The Complete On-Page SEO Checklist for 2026 (Every Element, Explained)

On-page SEO is everything you control on a page itself - the content, HTML tags, structure, links, media and markup - that helps search engines and AI answer engines understand, trust and rank it. In 2026 the priority order is: (1) match search intent with genuinely useful content, (2) make the page crawlable and indexable, (3) get the on-page fundamentals right (title, headings, URL, internal links, schema), (4) earn trust with real E-E-A-T signals, and (5) format the page so both Google and AI engines can lift a clean, quotable answer from it. Get those five in that order and you have done the work that actually moves rankings. This checklist walks through every element under each of them, explains why it matters, and tells you exactly how to do it.

How to use this checklist

This is built to be shared. Send it to your writer, your developer, and whoever owns your CMS, and treat it as the standard every page ships against. Each section has checkbox items you can tick off as you go. You do not need to do all of it on every page in one sitting - work top to bottom, because the order reflects impact. A page with perfect schema but the wrong intent will not rank; a page that nails intent and fundamentals will rank even with a few boxes unchecked.

A practical way to run it: use it as a pre-publish gate for new pages, and as an audit template for existing ones. If you are auditing a whole site rather than a single page, pair this with our SaaS SEO audit guide so you catch site-wide issues (crawl budget, sitemaps, index bloat) that no single-page check will surface. And keep Google Search Console open while you work - it is the ground truth for how Google actually sees each page.

One honest note before we start: on-page SEO is necessary but not sufficient. It is the half you fully control. The other half - authority earned through links and mentions - is covered in our breakdown of on-page vs off-page SEO. This checklist makes your page the best possible candidate; off-page is what gets it over the line in competitive niches.

---

1. Crawlability and indexability basics (page level)

Before any on-page optimization matters, a search engine has to be able to reach the page, render it, and be allowed to keep it in the index. If any of that breaks, nothing else on this list counts. These are page-level checks - site-wide crawlability lives in a technical SEO audit - but every page needs its own clean bill of health.

• The page returns a 200 status code. Not a soft 404, not a 301 chain, not a 500. Check the live URL, not a cached version. A page that resolves through two or three redirects wastes crawl budget and dilutes signals.
• The page is not accidentally blocked in robots.txt. robots.txt controls crawling, not indexing. A blocked page can still appear in results as a bare URL with no snippet, which looks broken. Confirm the page's path is not disallowed unless you genuinely never want it crawled.
• The meta robots tag says what you intend. For a page you want ranked, you want index, follow (or simply no robots meta at all, which defaults to indexable). Search your template and CMS for stray noindex tags - they are the single most common reason a good page silently disappears. A noindex in the HTML head or in the X-Robots-Tag HTTP header will keep the page out of the index no matter how good it is.
• noindex is applied deliberately, not by accident. Thin utility pages (internal search results, tag archives with one post, thank-you pages) should be noindex. Money pages and content should not be. Know which is which.
• The page has a self-referencing canonical tag. The <link rel="canonical"> in the head should point to the page's own clean, preferred URL. This tells Google which version is authoritative and consolidates signals when the same content is reachable through multiple URLs (trailing slashes, tracking parameters, uppercase variants). A missing or wrong canonical is how duplicate-content confusion starts.
• The canonical points to a live, indexable, 200 URL. A canonical that points to a redirected or noindexed URL sends a contradictory signal. Keep it clean.
• The URL is in your XML sitemap. The sitemap is your explicit list of "these are the pages I want indexed." A ranking page that is missing from the sitemap is a signal mismatch; a noindexed page that is still in the sitemap is another. Keep the sitemap and your indexing intentions aligned.
• The page is reachable by internal links, not only via the sitemap. A URL that appears only in the sitemap with no internal links pointing to it is effectively an orphan (more in section 8). Crawlers follow links first.
• Rendered content matches source content. If your site relies heavily on JavaScript, confirm the important text, links and headings appear in the rendered DOM, not only after a user interaction. Use the URL Inspection tool in Search Console to see the rendered HTML Google actually gets.
• No conflicting signals. A page should not be simultaneously canonicalized elsewhere, noindexed, and in the sitemap. Pick one intention per page and make every signal agree.

Why this matters: indexability problems are invisible from the front end. The page looks perfect to you and to your team, and it simply never ranks because Google was told - by a tag nobody remembered adding - to ignore it. Check these first, every time.

---

2. URL structure

The URL is a small ranking signal and a large usability and trust signal. People see it in search results, in AI citations, and when they copy and paste a link. A clean URL is easier to understand, easier to share, and easier for Google to categorize.

• Keep it short and readable. Aim for a URL a human can read aloud and roughly guess the content from. /blog/on-page-seo-checklist beats /blog/2026/08/23/p?id=48213.
• Use the target keyword in the slug. Include the primary keyword or a tight variant. It reinforces relevance and produces a bolded match in some result types. Do not stuff - one clean instance is enough.
• Use hyphens to separate words, never underscores or spaces. Google treats hyphens as word separators and underscores as word joiners. on-page-seo reads as three words; on_page_seo can read as one.
• Use all lowercase. Some servers treat /Checklist and /checklist as different URLs, which creates duplicates. Force lowercase.
• Strip unnecessary parameters. Tracking parameters, session IDs and faceted-navigation strings create infinite near-duplicate URLs. Keep the canonical URL parameter-free; handle tracking with fragments or canonical tags.
• Use a logical folder structure that mirrors your site's topics. /services/link-building-packages tells Google (and the reader) exactly where this sits. Shallow, logical folders help with topical grouping and breadcrumb schema.
• Keep the click depth shallow. Important pages should be reachable within a few clicks of the homepage. A URL buried five folders deep signals low importance.
• Avoid dates in evergreen URLs. A URL like /2024/03/on-page-seo looks stale the moment the year turns, and it makes updating the content awkward because the URL implies a publish date. Reserve dated URLs for genuinely time-bound news.
• Set the URL once and do not change it casually. Every URL change requires a 301 redirect and temporarily disrupts rankings. Decide the slug carefully before publishing, then leave it alone.
• Avoid stop words only if it stays readable. You can trim "a", "the", "and" from slugs for brevity, but never at the cost of clarity. Readability wins.

Why this matters: the URL is the most stable identity a page has. AI answer engines cite it, users judge it before they click, and Google uses its structure to understand your site's hierarchy. A messy URL is a small tax you pay on every impression forever.

---

3. Title tags

The title tag is the single highest-leverage on-page element after content quality. It is the clickable headline in search results, the label in browser tabs, and often the text an AI engine uses to reference your page. It influences both ranking and click-through rate, and a strong title can lift traffic on a page that already ranks without changing anything else.

• Front-load the primary keyword. Put the most important term near the start. Titles are read left to right, weight skews toward the front, and truncation cuts the end. "On-Page SEO Checklist for 2026" leads with the query; "The Ultimate 2026 Guide to On-Page SEO" buries it.
• Keep it within roughly 50 to 60 characters (about 575 pixels). Beyond that, Google truncates with an ellipsis and your carefully chosen ending vanishes. Write to the pixel limit, not a rigid character count, and check how it renders. A truncated title looks careless and costs clicks.
• Make every title unique across the site. Duplicate titles compete with each other and confuse Google about which page answers a query. Each page needs a distinct, specific title.
• Match the search intent of the target query. If the query is informational, the title should promise information ("checklist", "guide", "how to"). If it is a comparison, promise a comparison. A title that mismatches intent gets skipped in the results even when it ranks.
• Write for the click, not only the algorithm. The title competes against nine other results plus AI Overviews. Use specificity, numbers, brackets, and a clear benefit. "The Complete On-Page SEO Checklist for 2026 (Every Element, Explained)" tells the reader exactly what they get.
• Include the brand where it helps, at the end. For most content pages, append " | Brand" only if you have room. On homepage and money pages the brand can lead. Do not let the brand eat the character budget on a content page competing for a broad keyword.
• Use numbers and the current year for listicles and checklists. "2026" signals freshness and specificity, and it is a genuine, updatable fact - so keep it accurate when you refresh the page.
• Do not keyword-stuff the title. "On-Page SEO Checklist, On Page SEO, SEO Checklist 2026" reads like spam, triggers Google to rewrite your title, and repels clicks. One primary keyword plus one natural modifier is the ceiling.
• Avoid ALL CAPS and excessive punctuation. They look like spam and Google may rewrite them.
• Know that Google may rewrite your title. Google rewrites titles it judges unhelpful, truncated, or keyword-stuffed - often pulling from your H1 or on-page text instead. The fix is not to fight it but to write a clear, accurate, well-sized title so Google has no reason to override you. Check Search Console's performance data and the live SERP to see what is actually being shown.
• The title tag and the visible H1 can differ - and often should. The title is optimized for the SERP and CTR; the H1 is optimized for the reader on the page. They should be consistent in meaning but you can tune each for its context.

Why this matters: the title determines whether an impression becomes a visit. Two pages can rank in the same position; the one with the sharper title takes the majority of the clicks. It is also the most common element to be quietly wrong - truncated, duplicated, or auto-generated by a template nobody reviewed.

---

4. Meta descriptions

The meta description is not a ranking factor. Google has said so repeatedly, and it will frequently rewrite the description by pulling a snippet from your page that matches the query. So why bother? Because when Google does use it, a well-written description lifts click-through rate, and CTR feeds back into performance. Treat it as ad copy for your search result.

• Write a unique description for every important page. Duplicated or missing descriptions are a wasted opportunity and a common audit flag. Prioritize your money pages and top content.
• Keep it to roughly 150 to 160 characters. Longer and Google truncates it. Front-load the value so the important part survives truncation on mobile, where the limit is shorter.
• Include the primary keyword naturally. Google bolds query terms that appear in the description, which draws the eye. Use the keyword once, in a sentence that reads well.
• Lead with the benefit or the answer. Tell the searcher what they get and why this result beats the others. "Every on-page element explained, in a checklist your whole team can follow" is a reason to click.
• Match the intent and set an accurate expectation. A description that oversells and underdelivers raises bounce rate, which is a worse outcome than a lower CTR. Be compelling and honest.
• Include a soft call to action where it fits. "Learn how", "See the checklist", "Compare your options" - a gentle nudge, not a hard sell, for content pages.
• Do not stuff keywords. A list of keywords is not a sentence and will be rewritten or ignored.
• Accept that Google may ignore it. For many queries Google generates its own snippet from the page body. That is fine - it means your body copy answers the query directly, which is exactly what you want. Write the description anyway for the queries where it is used.

Why this matters: the description is the one place you get to make your sales pitch directly in the SERP. On a page ranking in position four or five, a better description can be the difference between a page that quietly earns traffic and one that does not.

---

5. Headings (H1 to H6)

Headings give a page its skeleton. They tell readers how the content is organized, they tell Google what each section is about, and - increasingly important - they let AI engines find and lift the specific passage that answers a question. A clean heading hierarchy is one of the highest-return, lowest-effort things you can fix.

• Use exactly one H1 per page. The H1 is the page's main title as seen by the reader. One H1 states the single topic clearly. Multiple H1s dilute the signal and confuse the hierarchy. (Modern HTML technically allows more than one, but for SEO clarity, use one.)
• Put the primary keyword in the H1. The H1 should closely reflect the page's target query. It can differ from the title tag, but it must be on-topic and keyword-relevant.
• Follow a logical, nested hierarchy. H2s are main sections, H3s are subsections of an H2, H4s subdivide H3s, and so on. Never skip levels for styling reasons - do not jump from H2 to H4 because you liked the font size. Style with CSS; structure with heading levels.
• Use H2s to break the page into scannable sections. Most readers scan headings before they read a word of body copy. Each H2 should make sense on its own and preview what follows.
• Phrase key H2s and H3s as the questions people actually ask. "How long should a title tag be?" as a heading, answered immediately underneath, is exactly the format AI engines and featured snippets reward. This is one of the most important overlaps between classic SEO and answer-engine optimization (section 13).
• Use secondary keywords and semantic terms in subheadings. Headings are a natural place to work in related terms and entities without stuffing the body. They reinforce topical breadth.
• Keep headings descriptive, not clever. "The nuts and bolts" tells Google nothing. "How to write a title tag" tells it everything. Save the wordplay for the body if you must.
• Front-load the meaningful words in each heading. Just like titles, headings are scanned left to right. Lead with the substance.
• Make headings concise. A heading that runs to three lines is a paragraph in disguise. Keep them tight so they work as an outline.
• Ensure the heading outline reads like a table of contents. Read only your headings top to bottom. If that alone tells a coherent story of the page, your hierarchy is right. If it jumps around, restructure.

Why this matters: headings are simultaneously a human usability feature, a Google relevance signal, and the anchor points AI systems use to chunk your content into answerable passages. A page with a clean question-and-answer heading structure is easier to rank, easier to feature in a snippet, and easier to cite in an AI answer - three wins from one edit.

---

6. Content quality and depth

This is the section that decides everything. You can perfect every tag on this list, but if the content does not satisfy the searcher better than the alternatives, it will not hold a ranking. Google's systems, and AI answer engines even more so, are built to reward content that genuinely and completely answers the query from a source the reader can trust. Everything else on this checklist is in service of getting great content seen.

Match search intent precisely

• Identify the dominant intent behind the query. Is the searcher looking to learn (informational), to compare (commercial), to buy or sign up (transactional), or to reach a specific page (navigational)? Look at what currently ranks - Google has already told you what it thinks the intent is.
• Match the content type to the intent. An informational query wants a guide or checklist, not a product page. A "best X" query wants a comparison. If you serve the wrong format, you will not rank regardless of quality. Our guide to why SaaS sites fail to rank covers intent mismatch as a top cause.
• Answer the core question early. Do not bury the answer under 800 words of preamble. Lead with the direct answer (see the TL;DR at the top of this page), then expand.

Be comprehensive without padding

• Cover the topic completely. Comprehensive pages rank because they answer the follow-up questions too. Look at the "People also ask" box, the related searches, and what competing pages cover, then make sure you address every legitimate sub-topic.
• Match the depth the query deserves, not an arbitrary word count. There is no magic length. A definition query needs a paragraph; a "complete checklist" query needs thousands of words. Write until the topic is fully covered and then stop. Length is a byproduct of completeness, never a target in itself.
• Cut the filler. Padding a thin page to hit a word count makes it worse, not better. Every paragraph should add information. AI engines in particular reward information density and punish fluff.

Originality and E-E-A-T

• Add something only you can add. First-hand experience, original data, a real example, a screenshot from your own tool, a genuine opinion. Google's helpful-content systems and AI engines both increasingly discount content that just re-summarizes what is already ranking. This is the "Experience" in E-E-A-T (covered fully in section 12).
• Demonstrate expertise. Get the details right, use the correct terminology, and show you understand the edge cases. Superficial content reads as superficial.
• Be accurate and cite sources. Link to primary sources for statistics and claims. This builds trust with readers, with Google, and with AI engines that check whether claims are supported.
• Avoid unsupported hype. "The ultimate revolutionary game-changing strategy" is a trust-killer. Plain, specific, honest claims win. This is a core reason data-honest content outperforms.

Freshness

• Keep time-sensitive content current. Update statistics, dates, screenshots, and claims that go stale. A "2026" in the title with 2023 data inside destroys trust instantly.
• Show a genuine last-updated date. Update the date only when you have made a substantive change, not as a trick. Fake freshness is transparent and erodes trust.
• Re-audit high-value pages on a schedule. Your best pages deserve a quarterly or biannual review. Rankings decay; refreshes reverse the decay.

Readability and formatting

• Write in plain language. Short sentences, everyday words, one idea per sentence. Most readers are skimming on a phone. Complexity is not sophistication.
• Use short paragraphs. Two to four sentences. Walls of text get abandoned.
• Break content up. Subheadings, bullet lists, numbered steps, tables, and bold key phrases all aid scanning and give AI engines clean chunks to extract.
• Use descriptive formatting for key facts. Put important stats, definitions, and takeaways in their own line or a bolded lead so they are easy to lift.
• Front-load value throughout. In every section, lead with the point, then support it. Do not make readers dig.

Multimedia

• Include relevant images, diagrams, or screenshots. Original visuals increase time on page, aid understanding, and give you image-search and AI-visual surface area. See section 9 for optimizing them.
• Add video where it genuinely helps. A short demo or walkthrough can raise engagement, but only when it serves the topic.
• Caption and label visuals. Captions are read and used; naked images are not.

Avoid thin and duplicate content

• No thin pages. If a page cannot say anything substantial, either build it out, merge it into a stronger page, or noindex it. Thin pages drag down the perceived quality of the whole site.
• No duplicate or near-duplicate pages. Two pages targeting the same intent cannibalize each other. Consolidate them and 301 the weaker one into the stronger.
• No boilerplate-heavy templates. If ninety percent of a page is the same navigation and footer text as every other page, the unique content is too thin.

Why this matters: content quality is the ceiling on everything else. Great fundamentals on mediocre content produce a page that ranks briefly and then fades. Mediocre fundamentals on genuinely excellent content produce a page that climbs anyway. Spend most of your time here.

---

7. Keyword and entity optimization

Modern search does not match strings; it matches meaning. Google and AI engines parse a page for the topics, entities, and relationships it covers, then judge whether it comprehensively addresses the searcher's need. Your job is to make the page's subject unmistakable without ever resorting to mechanical repetition.

• Define one primary keyword per page. Every page should have a single main target query it is built to win. If you cannot name it in one phrase, the page is unfocused.
• Support it with secondary and long-tail keywords. Identify the related terms and questions that share the same intent, and cover them in sections, subheadings, and naturally in the body. This is how one page ranks for dozens or hundreds of variations.
• Place the primary keyword in the high-signal spots. Title tag, H1, URL slug, the first 100 words, at least one H2, and the meta description. These placements are natural if the page is genuinely about the topic - if they feel forced, the page probably is not focused enough.
• Cover the entities, not just the keywords. Entities are the people, products, concepts, and organizations related to your topic. A page about on-page SEO should naturally reference entities like Google Search Console, Core Web Vitals, schema.org, canonical tags, and E-E-A-T. Covering the relevant entities signals genuine topical depth to Google's knowledge systems and to AI engines that map content to concepts.
• Use semantic and NLP-friendly language. Write the way an expert naturally would, using synonyms and related terminology. Google's language models understand that "click-through rate", "CTR", and "clicks per impression" are the same concept - you do not need to repeat the exact phrase.
• Never keyword-stuff. Repeating the exact keyword unnaturally is an old tactic that now actively hurts. It reads as spam to humans and as manipulation to algorithms. If you read a sentence aloud and it sounds robotic, rewrite it.
• Forget keyword density targets. There is no target percentage. Write naturally and comprehensively; density takes care of itself. Chasing a density number produces worse content.
• Use keyword variations in anchor text and headings. Distribute related terms across your subheadings and internal links so the page demonstrates breadth without repetition.
• Map keywords to pages, not the reverse. Maintain one target per page across your site so two pages never fight for the same query (keyword cannibalization). If two pages target the same term, consolidate.
• Answer the adjacent questions. Pull the "People also ask" questions for your keyword and make sure the page addresses them - ideally as question-form subheadings. This widens the range of queries the page can satisfy and feeds directly into AEO.

Why this matters: the goal is not to hit keyword quotas - it is to leave no doubt about what the page is about and to prove you have covered the topic more completely than the competition. Entity coverage and semantic depth are how you signal that in 2026, and they are exactly what AI engines evaluate when deciding whether your page is a reliable source to cite.

---

8. Internal linking

Internal links are the most underused high-leverage lever in on-page SEO, and they are entirely within your control. They pass authority (link equity) between your pages, they define your site's topical structure, and they help both crawlers and AI engines understand which pages are important and how they relate. A deliberate internal linking strategy can lift a whole cluster of pages at once.

• Use descriptive, keyword-relevant anchor text. The clickable text should describe the destination. "Read our anchor text optimization guide" tells Google what the target page is about; "click here" tells it nothing. Internal anchors are one place you can and should be descriptive with keywords - the risks that apply to external anchors do not apply the same way internally.
• Vary anchor text naturally. Do not use the exact same anchor to the same page every time. Natural variation reads better and avoids an unnatural pattern.
• Link from high-authority pages to the pages you want to rank. Your homepage, top blog posts, and most-linked pages have authority to share. Point that authority at your money pages and your priority content deliberately. This is the core of link sculpting - our guide to internal linking and authority sculpting goes deep on the mechanics.
• Support your money pages. Every commercial or conversion page (your services, your pricing, your signup) should receive relevant internal links from your content. A great blog post that never links to the product it relates to is a missed conversion path and a wasted authority flow.
• Build topic clusters. Organize content into a hub-and-spoke model: a pillar page covers a broad topic, and cluster pages cover subtopics, all interlinking. This signals topical authority and helps you own a subject rather than a single keyword. This checklist, for instance, sits in a cluster with pieces on AI search and getting cited by ChatGPT.
• Keep important pages shallow. The fewer clicks from the homepage, the more important a page appears. Do not bury priority pages deep in the structure.
• Eliminate orphan pages. An orphan page has no internal links pointing to it, so crawlers struggle to find it and it inherits no authority. Every indexable page should have at least a few relevant internal links pointing in. Audit for orphans regularly.
• Link contextually from within body content. In-content links carry more weight and relevance than links buried in footers or sidebars. Link where the topic naturally comes up.
• Do not over-link a single page. A page crammed with dozens of links dilutes the value each one passes and overwhelms the reader. Link purposefully.
• Fix broken internal links. A link to a 404 wastes authority and frustrates readers. Audit and repair them.
• Check for links to redirected or noindexed pages. Internal links should point to final, indexable URLs, not through redirect chains or to pages you have excluded from the index.

Why this matters: external links (backlinks) are hard to earn and mostly outside your control. Internal links are free, unlimited, and fully yours to direct. A site that links intelligently between related pages ranks better across the board than one where every page is an island - and it gives AI engines a clear map of your expertise, which helps them decide you are a source worth citing.

---

9. Image and media optimization

Images make content better for humans and open an additional search surface (image search, and increasingly AI visual answers). Unoptimized images are also one of the most common causes of slow pages and layout shift, which hurts the page experience signals in section 11. Optimize every image on every page.

• Write descriptive alt text for every meaningful image. Alt text describes the image for screen-reader users (accessibility, and a legal requirement in many contexts) and tells Google and AI engines what the image shows. Describe the image accurately and specifically; include a keyword only where it fits naturally. Decorative images can use empty alt (alt="") so screen readers skip them.
• Use descriptive, hyphenated file names. on-page-seo-checklist-diagram.webp beats IMG_4821.jpg. The file name is a ranking signal for image search and a small relevance signal for the page.
• Compress every image. Uncompressed images are the number one cause of slow load times. Compress to the smallest size that still looks sharp. This directly improves your Largest Contentful Paint.
• Serve next-generation formats. Use WebP or AVIF instead of JPEG and PNG where supported. They deliver the same visual quality at a fraction of the file size. Provide fallbacks for older browsers if needed.
• Specify width and height attributes (or use CSS aspect-ratio). When the browser knows an image's dimensions before it loads, it reserves the space, preventing the layout from jumping. This is the single biggest fix for Cumulative Layout Shift (section 11).
• Lazy-load below-the-fold images. Use loading="lazy" so off-screen images load only as the user scrolls. This speeds up the initial render. Do not lazy-load your LCP image (the main above-the-fold image) - that one should load immediately, or eagerly.
• Serve responsive images. Use srcset and sizes so mobile devices download appropriately sized images instead of a giant desktop file. This saves bandwidth and improves mobile speed.
• Add captions where they aid understanding. Captions are among the most-read text on a page and give context to both readers and search engines. Use them for diagrams, data, and screenshots.
• Use original visuals where you can. Original screenshots, diagrams, and charts are more citable and more useful than stock imagery, and they support your E-E-A-T and AEO efforts by demonstrating first-hand experience.
• Include images in your image sitemap or ensure they are crawlable. Images loaded only via JavaScript or CSS backgrounds may not be indexed for image search. Use standard <img> tags for content images.

Why this matters: images are a double lever. Done right they add a whole search channel, improve engagement, and support your content's credibility. Done wrong they are the most common reason a page is slow and unstable - which now directly affects rankings through Core Web Vitals.

---

10. Structured data and schema

Structured data is code (best implemented as JSON-LD) that describes your content to search engines in a machine-readable format. It does not directly boost rankings, but it makes your pages eligible for rich results - star ratings, FAQ dropdowns, breadcrumbs, and more - which increase visibility and click-through. Increasingly, it also helps AI engines parse and trust your content. You can validate any page's markup with our free schema checker.

• Use JSON-LD, not microdata or RDFa. JSON-LD is Google's recommended format. It lives in a single <script> block in the head or body, is easy to maintain, and does not tangle with your visible HTML.
• Add Article schema to blog posts and articles. Include the headline, author, publish date, modified date, publisher, and featured image. It reinforces authorship and freshness - both E-E-A-T signals.
• Add FAQPage schema to pages with a genuine FAQ section. When you include a real "Frequently asked questions" section with question headings and answers, FAQPage markup makes it eligible for expanded FAQ rich results and, importantly, gives AI engines clean question-answer pairs to lift. Only mark up FAQs that are actually visible on the page - marking up hidden content violates the guidelines.
• Add BreadcrumbList schema. Breadcrumbs show your site hierarchy in the search result instead of a raw URL, which improves clarity and CTR. They also reinforce your site structure to crawlers.
• Add Organization schema to your homepage. Define your organization's name, logo, official URL, and social/contact profiles. This feeds Google's knowledge graph and helps establish your brand as a recognized entity - which matters enormously for both E-E-A-T and AI citation.
• Add Product schema on product and pricing pages. For SaaS, use it for plans, pricing, and reviews where genuine. It can surface price and rating information directly in results.
• Be cautious with HowTo and other deprecated or reduced markup. Google has scaled back several rich-result types over time (HowTo and FAQ rich results have seen major visibility reductions for many sites). Add the markup where it is still valid and helpful for machine understanding, but do not build your strategy around a rich-result type that Google may show rarely or not at all. Check current documentation before relying on any one type.
• Match schema to visible content. Never mark up content that is not present on the page for users. Google treats mismatched or hidden-content markup as a spam signal and may issue a manual action.
• Validate every implementation. Use the Rich Results Test and Schema Markup Validator, or our schema checker, to confirm the markup is error-free and eligible. Broken schema is worse than none because it wastes the opportunity silently.
• Keep schema in sync with the page. When you update an author, a date, or a price on the page, update the schema too. Stale schema sends contradictory signals.
• Do not over-mark-up. Add schema types that genuinely describe your content. Piling on irrelevant types does not help and can look manipulative.

Why this matters: schema is how you translate your page into the structured language search engines and AI systems speak natively. It will not rescue weak content, but on strong content it wins you more visual real estate in the SERP and makes your page dramatically easier for machines to understand, extract, and cite - which is the whole game in AI search.

---

11. Page experience and Core Web Vitals

Google measures how a page feels to use, and it uses that measurement as a ranking signal, especially as a tie-breaker between pages of similar relevance. The framework is Core Web Vitals plus a handful of related requirements. These are largely technical, so this is the section to hand to your developer - but every content owner should understand the targets.

• Largest Contentful Paint (LCP) under 2.5 seconds. LCP measures how long the largest visible element (usually the hero image or main heading) takes to render. It is the closest metric to "how fast does the page feel to load." Fix it by compressing and preloading the LCP image, reducing server response time, and eliminating render-blocking resources.
• Interaction to Next Paint (INP) under 200 milliseconds. INP replaced First Input Delay as a Core Web Vital. It measures responsiveness - how quickly the page reacts when a user taps or clicks throughout the whole visit, not just the first interaction. Fix it by reducing heavy JavaScript, breaking up long tasks, and deferring non-critical scripts.
• Cumulative Layout Shift (CLS) under 0.1. CLS measures visual stability - how much the page jumps around as it loads. The usual culprits are images without dimensions, ads and embeds that push content down, and web fonts that reflow text. Fix it by setting explicit dimensions on all media (section 9), reserving space for dynamic content, and using font-display sensibly.
• Use real field data, not just lab scores. Lighthouse gives you a lab estimate; the Chrome User Experience Report (CrUX) and the Core Web Vitals report in Search Console show what real users actually experience. Optimize against field data.
• Be mobile-first. Google indexes and ranks based on the mobile version of your page. It must be fully responsive, with readable text without zooming, tap targets that are not cramped, and no content hidden or broken on small screens. Test on real devices, not just a resized desktop window.
• Serve the page over HTTPS. Secure connection is a baseline requirement and a light ranking signal. Ensure a valid certificate, no mixed content (HTTP resources on an HTTPS page), and that the HTTP version redirects to HTTPS.
• No intrusive interstitials. Pop-ups that cover the main content immediately on load - especially on mobile - are a demotion signal. Cookie and legal notices are allowed, but keep them minimal and non-obstructive. Do not hit a reader with a full-screen newsletter or discount modal before they have read a sentence.
• Minimize and defer JavaScript. Heavy JS is the most common cause of poor INP and slow LCP. Ship less of it, defer what you can, and code-split so each page loads only what it needs.
• Enable compression and caching. Gzip or Brotli compression and sensible cache headers reduce load times for repeat visits and are quick wins.
• Reduce third-party scripts. Every analytics tag, chat widget, and marketing pixel adds weight and can hurt INP. Audit them and remove what you do not need.

Why this matters: page experience rarely makes a slow, unhelpful page rank, but it regularly decides which of two equally helpful pages wins. And beyond rankings, it directly affects conversions - a slow, jumpy page loses signups regardless of where it ranks. For a SaaS site, that is money left on the table twice.

---

12. E-E-A-T on-page signals

E-E-A-T stands for Experience, Expertise, Authoritativeness, and Trustworthiness. It is not a single score Google assigns - it is a framework its systems and human quality raters use to judge whether content comes from a credible source. It matters everywhere and it matters most on "Your Money or Your Life" topics. In 2026 it also matters enormously for AI engines, which are built to prefer trustworthy sources. Much of E-E-A-T is earned off-page, but there is a lot you can and should signal directly on the page.

• Show a real author with a real byline. Every content page should name its author, not "admin" or "the team." A named, accountable author is a baseline trust signal.
• Add a substantive author bio. Link the byline to a bio that establishes why this person is qualified to write on the topic - their experience, credentials, and relevant background. This supports both Expertise and Experience.
• Demonstrate first-hand experience in the content. Show that you have actually done the thing. Real screenshots, original data, specific examples, and honest observations ("in our own testing we found...") are the "Experience" signal Google added specifically to reward content grounded in real use rather than rephrased research.
• Cite authoritative sources. Link out to primary sources, official documentation, and credible research for your claims and statistics. Linking to good sources is a trust signal, not a leak of authority.
• Have a clear, complete About page. Explain who is behind the site, your mission, and your track record. AI engines and users both check who they are dealing with.
• Have an accessible Contact page with real details. A real business address, contact method, and company identity (for us, Serpbays LLC) signal a legitimate operation, not a fly-by-night content mill.
• Be transparent and honest. State limitations, acknowledge trade-offs, and avoid overclaiming. Data-honest content that admits what it does not know is more trusted than content that pretends to certainty. This is a genuine competitive advantage in a web full of hype.
• Keep content accurate and current. Trust erodes fast when readers spot outdated or wrong information. Maintain your pages (see freshness in section 6).
• Display trust signals where relevant. Genuine reviews, testimonials, certifications, security badges, and clear policies (privacy, terms, refund) all reinforce trustworthiness, especially on commercial pages.
• Make it easy to verify your claims. Show your work. If you say a tactic produced a result, show the data. Verifiability is the strongest trust signal there is.

Why this matters: as AI-generated content floods the web, the differentiator is no longer information (everyone has that) but trust. Google and AI engines are both explicitly moving toward rewarding sources that demonstrate real experience and accountability. On-page E-E-A-T signals are how you prove, on the page itself, that a real, qualified, accountable source stands behind the content.

---

13. AEO and GEO: optimizing for answer and generative engines

This is the section that separates a 2026 checklist from a 2022 one. Search is no longer only a list of blue links. A large and growing share of queries are answered directly - by Google's AI Overviews, by ChatGPT with web search, by Perplexity, by Gemini, and by Microsoft Copilot pulling from the Bing index. Optimizing for these is a distinct discipline, and it is where the biggest opportunity currently sits because most sites have not adapted. If you want the deeper strategy, our GEO playbook for ranking in AI search and our guide to getting cited by ChatGPT go further; this is the on-page checklist.

What AEO and GEO are, and how they differ from classic SEO

• AEO (Answer Engine Optimization) is optimizing so that answer engines - AI Overviews, voice assistants, featured snippets - extract a direct answer from your page and present it, ideally citing you. The unit of success is the answer, not the ranking position.
• GEO (Generative Engine Optimization) is optimizing so that generative AI systems (ChatGPT, Perplexity, Gemini, Copilot) reference, quote, and cite your brand and content when they compose an answer. The unit of success is the citation and the brand mention.
• How they differ from classic SEO: classic SEO optimizes to win a ranked position that a human then clicks. AEO and GEO optimize to be the source a machine selects, synthesizes, and cites - often without a click at all. Classic SEO rewards the best page for a query; answer engines reward the most extractable, trustworthy, and clearly-stated passage. The good news: the two overlap heavily. Almost everything earlier in this checklist helps. But there are specific moves that tip a page from rankable to citable.

How AI engines select and cite content

AI engines do not read a page like a human. They retrieve candidate content, break it into passages (chunks), and select the passages that most directly and confidently answer the query from a source they assess as trustworthy. They favor content that: states the answer plainly and early, is well-structured, is backed by data and citations, comes from a recognized entity, and is fresh. They also lean heavily on sources that are already frequently mentioned and cited across the web - which is why brand authority and backlinks feed into AI visibility too (we cover that in do AI Overviews use backlinks and backlinks for AEO, GEO and SEO).

The on-page AEO/GEO checklist

• Write liftable direct-answer passages. For each key question your page addresses, provide a self-contained answer in the first sentence or two under the relevant heading - a passage that makes complete sense if lifted out and quoted alone. This is the single most important AEO tactic. The bolded TL;DR at the top of this page is a deliberate example.
• Lead the whole page with a TL;DR direct answer. Put a concise, complete answer to the page's main question at the very top, before any preamble. Answer engines frequently pull from the opening.
• Use question-form headings. Phrase key H2s and H3s as the exact questions users and AI systems ask ("How is AEO different from SEO?"), then answer immediately underneath. This maps your content directly onto the query-answer structure AI engines look for.
• Add a real FAQ section with FAQPage schema. A "Frequently asked questions" section with question headings and concise, self-contained answers gives AI engines clean, pre-chunked question-answer pairs - the easiest possible format to extract and cite. Mark it up with FAQPage schema (section 10).
• Optimize at the passage level (chunking). Make each section and paragraph independently coherent. AI systems retrieve chunks, not whole pages, so a paragraph that only makes sense in context of the previous three is hard to cite. Front-load each chunk with its point.
• Include specific statistics, numbers, and dates. AI engines strongly prefer to cite concrete, verifiable data over vague claims. "Bing's index feeds Copilot citations" is weaker than a specific, sourced figure. Add real numbers and cite where they come from.
• Cite authoritative sources inline. Content that references credible sources is judged more trustworthy and is more likely to be selected. It also signals that your claims are verifiable.
• Establish entity clarity. Make it unambiguous who you are and what you are an authority on. Use Organization schema, a consistent brand name, a strong About page, and consistent descriptions of your brand across the web. AI engines map content to entities; a clear entity is easier to cite by name.
• Build brand mentions across the web. GEO rewards brands that are frequently mentioned, not only linked. Unlinked brand mentions in relevant contexts feed the models' sense that you are a known, credible source. This is off-page work that pays off in on-page citability.
• Keep content fresh and dated. AI engines with web access prefer recent sources for anything time-sensitive. A genuine, recent last-updated date and current data raise your odds of being the source chosen for a live query.
• Structure for extractability. Bullet lists, numbered steps, comparison tables, and definition-style sentences ("X is...") are all formats AI engines find easy to parse and reproduce. Use them wherever the content naturally fits.
• Make sure AI crawlers can access the page. Confirm you are not blocking the major AI crawler user-agents in robots.txt unless you have made a deliberate choice to. A page an AI engine cannot fetch cannot be cited. Weigh this against your own content-licensing stance.
• Answer the full question and the obvious follow-ups. AI engines synthesize complete answers. A page that anticipates and answers the natural next questions is more useful as a single source and more likely to be cited across a multi-part answer.
• Be genuinely accurate and honest. AI engines are increasingly tuned to avoid citing unreliable sources. Overclaiming and thin accuracy get you filtered out. Data-honest content is not just ethical here, it is strategically optimal.

An honest note on llms.txt

You will hear that you should publish an llms.txt file - a proposed standard for a plain-text file telling AI models how to use your content. Be clear-eyed about it: as of 2026, there is no evidence the major AI engines actually use llms.txt to select or cite content, and Google's own 2026 guidance on AI features effectively dismisses it, comparing its lack of adoption to the long-ignored keywords meta tag. Publishing one is low-cost and harmless, so there is no strong reason not to, but do not mistake it for a ranking or citation lever. It is not one today. Spend your effort on the items above - liftable answers, structure, entity clarity, freshness, and trust - because those are what actually get content selected by AI engines right now. When you read a checklist that puts llms.txt near the top of AEO, treat that as a signal the author has not tested it against real crawler behavior.

Why this matters: answer engines are absorbing an ever-larger share of query volume, and they change the payoff of on-page work. The pages that win are not necessarily the ones ranked number one - they are the ones written so a machine can confidently lift a clean, trustworthy answer and attribute it to you. Most of your competitors are still writing for blue links only. Writing for both is the current edge.

---

Quick reference checklist

Use this condensed version as a pre-publish gate. Every item is expanded in the sections above.

Crawlability and indexability

• 200 status, no redirect chains
• Not blocked in robots.txt (unless intended)
• Correct meta robots (index, follow), no accidental noindex
• Self-referencing canonical to a clean, live URL
• URL present in the XML sitemap and reachable via internal links
• Rendered content matches source; no conflicting signals

URL structure

• Short, lowercase, hyphenated, keyword-relevant
• No unnecessary parameters, no dates in evergreen URLs
• Logical folder structure, shallow click depth

Title tag

• Keyword front-loaded, 50 to 60 characters, unique
• Matches intent, written for the click, brand where it fits
• No stuffing, no ALL CAPS, will not be rewritten

Meta description

• Unique, 150 to 160 characters, keyword included
• Compelling, accurate, soft CTA

Headings

• One H1 with the primary keyword
• Logical H2 to H6 hierarchy, no skipped levels
• Key headings phrased as questions and answered underneath

Content quality

• Matches dominant intent and content type
• Comprehensive, answer early, no filler
• Original insight, first-hand experience, accurate, cited
• Fresh with a genuine update date
• Readable: short paragraphs, lists, formatting, multimedia
• No thin, duplicate, or cannibalizing pages

Keywords and entities

• One primary keyword per page, placed in high-signal spots
• Secondary and long-tail terms covered naturally
• Relevant entities and semantic terms present
• No stuffing, no density targets, no cannibalization

Internal linking

• Descriptive, varied anchors
• Authority flowing to money pages and priority content
• Topic clusters interlinked, no orphan pages
• No broken links or links to redirected/noindexed pages

Images and media

• Descriptive alt text and file names
• Compressed, next-gen format (WebP/AVIF)
• Width/height set, lazy-load below the fold (not the LCP image)
• Responsive srcset, captions where useful, original visuals

Structured data

• JSON-LD, matched to visible content
• Article, FAQPage, BreadcrumbList, Organization, Product as relevant
• Validated and in sync with the page

Page experience

• LCP under 2.5s, INP under 200ms, CLS under 0.1 (field data)
• Mobile-first, HTTPS, no intrusive interstitials
• Minimal JS and third-party scripts, compression and caching on

E-E-A-T

• Named author with a real bio
• First-hand experience shown, sources cited
• About and Contact pages complete
• Honest, accurate, verifiable claims and trust signals

AEO and GEO

• Liftable direct-answer passages and a top TL;DR
• Question-form headings, real FAQ with FAQPage schema
• Passage-level coherence (chunking), stats and citations
• Entity clarity, brand mentions, freshness, extractable formatting
• AI crawlers not blocked; do not rely on llms.txt

Once a page clears this gate, the on-page work is done. From here, the lever is authority - see our guides on types of link building and on-page vs off-page SEO. You can check any page's current backlink profile with our free backlink checker, and when you are ready to build links that support your best pages, create a free account.

---