On-Page SEO Checklist 2026: 38 Checks to Rank Higher in Google
On-page SEO is everything you do directly on a page to improve its ranking potential — from the title tag to the last internal link. This checklist covers 38 checks across six categories, updated for 2026 to reflect current Google quality guidelines including E-E-A-T, Helpful Content, and AI search visibility.
Technical SEO gets your site crawled and indexed. Off-page SEO builds authority through links and mentions. On-page SEO is where everything comes together: it's the difference between a page Google trusts and ranks versus one it ignores despite adequate authority.
Work through this checklist page by page on your most important content. Prioritise HIGH items first — a single broken title tag structure can suppress an entire section of your site.
Quick Reference: Checklist Categories
| Category | Checks | Primary Tool |
|---|---|---|
| 1. Title Tags & Meta Descriptions | 6 | Screaming Frog, Google Search Console |
| 2. Heading Structure & URL | 6 | Screaming Frog, browser DevTools |
| 3. Content Quality & E-E-A-T | 9 | Manual review, Semrush Content Score |
| 4. Internal Linking | 5 | Screaming Frog, Ahrefs Site Audit |
| 5. Images & Media | 6 | Screaming Frog, PageSpeed Insights |
| 6. Schema Markup & Structured Data | 6 | Rich Results Test, Schema.org |
1. Title Tags & Meta Descriptions
Title tags and meta descriptions are the primary signals Google uses to understand page topic, and the primary content your clicks come from in search results. Getting these right is table-stakes on-page SEO.
- Title tag includes the primary keyword, preferably near the beginning. Google weights words at the start of the title more heavily. "Keyword: Brand Name" typically outperforms "Brand Name | Generic Category | Keyword." Aim to put your main target keyword in the first 3–4 words where natural.
- Title tag length is 50–60 characters (approximately 560px). Titles longer than 60 characters get truncated in SERPs on most screens. Use AuditDepot or Screaming Frog to bulk-check title lengths. Short titles (under 30 characters) often under-specify the page topic.
- Each page has a unique title tag. Duplicate title tags cause Google to choose between competing pages arbitrarily. Run a Screaming Frog crawl → filter "Duplicate" → resolve all duplicates before any other optimisation work.
- Meta description is 140–160 characters and includes a call to action. Meta descriptions don't directly affect rankings, but a well-written description with a clear CTA improves click-through rate — which is a meaningful signal. Include your primary keyword and one secondary keyword where natural. End with an action phrase: "Learn how," "Get the checklist," "Compare now."
- Meta description is unique per page. Like duplicate titles, duplicate meta descriptions indicate thin or duplicated content. Google often rewrites descriptions that don't match query intent — but having a good default still matters for unrecognised queries.
- Title tag avoids keyword stuffing and reads naturally. "Best Plumber London | Emergency Plumber London | 24hr Plumber London" is keyword stuffing. It's also penalised. Write for the reader first, then check that your keyword is present and near the front.
2. Heading Structure & URL
Headings provide semantic structure that both search engines and readers use to understand page content. URL structure affects crawlability, link equity, and user trust.
- Page has exactly one H1 tag that includes the primary keyword. Multiple H1s confuse crawlers about the primary topic. Missing H1s are a missed ranking signal. The H1 should closely match (but not necessarily be identical to) the title tag, and should include your primary keyword.
- H2s create a logical document outline. H2s should represent the main sections of the page. A reader skimming only the H2s should understand what the page covers. Each H2 is an opportunity to target a related keyword or question variant.
- H3s and below are used for sub-sections, not decoration. Heading tags have semantic meaning. Using H3 for a pull quote or styled callout block creates structural noise. Use CSS classes for visual styling; reserve heading tags for document structure.
- URL is short, lowercase, hyphen-separated, and keyword-rich. Best practice:
/blog/on-page-seo-checklistnot/blog/post?id=1234&category=seo. Remove stop words (a, the, of) where they don't add meaning. Never use underscores — Google treats hyphens as word separators, underscores as word joiners. - URL does not contain unnecessary parameters or dynamic segments. Query parameters (
?ref=,?utm_source=) in URLs can create duplicate content if not properly handled. Canonical tags or parameter exclusion in Google Search Console are the fix. - URL depth is three levels or fewer. Deep URL hierarchies (e.g.,
/blog/category/subcategory/year/month/slug) dilute link equity and increase crawl depth. Keep important content within 3 clicks of the homepage.
3. Content Quality & E-E-A-T
Google's quality evaluator guidelines and the Helpful Content system both emphasise Experience, Expertise, Authoritativeness, and Trustworthiness (E-E-A-T). These nine checks address the content signals that matter most in 2026.
- Content fully covers the search intent for the target keyword. Intent has four types: informational (how to), navigational (brand), transactional (buy), and commercial investigation (best, compare). Your content format and depth must match the intent. A how-to article doesn't rank for transactional keywords; a product page doesn't rank for informational queries. Check the top 3 ranking pages for your target keyword to calibrate expected format and depth.
- Content is substantially original — not paraphrased from top-ranking pages. Google's Helpful Content system specifically targets pages that aggregate information from other sources without adding original value. Include first-hand experience, original data, primary research, expert commentary, or unique framing wherever possible.
- Content demonstrates first-hand experience where relevant. The "E" for Experience is a 2022 addition to the E-E-A-T framework. For product reviews, service comparisons, and how-to guides, Google now explicitly favours content from someone who has done the thing. Screenshots, specific data points, named examples, and personal anecdotes all signal experience.
- Author is identified with a credible byline. Anonymous content is a trust signal weakness on YMYL (Your Money Your Life) topics — health, finance, legal. Include author name, title, and a brief bio. Linking to an author page with their full credentials strengthens the signal further.
- Content is up to date — dates are accurate and outdated claims are removed. Stale content with incorrect dates is a quality signal. "2023 Best Practices" pages that haven't been updated in three years are candidates for content pruning or update. Add a "Last updated" date and review content annually at minimum.
- Claims are supported with sources and links to authoritative references. Citing sources adds trust and creates helpful context for readers. Link to primary sources (government sites, academic papers, original research) rather than secondary aggregators where possible.
- Content length matches the complexity of the topic. There is no universal "correct" word count. Simple FAQ topics may need 500 words; comprehensive technical guides may need 3,000+. The right length is whatever it takes to fully answer the query without padding. Check competitor page lengths for calibration — but match intent, not word count.
- No keyword stuffing — natural language, not repetitive insertion of target phrases. Keyword density as a metric is outdated. Modern Google uses semantic understanding; stuffing a keyword 20 times signals low quality, not high relevance. Write for readers; use keyword variants, synonyms, and related terms naturally.
- Content is clearly written for humans, not for search engines. The Helpful Content system penalises pages that seem to be written "for search engines" — with unnatural keyword insertion, awkward phrasing, or content that satisfies no genuine reader need. Read your page aloud; if it sounds robotic, rewrite it.
4. Internal Linking
Internal links distribute page authority throughout your site, help Google understand your site structure, and create paths for readers to discover related content. Most sites dramatically under-invest in internal linking.
- Important pages receive internal links from other relevant pages. A page with no internal links pointing to it (an orphan page) gets crawled infrequently and is unlikely to rank well regardless of its quality. Every important page should have at least 3–5 relevant internal links pointing to it from elsewhere on the site.
- Anchor text is descriptive and includes relevant keywords. "Click here" and "read more" are wasted anchor text opportunities. Use descriptive anchors: "our guide to technical SEO" or "competitor pricing changes" tell Google and the reader what the linked page is about.
- No broken internal links (404 errors on internal targets). Broken internal links waste crawl budget and create a poor user experience. Run a Screaming Frog crawl monthly and fix all broken links. When pages are deleted or URLs change, always set up 301 redirects.
- Internal link structure reflects site hierarchy and content relationships. Links from high-authority hub pages (like a category page or pillar article) pass more equity than links from deep, low-authority pages. Make sure your most important target pages are linked from high-authority parents, not just peer pages.
- Each page links out to 3–10 relevant internal pages. Too few internal links means missed equity distribution; too many can dilute signals. For a long-form article, 5–10 contextual internal links are typical. For a short page, 3–5 may be appropriate.
5. Images & Media
Images affect page speed, accessibility, and keyword relevance. These six checks ensure your media assets support rankings rather than undermining them.
- All images have descriptive alt text that includes relevant keywords where natural. Alt text serves two purposes: accessibility (screen readers) and SEO (Google Images, page relevance signals). Describe the image content specifically: "graph showing organic traffic growth from January to April 2026" beats "chart" or leaving it blank. Don't stuff multiple keywords into alt text.
- Images are compressed and served in next-gen formats (WebP, AVIF). Uncompressed images are one of the most common causes of poor Core Web Vitals scores. Convert images to WebP or AVIF, compress to the smallest file size that maintains acceptable visual quality. Tools: Squoosh, ImageOptim, or automated CDN compression.
- Large images use lazy loading. Images below the fold should use
loading="lazy"to defer loading until the user scrolls near them. The hero image (above the fold) should NOT use lazy loading — it should load immediately to avoid LCP delay. - Image dimensions are specified in HTML (width and height attributes). Including explicit width and height prevents layout shift as images load, improving Cumulative Layout Shift (CLS) — a Core Web Vitals metric. Use the actual rendered dimensions, not the file dimensions.
- Images add genuine value — not just stock photos used as filler. Google's quality guidelines specifically mention that filler images with no informational value are a quality signal weakness. Use original screenshots, diagrams, data visualisations, and genuine product photos. Avoid generic stock photos that add no context.
- Videos are hosted externally (YouTube/Vimeo) and embedded, not self-hosted. Self-hosted video files dramatically increase page load times and bandwidth costs. Embed from YouTube or Vimeo; use a lightweight facade (thumbnail + play button) that loads the full player only on click to preserve LCP scores.
6. Schema Markup & Structured Data
Structured data (schema.org markup) helps Google understand your content and enables rich results in search — FAQ dropdowns, How-To steps, review stars, breadcrumbs, and more. It's also increasingly important for AI Overviews and answer engine visibility.
- Article pages use Article or BlogPosting schema with headline, datePublished, author. This is the baseline for informational content. Include
headline,datePublished,dateModified,author(with name and optionally sameAs links to Wikipedia/social profiles), andpublisher. - FAQ pages or sections with Q&A content use FAQPage schema. FAQPage schema can produce expandable FAQ dropdowns in Google results, significantly increasing click-through rate. Eligible when a page contains a list of questions with individual answers. Validate with Google's Rich Results Test.
- How-to content uses HowTo schema with step-by-step structure. Step-by-step guides eligible for HowTo schema can show step names (and optionally images and durations) as rich results. Each step should have a
nameandtext; images are optional but improve display. - BreadcrumbList schema matches the actual URL structure and breadcrumb navigation. Breadcrumb schema replaces the URL in SERP display with a more readable path (e.g., "Home > Blog > On-Page SEO"), improving CTR. It must match the actual breadcrumb navigation on the page.
- Product pages use Product schema with price, availability, and aggregate rating. For e-commerce and SaaS pricing pages, Product schema enables price display and review stars in search results — among the highest-CTR rich results available.
- All schema is validated with Google's Rich Results Test and has no errors. Invalid schema (wrong property types, missing required fields) is silently ignored by Google — you don't get an error, you just don't get the rich result. Test every schema implementation at search.google.com/test/rich-results before publishing.
On-Page SEO Priority Matrix
If you're working through a large site and need to triage, here's where to start:
| Priority | Action | Impact |
|---|---|---|
| 1 | Fix missing or duplicate title tags | High — directly affects ranking signals and CTR |
| 2 | Ensure each important page has a single, keyword-rich H1 | High — primary on-page keyword signal |
| 3 | Audit content for search intent match | High — mismatched intent can't rank regardless of other factors |
| 4 | Add internal links to orphan pages | High — orphan pages rarely rank |
| 5 | Add descriptive alt text to all images | Medium-High — accessibility + image search + relevance signal |
| 6 | Implement Article + FAQPage schema on eligible pages | Medium — enables rich results, improves AI search visibility |
| 7 | Compress and convert images to WebP/AVIF | Medium — affects Core Web Vitals and rankings |
| 8 | Add E-E-A-T signals (author bylines, sources, original research) | Medium — particularly important for YMYL topics |
Frequently Asked Questions
What is on-page SEO?
On-page SEO refers to all the optimisations made directly on a web page — including title tags, meta descriptions, headings, content quality, internal links, images, and schema markup — to help it rank higher in search results. It's distinct from off-page SEO (backlinks, brand mentions) and technical SEO (crawlability, site speed, indexation).
What is the most important on-page SEO factor in 2026?
Content quality and E-E-A-T signals are the most important on-page factors in 2026. Google has increasingly shifted ranking weight toward original, experience-based content that genuinely helps readers — and away from pages that optimise technical signals without adding real value. That said, title tags, intent match, and internal linking are foundational — they don't move the needle alone but their absence suppresses rankings that good content would otherwise earn.
How long should a title tag be for SEO?
50–60 characters (approximately 560 pixels wide on desktop). Title tags longer than 60 characters are truncated with "..." in search results, which typically reduces CTR. Use a SERP simulator or Screaming Frog to check your titles at display width, not just character count — pixel width is what actually matters.
How often should I update on-page SEO?
Review your highest-traffic pages quarterly. Run a full on-page audit of your entire site at least once per year. Additionally, trigger immediate reviews when: Google algorithm updates affect your rankings, competitors significantly outrank you on key terms, or you add major new content sections.
Run a Full On-Page SEO Audit
AuditDepot checks your pages against all 38 criteria above, identifies missing schema, duplicate tags, internal link gaps, and image issues — and prioritises fixes by estimated ranking impact.