Tag: SEO Best Practices

Current, practical SEO best practices.

The Complete Technical SEO Checklist for 2026

A website can have excellent products, useful content and a credible brand yet still struggle in search because of technical problems underneath the surface. Search engines must be able to discover a URL, access it, understand its content and select the correct version before rankings are even possible.

This technical SEO checklist gives website owners, marketers and developers a practical audit sequence for 2026. It covers the issues that matter across WordPress, Shopify, Wix, WooCommerce and custom websites, without treating every warning from an SEO tool as an emergency.

Quick answer: Start by confirming that important pages return a successful status code, are not blocked, can be crawled through internal links and contain indexable content. Then review canonicalization, sitemaps, redirects, mobile rendering, Core Web Vitals, structured data and ongoing monitoring.

1. Confirm that important pages are technically eligible for Google Search

Google describes three minimum technical requirements for a page to be eligible for indexing: Googlebot must not be blocked, the page must work and return an HTTP 200 success response, and it must contain indexable content. Meeting those requirements does not guarantee indexing or rankings, but failing one of them can prevent the page from appearing at all.

Open each priority URL without logging in.
Confirm it returns a 200 status code.
Check that the main content appears in the rendered HTML.
Verify that robots.txt does not block the URL or essential page resources.
Look for accidental noindex directives in the HTML or HTTP headers.

Use Google Search Console’s URL Inspection tool for important pages. Test the live URL when the indexed version looks outdated or when you suspect that Google sees something different from visitors.

Common mistake: Blocking a page in robots.txt does not reliably remove it from search. If you need Google to process a noindex rule, the crawler must be allowed to access the page and read that directive.

2. Audit robots.txt and indexing directives separately

Robots.txt controls crawling. Robots meta tags and X-Robots-Tag headers control indexing and search-result presentation. They solve different problems, so audit them separately.

Robots.txt checks

Confirm the file is available at /robots.txt.
Look for broad rules that block important folders, product pages, scripts, CSS or images.
Check staging rules after a migration. A site can accidentally launch with a sitewide disallow rule.
Do not block resources required to render the main content.
Include the sitemap location when useful, but do not assume this fixes crawlability problems.

Meta robots and header checks

Find pages carrying noindex, nofollow or restrictive snippet directives.
Confirm that intentional noindex rules are limited to pages that should not appear in search.
Check PDFs and other non-HTML files for X-Robots-Tag headers when relevant.
Review CMS search visibility settings after theme, plugin or platform changes.

3. Make every important page reachable through crawlable internal links

Googlebot discovers URLs through links, sitemaps and redirects. Important pages should be reachable through normal HTML links from other discoverable pages. A sitemap can support discovery, but it does not replace a coherent website structure.

Use real <a href> links rather than JavaScript-only click events.
Ensure every important page has at least one useful incoming internal link.
Keep important pages reasonably close to the homepage or a relevant category hub.
Use descriptive anchor text that explains the destination.
Check pagination, faceted navigation, related products and archive pages.
Find URLs known to the CMS, sitemap, analytics or Search Console but missing from a normal crawl.

For a detailed process, read our guide to finding and fixing orphan pages. It explains how to compare crawl data with sitemap, analytics, CMS and Search Console exports.

4. Clean up HTTP status codes, redirects and broken links

Status codes tell crawlers what happened when they requested a URL. A technical audit should identify errors and then evaluate their context rather than blindly forcing every URL to return 200.

Status	What it usually means	What to review
200	The request succeeded	Confirm the page contains the intended content and is not a soft 404.
301 or 308	Permanent redirect	Use for lasting URL changes and point directly to the final relevant destination.
302 or 307	Temporary redirect	Confirm the change is genuinely temporary.
404 or 410	The resource is unavailable	Restore, redirect or leave removed depending on whether a relevant replacement exists.
5xx	Server-side failure	Investigate hosting, application, database, timeout and deployment errors promptly.

Remove internal links that point to broken URLs.
Avoid redirect chains and loops.
Do not redirect unrelated removed pages to the homepage.
Update internal links to point directly to the final URL.
Monitor intermittent server errors, not only permanent failures.
Check whether custom 404 pages return the correct 404 status.

5. Review canonical tags and duplicate URLs

Duplicate or near-duplicate URLs can appear because of tracking parameters, product variants, filters, print pages, HTTP and HTTPS versions, trailing-slash variations or CMS-generated archives. Canonical signals help search engines choose a representative URL.

Use a self-referencing canonical on indexable primary pages when appropriate.
Point duplicate versions to the preferred canonical URL.
Keep canonical tags, internal links, sitemaps and redirects consistent.
Do not canonicalize a page to an unrelated destination.
Check that canonical URLs return 200 and are not blocked or noindexed.
Review parameter and faceted-navigation URLs on ecommerce sites.

Best practice: Treat canonicalization as a group of reinforcing signals. A canonical tag is less persuasive when the sitemap lists another URL and all internal links point somewhere else.

6. Validate the XML sitemap

An XML sitemap helps search engines discover new and updated URLs. It should be a clean list of canonical pages you want considered for indexing, not an archive of every URL your website has ever generated.

Include only canonical, indexable URLs that return 200.
Use absolute URLs with the correct HTTPS and hostname version.
Remove redirects, errors, noindexed pages and duplicates.
Keep modification dates accurate when your CMS supports them.
Split very large sitemaps when they approach protocol limits.
Submit the sitemap or sitemap index in Google Search Console.
Compare sitemap URLs with crawled URLs to find disconnected pages.

Google supports several sitemap formats and limits a single sitemap to 50 MB uncompressed or 50,000 URLs. Most small and medium websites will remain far below those limits, but large ecommerce and publishing sites should use sitemap index files.

7. Check mobile rendering and JavaScript dependencies

Google primarily uses mobile crawling. Your audit should therefore focus on what the mobile crawler can access and render, not only how the desktop page looks to a logged-in administrator.

Compare mobile and desktop main content.
Ensure important text is not omitted from the mobile version.
Check navigation, accordions, filters and product options on small screens.
Confirm lazy-loaded content appears when a crawler renders the page.
Give each important application screen or content state a stable URL.
Make sure JavaScript-generated links are crawlable.
Inspect rendered HTML for text, links, canonical tags and structured data.

JavaScript is not automatically bad for SEO. The risk appears when essential content or navigation depends on scripts that fail, require unsupported interactions or load differently for crawlers.

8. Measure Core Web Vitals with field and lab data

Core Web Vitals measure real-world loading performance, responsiveness and visual stability. Google’s current good-experience targets are:

Largest Contentful Paint (LCP): within 2.5 seconds.
Interaction to Next Paint (INP): under 200 milliseconds.
Cumulative Layout Shift (CLS): 0.1 or less.

Use field data to understand actual user experience and lab tools to diagnose causes. A single fast test from your office does not represent visitors using slower phones, congested networks or distant locations.

Common improvements

Compress and correctly size hero images.
Preload the genuine LCP resource when appropriate.
Reduce unused JavaScript and third-party scripts.
Break up long main-thread tasks.
Reserve dimensions for images, embeds and advertisements.
Use efficient caching and content delivery.
Improve server response time and database performance.

9. Validate structured data without treating it as a ranking shortcut

Structured data gives search engines explicit information about a page and may make the page eligible for supported rich results. It does not guarantee a rich result and should accurately represent visible page content.

Use a supported type that matches the page.
Include required and useful recommended properties.
Ensure marked-up information is visible and accurate.
Do not add reviews, prices, availability or author details that are not present on the page.
Use Google’s Rich Results Test for supported features.
Monitor enhancement reports in Search Console.
Retest templates after theme or plugin updates.

Common useful types include Product, Article, LocalBusiness, Organization, BreadcrumbList, Event, JobPosting, Recipe and VideoObject. Use only the types that genuinely fit your content and Google’s current search feature support.

10. Audit titles, headings, images and metadata at scale

Technical SEO overlaps with on-page SEO when templates generate missing, duplicated or misleading metadata. Review patterns across page types rather than editing random URLs one by one.

Give each important page a descriptive title.
Write useful meta descriptions for pages that deserve search traffic.
Use a clear main heading that matches the page purpose.
Keep heading order understandable for readers.
Add meaningful alt text to informative images.
Use descriptive filenames where practical.
Prevent templates from creating thousands of near-identical titles.
Check social sharing titles and descriptions for important pages.

11. Check international and local technical signals where relevant

Businesses serving several countries or languages need consistent regional architecture. Local businesses should also ensure that search engines can understand their location and business identity.

Use separate, crawlable URLs for language or regional versions.
Implement reciprocal hreflang annotations where needed.
Include an x-default version when it fits the experience.
Avoid automatic location redirects that prevent crawlers or travellers from accessing another version.
Keep business name, address, phone and service information consistent.
Use relevant LocalBusiness structured data without inventing information.

12. Review technical readiness for AI search experiences

Technical SEO remains foundational for AI-powered search. Google states that its established SEO best practices continue to apply to AI Overviews and AI Mode, with no special markup required for inclusion.

Keep important content crawlable and indexable.
Use clear internal linking and logical topic relationships.
Publish original, useful information rather than scaled variations of the same page.
Support text with relevant images and video when they improve the answer.
Make factual claims easy to verify.
Keep product, local and organizational information current.
Do not create hundreds of thin pages for every imagined AI query variation.

For more detail, see our guide to optimizing a website for generative AI features on Google Search and our practical GEO strategies for AI visibility.

13. Create a monitoring and change-control process

A technical SEO audit is a snapshot. Websites change every week through new content, plugin updates, product imports, redirects, theme releases and developer deployments. The stronger approach is continuous monitoring with clear ownership.

Crawl the site after migrations, redesigns and major releases.
Monitor Search Console indexing, crawl and enhancement reports.
Track changes in organic landing pages and conversions.
Alert the team to new 5xx errors, broken internal links and accidental noindex rules.
Keep backups before changing templates, metadata or redirect rules.
Document what changed, why it changed and how success will be measured.
Verify fixes after deployment and retain a rollback path.

Technical SEO audit priority order

Critical access, security and server errors
Accidental blocking or noindex directives
Important pages missing from the index
Broken links, redirect loops and incorrect canonicals
Site architecture and orphan pages
Mobile rendering and JavaScript issues
Core Web Vitals and performance
Structured data and search appearance
Template metadata and image optimization
Ongoing monitoring and prevention

Common technical SEO mistakes

Trying to fix every crawler warning without checking business impact.
Using robots.txt to remove pages from the index.
Submitting a sitemap full of redirects, errors and noindexed URLs.
Redirecting every deleted page to the homepage.
Changing URLs without mapping redirects first.
Optimizing only the homepage while product, service or article templates remain broken.
Relying on one speed-test result instead of real-user data.
Adding structured data that does not match visible content.
Publishing pages without internal links.
Applying automated changes without backups, verification or rollback.

Technical SEO checklist

Priority pages return 200 and contain indexable content.
Googlebot is not unintentionally blocked.
Noindex rules are intentional and crawlable.
Robots.txt is valid and does not block required resources.
Important URLs have crawlable internal links.
Broken links, redirect chains and loops are fixed.
Canonical tags point to valid preferred URLs.
The XML sitemap contains clean canonical URLs.
Mobile pages contain the full important content.
JavaScript does not hide essential links or text.
Core Web Vitals are measured with field data.
Structured data is valid, visible and appropriate.
Titles, headings and image metadata work across templates.
International and local signals are consistent.
Important changes have backups, verification and rollback.

Frequently asked questions

How often should you run a technical SEO audit?

Small, stable websites may need a full review every quarter, with lighter monthly monitoring. Ecommerce, publishing and frequently updated websites should monitor critical issues continuously and run deeper audits after major releases.

Does a perfect technical audit guarantee higher rankings?

No. Technical SEO makes content accessible and understandable, but rankings also depend on relevance, usefulness, competition, reputation and many other signals. A technically perfect empty page is still not useful.

Which technical SEO issue should be fixed first?

Start with issues that prevent important pages from loading, being crawled or being indexed. Server failures, accidental sitewide blocking and incorrect noindex rules normally take priority over cosmetic metadata warnings.

Can WordPress, Shopify or Wix handle technical SEO automatically?

Modern platforms handle many defaults, but they cannot know your preferred site architecture, which duplicate URL should win, whether a removed page needs a redirect or whether an important page has become orphaned. Platform automation reduces manual setup; it does not remove the need for review.

Is technical SEO different for AI search?

The fundamentals are largely the same. Content must be accessible, indexable, useful and connected to the rest of the site. Clear structure and reliable information support visibility in both conventional and AI-powered search experiences.

Conclusion

The best technical SEO audit does not produce the longest list of errors. It identifies which issues prevent valuable pages from being discovered, understood and used, then fixes those issues in a safe order.

Begin with access and indexability. Then improve site structure, canonicalization, sitemaps, mobile rendering, page experience and structured data. Finally, put monitoring in place so the same problems do not quietly return.

If you would rather automate ongoing checks instead of repeating the entire process manually, tools like Visible Pilot can monitor a connected website, prioritize new SEO opportunities and apply approved improvements with backups and rollback available.

Authority references: Google Search technical requirements; Google crawling and indexing documentation; Google sitemap guidance; Google Core Web Vitals guidance; Google guidance for generative AI features.

July 24, 2026

How to Get Cited in AI Search Results: 12 Practical Steps

Ranking on Google is no longer the only way a website can earn visibility. Search experiences now generate direct answers, summaries, comparisons and recommendations—and many of those answers include citations to source pages.

That creates a new question for website owners: how do you make your content more likely to be selected as a source?

The answer is not to write for robots or repeat keywords more often. AI search citations are more likely when a page is technically accessible, clearly focused, easy to extract, supported by evidence and useful for the exact question being asked.

Quick answer

To improve your chances of being cited in AI search results, publish pages that answer specific questions directly, support important claims with reliable evidence, use clear headings and comparison formats, maintain accurate entity information, strengthen internal links and keep content technically crawlable and current.

There is no guaranteed method for earning a citation. Different AI systems use different retrieval, ranking and generation methods. Your goal is to make each important page a strong, trustworthy source that can be found, understood and reused without ambiguity.

What is an AI search citation?

An AI search citation is a visible reference or link that identifies a webpage as a source used to support a generated answer. Citations may appear beside a sentence, below an answer or inside a source panel, depending on the search platform.

A citation is different from a traditional organic ranking. A page may rank well but not be cited in a particular generated answer. Another page may be selected because it provides a more precise definition, clearer comparison, stronger evidence or more current information for the specific prompt.

Traditional search result	AI search citation
A webpage appears as a ranked result	A webpage is referenced as evidence inside a generated answer
Usually evaluated at query level	May be selected for a particular claim or section of an answer
Success is measured with rankings, impressions and clicks	Success may include citations, brand mentions, referenced pages and assisted conversions
The user chooses which result to open	The AI system summarizes information before the user visits a source

How AI systems choose pages to cite

There is no universal citation algorithm. However, most citation opportunities depend on four broad stages:

Discovery: the system must be able to find and access the page.
Retrieval: the page must appear relevant to the prompt or grounding query.
Understanding: the system must be able to identify the page’s subject, entities and key claims.
Selection: the content must be useful and trustworthy enough to support the generated answer.

This is why AI visibility still depends on strong SEO foundations. A page that is blocked, orphaned, duplicated or poorly structured creates friction before content quality can even be evaluated.

12 ways to make content more citation-worthy

1. Target a precise question or decision

Broad articles often contain too little depth on any single question. A stronger page has a clear job: define a concept, compare options, explain a process, diagnose a problem or help the reader make a decision.

For example, “AI search optimization” is broad. “How to measure citations in AI search results” is more precise. The second topic gives the page a clear retrieval purpose and makes it easier to match with a specific prompt.

2. Put the direct answer near the beginning

Do not make readers work through a long introduction before reaching the useful information. Give a concise answer first, then explain the conditions, evidence and practical steps.

A good opening answer usually states what the concept means, what matters most and what the reader should do next. This format helps people scan the page and creates a self-contained passage that can be understood without missing context.

3. Use headings that match real follow-up questions

Descriptive headings improve navigation and help separate the page into distinct answer units. Instead of vague headings such as “More details” or “Key factors,” use headings like “Does schema markup improve AI citations?” or “How often should citation-focused content be updated?”

Each section should answer the heading immediately. This makes the article useful for conversational searches where one initial question produces several follow-up questions.

4. Support factual claims with primary sources

Unsupported claims are difficult to trust and easy to replace. When you discuss search engine behaviour, platform features, standards, regulations or research findings, link to the original documentation or study whenever possible.

For example, Google explains that crawlable internal links and descriptive anchor text help it discover pages and understand their context. Microsoft now provides AI citation reporting in Bing Webmaster Tools, including cited pages and grounding queries. These are stronger foundations than repeating unverified advice from another marketing article.

Useful official references include Google’s link best practices, its Article structured data documentation, and Microsoft’s explanation of AI Performance in Bing Webmaster Tools.

5. Add original information competitors cannot copy easily

A page becomes more valuable when it contributes something beyond a generic summary. Original information may include first-hand tests, anonymized audit findings, customer questions, screenshots, templates, calculations, expert commentary or a documented process.

You do not need a large research budget. A small study of 20 websites can be more citation-worthy than a broad article that repeats common advice without evidence. Explain the methodology, limitations and date so readers can judge the result properly.

6. Make every important claim unambiguous

A sentence should identify who or what it refers to. Avoid unclear phrases such as “this tool,” “they say” or “it works better” when the reader cannot determine the subject, comparison or condition.

Use exact product names, dates, locations, units and definitions where they matter. Clear entity references help prevent a statement from being separated from the context needed to interpret it accurately.

7. Use tables for comparisons and decision criteria

Tables are useful when readers need to compare tools, tactics, costs, requirements or outcomes. They make relationships explicit and reduce the amount of interpretation required.

Content format	Best use	Common mistake
Definition	Explaining a term quickly	Using jargon inside the definition
Comparison table	Helping users choose between options	Comparing inconsistent criteria
Step-by-step process	Explaining how to complete a task	Skipping prerequisites or decision points
Checklist	Auditing whether requirements are met	Listing actions without explaining priority
FAQ	Answering natural follow-up questions	Adding questions only to repeat keywords

8. Strengthen internal links and topic relationships

Internal links help crawlers discover pages and help users move from a broad explanation to a more specific answer. Google recommends linking important pages from other relevant pages and using concise, descriptive anchor text.

For a deeper foundation, read our guide to generative engine optimization strategies. Website owners troubleshooting brand visibility should also review why ChatGPT may not mention a brand.

Avoid linking every occurrence of a keyword. Add a link when the destination gives the reader meaningful next-step information.

9. Prevent orphan pages and duplicate intent

A high-quality article can remain weak if no other page links to it. Orphan pages receive little internal context and may be harder for crawlers and users to find. Our guide to finding and fixing orphan pages explains how to detect and reconnect them.

Also avoid publishing several pages that answer the same question with only minor wording changes. Overlapping pages divide links, relevance and maintenance effort. Consolidate them or give each page a clearly different intent.

10. Use structured data accurately

Structured data can help search engines identify page types and attributes. For articles, relevant properties may include the headline, author, publication date, modification date and representative images.

Schema does not make weak content authoritative and does not guarantee a citation or rich result. It should accurately describe visible content rather than introduce claims that readers cannot see on the page.

11. Keep important pages current

Outdated facts reduce usefulness and trust. Review pages that contain prices, product capabilities, regulations, platform features, statistics or recommendations more frequently than evergreen definitions.

Update the visible content, not only the modification date. Record meaningful changes, replace obsolete sources and check that external and internal links still resolve correctly.

12. Make the page technically easy to access

Confirm that the preferred URL returns a successful response, is not unintentionally blocked, has the correct canonical, appears in a sitemap and can be reached through crawlable links. Important information should be available in rendered text rather than existing only inside an image or inaccessible script.

For ecommerce sites, technical visibility problems often affect product and collection pages at scale. The diagnostic steps in why a Shopify website may not appear on Google provide a useful starting point.

What does not guarantee an AI citation?

Several popular tactics are presented as shortcuts, but none creates a guaranteed citation.

Adding an FAQ section: useful when it answers real questions, but ineffective when it repeats the article.
Adding schema markup: helpful for machine understanding, but not a substitute for evidence or relevance.
Increasing word count: more words do not automatically create more value.
Repeating brand names and keywords: repetition can reduce readability without strengthening proof.
Publishing large volumes of AI-generated content: scale does not compensate for duplication, weak sourcing or unclear intent.
Blocking all crawlers except Google: this may prevent other search and answer systems from accessing your public content.

How to measure AI search citations

AI visibility measurement is still developing, so use several signals rather than one headline metric.

Metric	What it tells you	Limitation
Citation count	How often supported platforms reference your domain or page	Does not always show prominence or influence
Cited pages	Which URLs are selected most frequently	A citation does not guarantee a visit
Grounding queries	Topics or phrases associated with retrieval	Reporting may be sampled or platform-specific
AI referral traffic	Visits sent from identifiable AI platforms	Some journeys may not pass clear referral data
Brand mentions	Whether the brand appears in relevant answers	Results can vary by wording, time and location
Conversions	Whether AI-assisted visits lead to business outcomes	Attribution may involve several channels

Microsoft’s AI Performance report is especially useful because it exposes total citations, cited pages and grounding queries across supported experiences. Combine platform reporting with analytics, server logs, conversion tracking and a controlled set of recurring test prompts.

A practical AI citation optimization workflow

Choose one valuable prompt cluster. Identify the main question and the follow-up questions a buyer is likely to ask.
Audit the existing results. Review which source types are cited and what evidence or formats they provide.
Select the best page. Improve an existing relevant page before creating a competing URL.
Write the direct answer. State the core answer near the beginning and clarify important conditions.
Add evidence. Include primary sources, original examples and transparent methodology.
Improve extraction. Use descriptive headings, tables, steps and concise definitions.
Confirm technical access. Check status codes, canonical signals, robots rules, sitemaps and rendered content.
Add internal links. Link from relevant existing pages and link out to deeper supporting resources.
Measure over time. Track citations, mentions, referral traffic and conversions rather than checking one prompt once.
Refresh based on evidence. Improve weak sections, update facts and expand the page only where users need more help.

Frequently asked questions

Can a new website get cited in AI search?

Yes, but a new website usually has fewer authority and entity signals. It can compete by targeting narrow questions, publishing genuinely original information, earning relevant third-party mentions and making each page technically accessible and easy to verify.

Does ranking first on Google guarantee an AI citation?

No. Strong rankings may improve discoverability, but generated answers can cite several sources and may select a page that better supports a specific claim, comparison or recent fact.

Does schema markup improve AI visibility?

Accurate structured data can clarify page type and entity attributes, but it does not guarantee inclusion. It works best as part of a broader strategy involving crawlability, helpful content, evidence, internal links and consistent entity information.

How long does it take to earn AI citations?

There is no fixed timeline. Discovery, indexing, source selection and reporting differ by platform. Technical changes may be discovered relatively quickly, while authority, third-party mentions and stable citation patterns usually take longer to develop.

Should every blog post be optimized for AI citations?

Every post should be clear and useful, but not every page needs to target a citation opportunity. Prioritize pages that answer valuable questions, contain distinctive evidence or support an important customer decision.

Final takeaway

The most reliable way to improve AI citation potential is to become a better source. Make your content accessible, precise, well structured, evidence-led and current. Build clear relationships between related pages, remove duplication and publish information that adds something genuinely useful to the web.

Visible Pilot is being built to help website owners identify, prioritize and safely complete SEO and GEO improvements across their sites. The principle remains the same whether the work is manual or automated: visibility improves when search systems can confidently find, understand and trust the right page.

July 24, 2026