to users and should align with the primary keyword intent of the page. Common failures: pages with no H1 (particularly on dynamically generated pages), pages with two or more H1s (often caused by CMS templates that insert a default H1 alongside the content-area H1), and H1 text that doesn't match the page's target keyword intent. The H1 is one of the strongest on-page signals Google uses to understand what a page is about -- it deserves attention every audit cycle.
Want us to run this audit on your site?
We'll run every one of these 15 checks on your site, score each metric against our benchmarks, and give you a prioritised fix list in a 30-minute strategy call.
Book a Free Strategy Call →Benchmarks We Use to Grade Each Metric
Audit findings without benchmarks are observations without context. Here are the exact benchmark thresholds we use to classify each metric as Pass, Needs Work, or Fail. These align with Google's published guidance where available and our own client data where Google has not published specific thresholds.
- LCP (mobile): Pass = under 2.5s | Needs Work = 2.5s to 4.0s | Fail = over 4.0s
- INP (mobile): Pass = under 200ms | Needs Work = 200ms to 500ms | Fail = over 500ms
- CLS: Pass = under 0.1 | Needs Work = 0.1 to 0.25 | Fail = over 0.25
- TTFB: Pass = under 600ms | Needs Work = 600ms to 1,800ms | Fail = over 1,800ms
- Mobile PageSpeed Score (PSI): Pass = 80 or above | Needs Work = 60 to 79 | Fail = under 60
- XML sitemap URLs returning 200: Pass = 100% | Needs Work = 95% to 99% | Fail = under 95%
- Schema validation errors: Pass = zero | Needs Work = warnings only | Fail = one or more errors
- Pages with missing H1: Pass = zero | Needs Work = 1% of pages | Fail = over 1% of pages
- Redirect chains (2+ hops): Pass = zero | Needs Work = 1 to 5 | Fail = over 5
The benchmark that causes the most friction with clients is the mobile PageSpeed score. Many sites with excellent desktop scores have mobile scores in the 40-60 range, and the gap is almost always explained by three things: unoptimised images, render-blocking scripts, and fonts loaded synchronously. Fixing these three items is the fastest path from a failing mobile score to a passing one.
The Tools We Use for Every Audit
We use a consistent tool stack for every quarterly audit. No single tool catches everything, and the combination we've landed on gives us comprehensive coverage across all 15 checkpoints.
- Google Search Console (free) -- Core Web Vitals field data, crawl stats, index coverage, sitemap status, URL inspection. The starting point for every audit. Field data in GSC is the most accurate reflection of real-user experience because it comes from Chrome users visiting the actual site.
- PageSpeed Insights (free) -- Lab and field data for individual URL performance. We run PSI on the homepage, the top 5 traffic pages, and any page flagged as Poor in GSC Core Web Vitals. PSI's Opportunities section gives specific, actionable fix recommendations.
- Screaming Frog SEO Spider (paid) -- Full site crawl for redirect chains, missing canonical tags, H1 audit, noindex audit, robots.txt validation, and broken internal links. Screaming Frog is the closest thing to Googlebot in a desktop application. We run it on every client site at the start of each quarterly audit cycle.
- Ahrefs Site Audit (paid) -- Automated crawl with prioritised issue detection, LCP and Core Web Vitals monitoring, and historical trend data. Ahrefs Site Audit gives us a 90-day trend view that shows whether issues are improving, stable, or deteriorating between audit cycles.
- Schema.org Validator (free) -- Schema markup validation for all structured data types. We paste the JSON-LD from each page template and verify zero errors before considering any schema implementation complete.
- Semrush Site Audit (paid) -- Crawlability issues, duplicate content detection, HTTPS status, and (in the AI Toolkit) AI Overview appearance tracking. We use Semrush alongside Ahrefs for cross-validation on crawl issues, since each tool's crawler occasionally misses things the other catches.
The One Fix That Moves the Needle Most
If you're working through this checklist and need to prioritise, here is the honest answer based on our Houston TX client work over the past 12 months: fixing INP on mobile drives the biggest ranking improvement in 2026.
INP is the Core Web Vital that most sites are currently failing, and it is the one that Google has most recently added to its ranking signals. Sites that are failing INP -- particularly on mobile -- are competing at a handicap against sites that pass it, even if everything else about the sites is equivalent. The ranking lift from taking a mobile INP from the Needs Work range (200-500ms) to Pass (under 200ms) is, in our experience, the single highest-return technical SEO investment available right now.
The most common cause of failing INP in our client base is not poorly written application code -- it's accumulated third-party scripts. Chat widgets, analytics platforms, advertising tags, heatmap tools, and conversion tracking scripts all execute on the main JavaScript thread and all contribute to INP degradation. A site that has added three or four of these over two years without ever reviewing their performance impact will almost always fail INP on mobile.
The fix process we use: run a Chrome DevTools performance profile on the pages with the worst INP scores, identify the long tasks on the main thread (anything over 50ms qualifies), trace each long task back to its source script, then either defer that script until after the page is interactive, load it asynchronously, or (if it's not providing measurable business value) remove it entirely.
For a Houston e-commerce client we worked with in late 2025, this process -- combined with image optimisation and eliminating one redundant analytics tag -- took their mobile PageSpeed score from 45 to 91 in three weeks. Their organic clicks increased 34% in the subsequent 60 days, which we attribute primarily to the INP and LCP improvements restoring competitive ranking positions they had lost over the preceding year.
"We went from a 45 to 91 mobile PageSpeed score in three weeks and saw a 34% increase in organic clicks over the following 60 days. The INP fix was the change that made the biggest difference." -- Houston e-commerce client, Q4 2025
Technical SEO is the foundation that everything else depends on. Your AEO content strategy, your GEO schema stack, your local citation programme -- none of it performs at its potential if Googlebot can't efficiently crawl and index your site, or if users bounce from poor Core Web Vitals performance before they ever engage with your content. The 15-point checklist in this article is the systematic process we use to keep that foundation solid.
If you want to know where your site currently stands against these benchmarks, our SEO & AEO services include a full technical audit as part of onboarding. You can also book a free strategy session with our Houston team to talk through your specific situation before committing to anything.