On-page SEO is the foundation that determines whether search engines — and now AI platforms — can find, understand, and cite your content. Every page you publish either earns visibility or wastes the effort you put into writing it, and the difference usually comes down to a handful of optimisations that take minutes to implement but months to recover from when skipped.
This checklist covers every on-page task that matters in 2026, organised by category. It includes the traditional fundamentals that still drive Google rankings and the AI visibility signals that most checklists ignore — because in 2026, optimising for AI search engines is no longer optional. Sites with properly implemented structured data get cited in AI responses 3.2 times more often than those without.
Key Takeaways
- Sites with properly implemented structured data get cited in AI responses 3.2 times more often than those without.
- Pages using 120 to 180 words between headings receive 70% more ChatGPT citations than pages with sections under 50 words.
- Meta descriptions influence click-through rates by 20 to 30% without changing ranking position, and AI systems parse them when evaluating whether to read further.
- On-page SEO in 2026 serves two audiences simultaneously: search algorithms and AI models — the same fundamentals matter for both, but AI adds a citability layer most checklists miss.
- 62.73% of all web traffic comes from mobile devices, making mobile-first design a baseline requirement rather than an optimisation.
Keyword and Topic Foundation
Every on-page optimisation starts with knowing what you're optimising for. Get the foundation wrong and nothing else matters.
Choose a primary keyword with clear intent. Use tools like Google Keyword Planner, Ahrefs, or Semrush to identify a primary keyword with meaningful search volume and intent that matches your page's purpose. Informational, commercial, and transactional queries each require different page structures.
Identify secondary and related keywords. Map 3–5 closely related terms that support your primary keyword. These aren't stuffing targets — they're semantic signals that help both search algorithms and AI models understand the full scope of your topic.
Research the prompts, not just the keywords. Traditional keyword research tells you what people type into Google. Prompt research tells you what people ask ChatGPT, Perplexity, and Gemini. In 2026, both matter. Check how your target topic gets discussed in AI platforms and structure your content to answer those conversational queries directly.
Analyse the top-ranking pages. Before writing, review the top 5 results for your primary keyword. Note their structure, depth, and angle. Your content needs to match their quality baseline and add something they don't cover — Google's algorithms now prioritise "Information Gain," and AI models favour sources that contribute original insights.
Title Tags and Meta Data
Title tags and meta descriptions are your first impression in search results and the metadata AI systems parse before deciding whether to read further.
Write a title tag under 60 characters. Front-load your primary keyword. "On-Page SEO Checklist: Complete Task List for 2026" works because the keyword appears first and the title communicates scope. Avoid vague titles that force searchers to guess what the page covers.
Write a meta description under 155 characters. Meta descriptions aren't a direct ranking signal, but they influence click-through rates by 20–30% without changing your ranking position. Include your primary keyword, state the specific value the page delivers, and end with an implicit call to action. "A 20-point on-page SEO checklist covering..." tells searchers exactly what they'll get.
Set a clean, keyword-rich URL slug. Use hyphens to separate words. Keep it descriptive but brief — /on-page-seo-checklist-2026 outperforms /the-complete-on-page-seo-checklist-you-need-for-2026. Skip dates that will become stale unless the year is part of the value proposition.
Use exactly one H1 tag per page. Your H1 should include the primary keyword and clearly describe the page content. It doesn't need to match your title tag exactly, but it should be closely aligned. AI agents use the H1 as a primary signal for understanding page topic.
Content Structure and Quality
Content structure determines whether humans read your page and whether AI systems can extract citable information from it. Pages that use 120–180 words between headings receive 70% more ChatGPT citations than pages with sections under 50 words.
Lead every section with a direct answer. AI agents extract the first clear statement that answers a question. Start each H2 section with a concise, factual answer before expanding with context, examples, and nuance. This pattern serves both human readers scanning your page and AI models deciding whether to cite it.
Use descriptive, question-based H2 headings. Write headings that mirror how people search. "How Do Title Tags Affect AI Visibility?" outperforms "Title Tags" because it matches both search queries and conversational AI prompts. Each heading should contain a secondary keyword naturally.
Write self-contained sections. Each H2 should make complete sense independently. AI agents cite individual sections, not full articles — a section about internal linking should define the concept, explain why it matters, and give the practical step, all within that section.
Include specific data and numbers. "62.73% of all web traffic comes from mobile devices" is citable. "Most traffic is mobile" is not. Every factual claim with a specific number becomes a potential citation for AI systems building their responses.
Define terms inline the first time you use them. Don't assume readers — or AI agents — know industry jargon. "Schema markup — the structured code that helps search engines and AI agents understand your page content — is foundational to visibility" serves both audiences simultaneously.
Internal and External Links
Links are how search engines discover your site structure and how AI systems map relationships between your content.
Add 3–5 internal links per page. Link to related content using descriptive anchor text. "See our AI citation playbook" is effective. "Click here" wastes an opportunity to signal topical relevance. Distribute links to your most important pages — each internal link passes authority and helps search engines understand your content hierarchy.
Link to authoritative external sources. Cite credible sources for data and claims. External links to authoritative domains signal trustworthiness to both search algorithms and AI models. Google's E-E-A-T framework rewards pages that reference established expertise rather than making unsourced claims.
Fix broken links regularly. Broken links damage user experience, waste crawl budget, and signal neglect to search engines. Run a crawl check quarterly at minimum — tools like Screaming Frog or Ahrefs Site Audit catch broken links before they accumulate.
Use descriptive anchor text, never generic phrases. "Learn more about generative engine optimisation" gives search engines and AI platforms explicit context about the linked page. "Learn more [here]" gives them nothing.
Image Optimisation
Images affect page speed, accessibility, and search visibility across both Google Image Search and AI platforms.
Add descriptive alt text to every image. Alt text serves accessibility, SEO, and AI comprehension. Describe the image's content and meaning — "on-page SEO checklist showing technical optimisation tasks" tells search engines and AI agents what the image contributes to the page.
Compress images before uploading. Unoptimised images are the most common cause of slow page loads. Use tools like TinyPNG, Squoosh, or your CMS's built-in compression. Target under 200KB for in-content images and under 300KB for hero images without visible quality loss.
Use descriptive file names. on-page-seo-checklist-hero.jpg outperforms IMG_4592.jpg for both search indexing and AI understanding. Include your target keyword where it fits naturally.
Serve images in modern formats. WebP and AVIF formats deliver 25–50% smaller file sizes than JPEG at equivalent quality. If your CMS supports it, serve modern formats with JPEG fallbacks for older browsers.
Schema Markup and Structured Data
Structured data is the single highest-impact optimisation that most websites still skip. It gives search engines and AI agents machine-readable context about your content — who you are, what the page covers, and how information relates to your broader site.
Implement Article schema on all editorial content. Include headline, datePublished, dateModified, author, and publisher fields. This schema type directly maps to the metadata AI systems use when evaluating whether to cite a source. For a complete list of the schema types that impact AI visibility, see our AI visibility checklist.
Add Organisation schema to your homepage. Include name, description, URL, logo, and social profiles. This establishes your brand as a known entity — a signal that matters for both Knowledge Graph inclusion and AI citation decisions.
Use FAQ schema for question-and-answer content. FAQ schema maps directly to the query patterns AI engines handle. When ChatGPT or Google AI Overview encounters a question matching your FAQ, structured data makes extraction trivial rather than inferential.
Validate your implementation. Use Google's Rich Results Test or Schema.org Validator to confirm your structured data parses correctly. Malformed schema is worse than no schema — it introduces conflicting signals. SwingIntel's AI Readiness Audit checks 6 structured data signals specifically because they account for a disproportionate share of AI citation decisions.
Technical Signals
Technical performance determines whether search engines can crawl your pages efficiently and whether users stay long enough for your content to matter.
Optimise Core Web Vitals. Google's three performance metrics remain ranking factors in 2026:
- Largest Contentful Paint (LCP) — main content should load within 2.5 seconds
- Interaction to Next Paint (INP) — pages should respond to user input within 200 milliseconds
- Cumulative Layout Shift (CLS) — visual stability score should stay below 0.1
Ensure mobile responsiveness. With 62.73% of web traffic coming from mobile devices, mobile-first design isn't a recommendation — it's a baseline requirement. Test every page on actual mobile devices, not just browser dev tools.
Check your robots.txt and crawlability. Verify that your robots.txt at yourdomain.com/robots.txt allows both search engine crawlers and AI agents to access your content. Blocking Googlebot or AI-specific crawlers (GPTBot, ClaudeBot, PerplexityBot) cuts off visibility at the source.
Implement HTTPS across your entire site. HTTPS is a confirmed ranking signal and a trust requirement for AI systems evaluating source credibility. Mixed content warnings — where some resources load over HTTP — can trigger security flags that reduce both rankings and AI citation likelihood.
Minimise render-blocking resources. JavaScript and CSS that block initial rendering slow LCP and frustrate users. Defer non-critical scripts, inline critical CSS, and lazy-load below-the-fold images. Around 35% of domains have speed issues traceable to render-blocking resources alone.
The AI Visibility Layer Most Checklists Miss
Traditional on-page SEO checklists stop at Google. In 2026, that's only half the picture. AI platforms like ChatGPT, Perplexity, Gemini, and Claude now process a growing share of search queries, and the on-page signals that earn their citations overlap with — but extend beyond — standard SEO.
Write citable statements. AI agents need declarative sentences they can confidently extract. "On-page SEO covers 14 distinct optimisation categories" is citable. "On-page SEO covers many areas" is not. Embed these throughout your content, not just in introductions.
Build topical authority with content clusters. Single pages rank. Content clusters earn citations. When AI systems see 10 interlinked pages covering different aspects of your topic, they assign your domain higher topical authority — making every page in the cluster more likely to be cited. Our guide on creating content for AI search engines covers the clustering strategy in detail.
Keep content current with visible dates. AI models check publication dates. A page dated 2023 competes poorly against a 2026 update covering the same topic. Add datePublished and dateModified in your Article schema, and display the date visibly on the page.
Test your actual AI visibility. You can follow every item on this checklist and still not know whether AI platforms actually cite your content. The only way to know is to test — run a free scan to see your AI Readiness Score across the signals that determine whether ChatGPT, Perplexity, and Gemini reference your brand.
Frequently Asked Questions
What is the single most important on-page SEO task in 2026?
Implementing structured data (JSON-LD schema markup) is the single highest-impact optimisation that most websites still skip. It gives search engines and AI agents machine-readable context about your content. Sites with properly implemented structured data get cited in AI responses 3.2 times more often than those without.
How does on-page SEO differ for AI search engines versus Google?
Traditional SEO rewards page authority and keyword relevance. AI search adds a citability layer — AI agents need self-contained sections, declarative factual sentences, and structured data they can confidently extract and quote. Pages using 120 to 180 words between headings receive 70% more ChatGPT citations than pages with very short sections.
How many internal links should each page have?
Add 3 to 5 internal links per page using descriptive anchor text. Link to related content that signals topical relevance — "See our AI citation playbook" is effective, while "Click here" wastes an opportunity. Each internal link passes authority and helps search engines understand your content hierarchy.
On-page SEO in 2026 isn't harder than it was in 2024. It's broader. The same fundamentals still matter — clean title tags, structured content, fast pages, descriptive links. The difference is that these fundamentals now serve two audiences: search algorithms and AI models. The checklist above covers both. Work through it page by page, starting with your highest-traffic content, and you'll build visibility across every surface where your audience searches.
