Here is an uncomfortable truth most SEO guides skip: you can do everything “right” on a WordPress article, pick the right keyword, hit the Rank Math green score, publish on schedule, and still rank on page three for years.
The reason almost always comes down to the same set of mistakes. Thin headings that tell Google nothing. Content that answers the surface question without covering the real topic. Internal links that go nowhere useful. Images with generic names that waste a ranking opportunity. These are not beginner errors. Experienced WordPress publishers make them every week.
This guide covers every on-page SEO element that actually moves rankings in 2026. Not the theory, but the exact decisions you make inside the WordPress editor before you hit publish. We cover title tags, meta descriptions, heading structure, content quality, search intent, semantic keywords, internal linking, and image SEO, all in one place.
+32% more clicks from moving up just one position in resultsAIOSEO, 2026
What On-Page SEO Actually Means in 2026
On-page SEO covers every decision you make on an individual page to help Google understand what it is about, who wrote it, whether it answers the query, and whether the reader leaves satisfied.
That last part has grown in importance significantly. For most of SEO history, on-page SEO meant placing keywords in the right spots. Title tag, H1, first paragraph, done. Google now evaluates meaning, topic completeness, content structure, author credibility, and behavioural signals that show users actually found what they needed.
Think of SEO as three layers building on each other. Technical SEO is the foundation, Google needs to crawl and index your pages before any content decision matters. On-page SEO is the content layer you build on top of that foundation. When both work together, you create the conditions for appearing in AI-generated answers. We cover that in full detail in our guide on how to appear in Google AI Overviews.
Where most WordPress guides fail is treating on-page SEO as a checklist of boxes to tick. It is a set of editorial decisions about clarity, structure, sourcing, and relevance that you make before every publish. The checklist supports those decisions. It does not replace them.

Google’s Helpful Content system asks one question about every page: did the person who searched for this leave satisfied? On-page SEO is the craft of making sure the answer is yes, clearly enough that Google’s systems can measure it.
How E-E-A-T Changes the Way You Write
E-E-A-T stands for Experience, Expertise, Authoritativeness, and Trustworthiness. Google introduced it in its Quality Rater Guidelines, and its influence on rankings has increased with every core update since 2023. If you want a dedicated breakdown of how E-E-A-T signals affect AI citation specifically, read our guide on E-E-A-T for Google AI Overviews.
For on-page SEO purposes, E-E-A-T changes four specific writing decisions you make on every article.
Experience: Write From What You Have Actually Done
Google added “Experience” to the original E-A-T framework in December 2022 because AI tools had made it trivial to produce fluent, accurate-sounding text with zero firsthand knowledge behind it. The addition sent a clear signal: Google wants content from people who have actually done what they describe.
For Technexies, this means writing from the perspective of someone who has installed Rank Math, fixed a real crawl error, and watched a Core Web Vitals score change after adding a caching plugin. Not someone summarising what other articles say about those things. The difference in tone is immediately obvious to any reader, and Google’s systems have learned to detect it.
Practical application: replace general advice with specific observations. Instead of “you should optimise your title tag,” try “when we moved the primary keyword to the start of this article’s title, the click-through rate in Search Console improved within four weeks.” Specific. Observed. Verifiable. That is experience on the page.
Expertise: Show the Depth Behind the Advice
Expertise means you understand your topic deeply enough to explain the things other guides skip. You know why the rule exists, not just what the rule is. You can explain when an exception applies and back it up with evidence.
On an SEO article for WordPress users, expertise looks like explaining that Google rewrites title tags 76% of the time and what that means for how you should write them, rather than just saying “keep your title under 60 characters.” The first sentence shows understanding of the underlying mechanism. The second repeats a surface rule anyone can find in a basic SEO post.
Authoritativeness: Build the Author Layer
Authoritativeness is about your reputation outside your own site. Other sites cite you. Your author name appears consistently across well-regarded content. Google can verify the author is a real person with a publishing history.
For WordPress sites, the most actionable step is making sure every article carries a detailed author bio with a name, demonstrated experience, and ideally a link to a profile that exists somewhere outside your own domain. Google’s Quality Raters use the author bio to assess whether the person who wrote the article is qualified to have written it. It is not decoration. It is a trust signal.
Trustworthiness: Source Every Claim You Make
According to Google’s quality evaluator guidelines, trustworthiness is the most critical E-E-A-T component because untrustworthy pages have low E-E-A-T regardless of how expert or authoritative they may otherwise appear.
The practical fix is simple: link to the source every time you cite data, a study, or a specific factual claim. Not just “research shows” but a named study from a named organisation with a year attached. If you cannot find a credible source for a claim, either find one or remove the claim. Invented statistics destroy trust faster than almost any other content decision.
Keyword Research and Search Intent
Before you write a single word, you need to understand what your reader is actually searching for and what they expect to find when they click. Keyword research in 2026 starts with intent, not volume.
Search Intent Determines Your Content Format
Search intent is the reason behind a query. Someone searching “how to noindex author archives WordPress” wants step-by-step instructions. Someone searching “Rank Math vs Yoast 2026” wants a comparison with a clear recommendation. If your content format does not match what those searchers expect to find, you will not hold their attention long enough to rank regardless of how well everything else is optimised.
There are four types of search intent. Informational intent covers “how to” and “what is” queries. Navigational intent means someone is looking for a specific site or brand. Commercial intent covers comparison and review queries. Transactional intent means the searcher is ready to act. For an SEO and WordPress blog, informational and commercial intent make up the vast majority of relevant traffic. Match your content format to the intent type before you set a word count or structure your headings.
Where to Find Keywords Without Guesswork
Start with Google Search Console. Go to Performance, then Queries. Every term in that report is a real search query someone used to find your site. These are more valuable than anything a keyword tool estimates, because they reflect actual demand from your actual audience.
For new keyword discovery, use Google directly. Type your main topic and observe three things: what autocomplete suggests (common phrasing patterns), what appears in the People Also Ask box (related questions worth covering in your article), and what the Related Searches section at the bottom reveals (adjacent topics your audience cares about). These three free sources reflect real search behaviour from billions of queries.
For competitive data and volume estimates, Ahrefs and Semrush both have free tiers. Even limited free access lets you check whether a keyword has meaningful volume and how competitive the ranking landscape is before you invest thousands of words in covering it.
One Primary Keyword Per Page
Each page targets one primary keyword. Not a group of similar keywords, not a primary and a secondary with equal weight. One clear topic that the entire page covers thoroughly.
A page that tries to rank for “on-page SEO for WordPress,” “WordPress on-page optimization,” and “how to do on-page SEO” simultaneously ends up serving none of those queries particularly well. A page committed entirely to one topic covers it more thoroughly, structures its headings more coherently, and satisfies the reader more completely.
In Rank Math, set your focus keyword before you write. The panel flags if the keyword is missing from critical positions: title, URL, opening paragraph, H2, and meta description. Use those flags as editorial guidance, not as a score to optimise toward. A Rank Math score of 80 on a genuinely excellent article beats a score of 100 on a mechanically optimised but hollow one every time.
Two pages on the same site targeting the same keyword split the ranking signal between them. Google picks one to rank and suppresses the other, and it may not pick the one you prefer. Before publishing any new article, check whether an existing page already targets the same query. If it does, either merge both pages or clearly differentiate their angles so each serves a distinct version of the intent.
Semantic and NLP Keywords: The Layer Most Sites Miss
Most WordPress publishers focus entirely on their primary keyword and miss the layer underneath it that now drives the majority of how Google evaluates topical relevance. That layer is semantic coverage.
Google does not use Latent Semantic Indexing (LSI) as a ranking algorithm. Google’s own John Mueller confirmed this explicitly: “There’s no such thing as LSI keywords — anyone who’s telling you otherwise is mistaken.” What Google does use is far more sophisticated: BERT, MUM, and entity-based NLP models that analyse the meaning and context of your entire page, not just the keywords on it.
The practical outcome is the same. Pages that naturally cover the semantic neighbourhood of a topic — the related concepts, entities, questions, and vocabulary that expert writers use when covering that topic — signal genuine topical authority. Pages that repeat one keyword and ignore the surrounding context look thin to Google’s NLP systems regardless of how many times the keyword appears.
The Semantic Keywords for On-Page SEO
For a page targeting “on-page SEO for WordPress,” Google’s BERT and MUM systems expect to see the following terms and concepts present because they are the natural vocabulary of this topic. These should appear in your content not as a checklist but because they are part of genuinely covering the subject.
| Semantic Term or Entity | Why It Belongs on This Page |
|---|---|
| Search intent | Core concept in modern on-page SEO. Pages that ignore intent cannot rank regardless of technical optimisation. |
| Title tag optimisation | One of the primary on-page elements Google evaluates for relevance |
| Meta description | Affects click-through rate, which influences ranking signals indirectly |
| Heading structure (H1, H2, H3) | Signals content hierarchy and topic coverage to Google’s crawlers |
| Focus keyword | Rank Math terminology every WordPress publisher uses daily |
| Keyword density | Commonly misunderstood concept this page needs to clarify accurately |
| Content structure | How information is organised and how it signals quality to Google |
| Click-through rate (CTR) | Behavioural signal that connects title tags and meta descriptions to rankings |
| Internal linking strategy | Core on-page element for distributing authority and guiding crawlers |
| Anchor text | Specific attribute of internal and external links that signals page topic |
| Alt text | Image SEO element that contributes to accessibility and relevance signals |
| URL structure / URL slug | On-page element that contributes to both crawlability and relevance |
| Rank Math | The entity (tool) most relevant to WordPress on-page SEO configuration |
| Google Search Console | Primary entity for measuring the outcomes of on-page SEO decisions |
| Helpful Content system | Google’s content quality evaluation framework directly relevant to on-page decisions |
| BERT and NLP | The underlying technology that evaluates semantic coverage on the page |
| Schema markup | Structured data that extends on-page signals beyond raw text |
| Topical authority | The outcome of consistent, semantically rich coverage across a topic cluster |
How to Use Semantic Keywords Without Forcing Them
You do not add semantic keywords by inserting them into sentences where they do not belong. You add them by covering the topic completely. A page about on-page SEO for WordPress that genuinely covers title tags, heading structure, content quality, and internal linking will naturally contain every term in that table above, because those are the concepts you discuss when covering the topic well.
The test is simple: read your article and ask whether an expert on this topic would feel any major concept is missing. If yes, cover it. When you do, the semantic vocabulary follows automatically.
Semrush’s research on semantic keywords confirms that using contextually related terms improves Google’s understanding of your page’s topic and scope, which expands the range of related queries your page can rank for beyond the primary keyword alone. That is the real value of semantic coverage: not a slight bump on one keyword but a broader footprint across an entire topic cluster.
Entity Coverage: Mention What Google Knows
Entities are specific, universally recognised concepts, people, tools, organisations, and places that Google’s Knowledge Graph stores and understands. When your content mentions relevant entities naturally, it gives Google’s NLP systems additional context anchors to verify your page is genuinely about its stated topic.
For on-page SEO content aimed at WordPress users, the relevant entities include: Google, Rank Math, Yoast, Google Search Console, PageSpeed Insights, Screaming Frog, BERT, John Mueller, WordPress, Gutenberg editor, and the Google Helpful Content system. These are not keywords to stuff into the article. They are real concepts and tools that belong in an expert discussion of this topic and appear naturally when you write from genuine knowledge.
Writing Title Tags That Survive Google Rewrites
Your title tag is the headline in Google search results. It is often the first and only thing a potential reader sees before deciding whether to click. There is one significant problem with it: Google rewrites most title tags before displaying them.
The Rewrite Data You Need to Know
SEO consultant John McAlpin’s Q1 2025 study, published in Search Engine Land, found that Google rewrites title tags 76% of the time, up from 61% in 2023. When Google does rewrite, it retains only around 35% of the original words on average.
Your title tag still matters as a ranking signal even when Google shows a different version. These are two separate processes. The original title influences whether the page ranks for a query. The displayed version influences whether a searcher clicks. Design your title to do both jobs.
When Google rewrites a title, it pulls from three sources: your H1 tag, your page content, and anchor text from links pointing to the page. This means your H1 and your title tag should say the same thing in closely aligned language. When they align, Google has less reason to generate something different.
The Specific Rules That Reduce Rewrite Rates
The Zyppy study of 80,000 title tags found that titles between 51 and 60 characters have the lowest rewrite rate, approximately 39% to 42%. Titles under 20 characters get rewritten over 50% of the time. Titles beyond 12 words are rewritten more often than not. The 51 to 60 character range sits in the sweet spot: long enough to be informative, short enough to display completely on every device.
Numbers reduce rewrite rates significantly. The same study found that when a title contains a number and the H1 also contains a matching number, Google preserves that number 97.3% of the time. “7 Fixes for WordPress Crawl Errors in 2026” is harder for Google to rewrite than “How to Fix WordPress Crawl Errors in 2026” because the specific number anchors the title to the page content in a way that generic phrasing does not.
Front-load the primary keyword. Not because keyword placement in title tags is a powerful ranking signal on its own — Google’s John Mueller has described it as a minor factor — but because readers scan the beginning of headlines first. If your keyword appears at the end of a title that gets truncated at character 60, some readers may never register it before moving to the next result.
Write the Title for the Reader, Not the Algorithm
The most common title tag mistake is optimising for Google first and the human reader second. Titles written that way feel generic and slightly awkward. They fail to stand out in a results page where eight other titles contain the same phrase.
Ask what the reader is trying to accomplish and make the title promise that specific outcome. “How to Fix 404 Errors in WordPress: A Step-by-Step Guide” works because it tells a problem-aware reader that this page delivers a practical solution. “WordPress 404 Error Fix 2026” tells them almost nothing about whether clicking is worth their time.
Primary Keyword or Problem + What the Reader Gets + Year Where RelevantWeak: “WordPress SEO Title Tags Guide 2026”
Strong: “How to Write Title Tags That Google Won’t Rewrite — WordPress 2026″The second version answers the question the reader actually has and promises a specific, useful outcome.
Set your title in Rank Math’s SEO Title field in the post editor. The character counter shows your length in real time. Check the SERP preview directly below it to see exactly how your title and meta description will appear in Google before you publish.
Meta Descriptions: Your One Shot at the Click
Meta descriptions do not affect rankings. Google confirmed this officially, and multiple independent studies have verified it. They are not a direct ranking signal.
They are, however, your primary tool for convincing a reader who has already seen your title to actually click through. That makes them a CTR driver, and higher CTR is a behavioural signal that Google does factor into rankings indirectly. A page that earns more clicks than its position predicts sends a relevance signal to Google that compounds over time.
Google Rewrites These Too — But You Still Write Them
An Ahrefs study of 20,000 keywords found that Google rewrites meta descriptions roughly two thirds of the time, typically pulling a passage from the page body that more closely matches the specific query a user typed.
When Google uses your description, it means your version was already the best match for the most common version of that query. That is exactly what you want. Write it for the most likely searcher and the most likely query. When Google replaces it with a body excerpt, it means the searcher used a variant you did not specifically write for, which is fine.
Never leave the meta description blank and trust Google to fill it in automatically. When Google generates a description from unguided page content, it often pulls a mid-article sentence written for a reader who is already on the page, not a searcher deciding whether to visit it.
What Strong Meta Descriptions Actually Do
Keep it under 160 characters. Descriptions beyond this get truncated with an ellipsis on desktop. Put the most important information in the first 120 characters to handle mobile truncation safely.
Include your primary keyword naturally, once. Google bolds matching terms in the meta description when they match what the user searched. A description with bolded keywords stands out visually against competitors that do not match the query as closely.
Tell the reader what they will get, not what the page is. “This guide covers on-page SEO for WordPress” is a description of content. “Learn exactly how to write title tags, structure headings, and build internal links that Google rewards in 2026” is a promise of specific, applicable value. One describes. The other sells the click.
Weak: “On-page SEO for WordPress is important for rankings. Learn about keywords, titles, and headings in this complete guide.”
Strong: “Stop guessing what Google wants from your WordPress pages. This guide covers title tags, headings, semantic keywords, internal links, and image SEO with the exact steps to apply each one before you publish.”
The second version is active, specific, and addresses the reader’s real problem. It earns the click because it earns the trust first.
Heading Tags Done Right on WordPress
Heading tags do two things for SEO. They create a content hierarchy that tells Google which sections cover which topics. And they give readers navigation landmarks they use to decide where to read carefully and where to skim.
Neither function is served by headings used as visual styling. If you use an H3 tag because it looks good on the page rather than because it genuinely introduces a subsection of the content above it, you have broken the structure for both Google and your reader.
The H1: One Per Page, Matched to the Title
Every WordPress post has exactly one H1 — the post title. WordPress wraps it automatically when you write in the block editor. Never add a second H1 as a styled heading block inside your article content. Check by viewing the page source on any published post and searching for <h1>. You should find exactly one instance.
Your H1 should closely match your title tag without being identical. The title tag fits within 60 characters and serves searchers on the results page. The H1 has more room and serves readers who have already clicked through. Use that extra space to be slightly more descriptive, but cover the same topic with the same keyword so they align clearly. Aligned title and H1 together reduce Google’s incentive to rewrite your title in search results.
H2 Tags: Section Headings Written as Real Questions
Write every H2 as if a reader might arrive at that section directly without reading anything above it. Each H2 should be a complete, clear statement of what that section covers or answers.
“Introduction” is a useless H2. “What Is On-Page SEO and Why It Still Matters in 2026” is a strong H2. The second version contains the section’s primary question, signals the keyword cluster to Google, and tells every reader exactly what value that section delivers.
Google’s systems use H2 content to match pages against related queries beyond the primary keyword. A page with H2s written as real, specific questions or statements naturally ranks for more long-tail variations of its main topic. According to Backlinko’s ranking factor research, heading tags help Google understand page structure and represent a genuine, if moderate, on-page relevance signal.
Include your primary keyword or a close semantic variation in at least one H2. Do not force it into every H2 because that looks unnatural and reduces the clarity of your headings for actual readers. One natural placement is the standard.
H3 Tags: Subsections Only When the Hierarchy Genuinely Exists
Use H3 tags when a section has multiple distinct sub-points that each need their own heading. Do not use them to add visual variety or make a page look more structured than it actually is. A section with three short paragraphs does not need three H3 headings. That fragments the reading experience and creates false hierarchy.
Correct hierarchy: H1 for the page title, H2 for main sections, H3 for subsections within those sections. Never skip a level. Jumping from H2 to H4 creates a structural error that damages both accessibility and crawl clarity. Screen readers traverse headings sequentially, and Google’s crawlers interpret the hierarchy the same way.
Using the Heading block in Gutenberg for pull quotes, callout text, or visual dividers. These are not headings. They do not introduce new sections of content. Use a paragraph block with bold formatting, or a custom callout block, for any text that is meant to stand out visually but does not start a new content section. Misusing heading tags breaks your semantic structure for both readers and crawlers.
How to Write Content Google Rates as Genuinely Helpful
Google’s Helpful Content system launched in 2022 and has been refined through every major core update since. Its purpose is to identify content written for people versus content written to satisfy an algorithm. In 2026, after the December core update and the continued rollout of AI-powered search evaluation, the distinction has never mattered more.
Answer First, Then Explain
Start every section by answering the question its heading asks. Then explain the answer. Then support it with evidence or a real example.
Most writers build up to the answer. They provide context, background, and caveats, then arrive at the point at the end of the paragraph. That structure works in academic writing. It fails completely for search traffic. A reader who arrived from Google wants to confirm they are on the right page within the first sentence. If the opening does not deliver that confirmation, they leave — and Google’s behavioural signals register that the page did not satisfy the query.
Short Paragraphs, One Idea Each
Keep paragraphs to two to four sentences. One idea per paragraph. Start a new paragraph whenever you shift to a new point or angle.
Research from the Nielsen Norman Group on web reading behaviour established that users scan content before committing to reading it. Short paragraphs with clear opening sentences allow a reader to scan efficiently and choose where to read carefully. Long, dense paragraphs interrupt that scanning pattern and cause readers to skip entire sections — which reduces time on page and increases bounce rate, both signals Google measures.
Cover the Topic Completely Without Padding
Comprehensive content means covering the topic without leaving the reader needing to search for something else. It does not mean adding word count for its own sake. Google’s Helpful Content guidelines specifically ask whether “a reader would feel they got what they came for” from the page.
Before you publish, ask: what is the most likely follow-up question someone would search after reading this article? Either answer it in the article, or link to a page that does. If the follow-up question is genuinely out of scope for this page, a brief mention with an internal link satisfies the reader without inflating the article’s length with off-topic content.
Semantic Completeness for AI Citations
AI systems — Google’s AI Overviews, ChatGPT, Perplexity, and others — cite pages that answer queries in self-contained, semantically complete passages. They do not cite pages that make readers scroll through five sections to assemble the answer from fragments.
A study of 15,847 AI Overview results found that semantic completeness was the strongest predictor of citation, with pages that scored highly on this factor being 4.2 times more likely to appear as a cited source. Write each section of your article so it could stand alone as a direct answer to a specific question. If a section requires context from three paragraphs above it to make sense on its own, restructure it.
For the complete breakdown of what Google and AI systems look for when choosing which pages to cite, read our guide on how to appear in Google AI Overviews. Our guide on AI Overview optimisation mistakes covers the specific content patterns that disqualify pages from citation even when the content itself is solid.
Use Tables and Lists With Discipline
Tables work for comparisons, configuration settings, and structured data with multiple attributes per item. Lists work for sequential steps, sets of items with clear membership, and anything where order matters.
Do not convert prose into bullet points just to break up the visual weight of a page. A paragraph that builds an argument — where each sentence connects logically to the next — does that job better as a paragraph. Fragmenting it into bullets strips the reasoning out and leaves a list of disconnected statements. Google’s quality systems distinguish between structured content that organises information clearly and bullet-pointed content that avoids forming coherent sentences.
Internal Linking: The Biggest Win Most Sites Ignore
Internal linking is the on-page SEO element with the best return on effort. It costs nothing beyond the time to add the link. It requires no external relationships or budget. And it delivers measurable benefits that compound as your site grows.
What Internal Links Do for Your Rankings
Internal links serve three distinct functions simultaneously.
They guide Google’s crawlers to pages that might not otherwise be discovered quickly. Google’s own crawling documentation explains that Googlebot follows internal links to discover content. A page with no internal links pointing to it — what SEOs call an orphan page — may take weeks to index even if it appears in your sitemap, because Google has no clear crawl path prioritising it.
They distribute authority. When a page earns backlinks from external sites, some of that ranking power flows through internal links to connected pages. A pillar article that earns several quality backlinks can accelerate the ranking of cluster articles through well-placed, contextual internal links.
They signal topical relationships. Google uses internal link patterns to understand which pages on your site cover related topics, which pages the site considers most important, and what subject areas your site claims expertise in. Google’s Search Advocate John Mueller described internal linking as “one of the biggest things you can do on a website to guide Google and guide visitors to the pages that you think are important.”
The Pillar-Cluster Model for WordPress Blogs
For a content-focused WordPress site, the pillar-cluster model produces the most effective internal link architecture. One pillar page covers a broad topic comprehensively. Cluster articles each go deep on one specific subtopic within that broader theme. Every cluster links back to the pillar. The pillar links out to every cluster.
This creates a closed authority loop. Each page reinforces the others. Google sees a clear content hierarchy. Readers have natural navigation paths between related content. And your site builds topical authority across an entire subject area rather than on isolated articles.
A Semrush case study from August 2025 documented a startup that used a well-organised, contextually relevant internal linking strategy and achieved four times the monthly organic traffic of a competitor with similar domain authority that used unrelated or inconsistent internal links. The content was comparable. The internal link architecture was not.
Your Technexies Internal Linking Structure
Every article on Technexies connects to the broader topic clusters that have already been built. Here is how each live pillar and cluster relates to on-page SEO and where natural internal links belong.
| Article | What It Covers | How It Connects to On-Page SEO |
|---|---|---|
| Technical SEO for WordPress | Crawlability, Core Web Vitals, sitemaps, duplicate content | The infrastructure layer that on-page SEO builds on top of |
| How to Appear in Google AI Overviews | AI citation signals, content structure, schema | The outcome strong on-page SEO creates conditions for |
| E-E-A-T for Google AI Overviews | Experience, expertise, authoritativeness, trust signals | The credibility layer that on-page SEO must demonstrate through sourcing and author signals |
| Structured Data for AI Overviews | Schema markup, JSON-LD, FAQPage, Article schema | The structured data extension of on-page optimisation beyond raw text |
| AI Overview Optimisation Mistakes | Content patterns that prevent AI citation | The practical audit checklist for on-page content decisions |
| Rank Math SEO Setup Guide | Every Rank Math setting that affects indexing and on-page configuration | The tool layer where on-page SEO decisions get implemented in WordPress |
| WordPress Core Web Vitals | LCP, INP, and CLS are confirmed Google ranking factors | Page experience signals that extend beyond content into performance |
| Fix Crawl Errors in WordPress | 404s, noindex errors, robots.txt issues | Ensures the pages you optimise on-page are actually indexed by Google |
| WordPress Duplicate Content Fix | Archive pages, canonical tags, tag pages | Ensures on-page authority concentrates on the right URLs |
Anchor Text: Make Every Link Descriptive
Anchor text — the words that carry the link — tells Google what the destination page is about. “Click here” and “read more” are wasted anchor text because they convey no topical information. “WordPress duplicate content fix” and “how to fix crawl errors in WordPress” are descriptive anchors that reinforce the destination page’s topic and keyword focus every time Google crawls the link.
Use natural anchor text that fits the sentence without reading as forced. Do not add links just to have links somewhere on the page. Add them when the connection genuinely serves the reader. A reader learning about on-page SEO will find a link to the Rank Math setup guide useful, because that is where on-page settings get implemented. A link inserted purely to move authority serves no one.
Every time you publish a new article, go back to two or three relevant existing posts within the same hour and add an internal link to the new page. This prevents orphan pages, speeds up Google’s discovery of new content, and ensures your newest article benefits from the authority already built in older posts. Make this a non-negotiable step in your publishing process, not something you get to later.
How Many Internal Links Per Page
There is no ideal number. The correct answer is as many as genuinely serve the reader, and no more. For a 4,000 to 6,000 word pillar article, four to eight internal links to related cluster articles is reasonable. For a cluster article of 2,000 to 3,000 words, two to four links back to the pillar and to one or two related clusters is appropriate.
Too many links dilute the signal each individual link sends. If every paragraph contains a link, nothing is prioritised. Think of internal links as editorial recommendations. You recommend something because you genuinely believe the reader should follow it, not because a quota says this section needs a link.
Image SEO for WordPress
Images affect three separate ranking factors simultaneously. They are indexable content that can appear in image search. They affect page load speed, which influences Core Web Vitals scores. And they send on-page relevance signals through alt text and file names. Most WordPress sites handle all three badly by default.
Name Your Image Before You Upload It
The file name of an image is the first signal Google uses to understand what the image shows. “IMG_5047.jpg” communicates nothing. “rank-math-seo-setup-titles-metas-screenshot.webp” tells Google the exact content and context of the image.
Name every image file descriptively before uploading to the WordPress media library. Use hyphens between words, lowercase letters, and include your primary keyword or a closely related term where it fits naturally. This takes ten seconds per image and is one of the highest-return, lowest-effort on-page tasks available to any WordPress publisher.
Alt Text: Write It for Both People and Crawlers
Alt text is the written description that replaces an image when it cannot load and is read by screen readers for accessibility. Google also uses it to understand what an image depicts and to match images to relevant search queries.
Write alt text that describes the image specifically and accurately. “Screenshot of Rank Math Titles and Metas panel with noindex toggle enabled for author archives” is good alt text — specific, descriptive, and naturally includes relevant terms. “Rank Math settings” is too vague. “Rank Math SEO settings 2026 noindex author archives WordPress plugin” is keyword-stuffed and reads as unnatural — avoid it.
In Rank Math’s General Settings under Image SEO, enable “Add Missing ALT Attributes” and set the format to the filename. This auto-generates fallback alt text from file names for images without manual alt text. For featured images, hero images, and any image central to the content’s argument, always write the alt text manually.
WebP Format and File Size
WebP is the correct image format for WordPress in 2026. WebP files run 25 to 35% smaller than JPEG at equivalent visual quality, which directly reduces page load time and improves your LCP score. ShortPixel and Imagify both convert images to WebP automatically on upload and can batch-convert your existing media library. Both have free tiers sufficient for most new WordPress sites.
Width and Height Attributes Prevent Layout Shifts
Every image tag needs explicit width and height attributes. Without them, the browser does not know how much space to reserve for the image before it loads. This causes the page layout to shift as images appear, creating a high CLS score.
A high CLS score is a Core Web Vitals failure. Core Web Vitals are confirmed ranking signals. An image decision made in thirty seconds — adding explicit dimensions — directly affects whether your page passes or fails a factor Google explicitly measures and weights in rankings. Rank Math’s Image SEO feature adds these attributes automatically when enabled. Verify by inspecting any image on your live pages. You should see width and height values in the img tag.
Pre-Publish Checklist: Every Item Before You Hit Publish
Run through every section below before publishing any article. These are not suggestions. They are the minimum standard for a page worth indexing.
Keyword, Intent, and Semantic Coverage
- One primary keyword identified, confirmed, and checked against existing pages for cannibalism
- Search intent confirmed: does your content format match what searchers expect to find?
- Focus keyword entered in Rank Math’s Focus Keyword field before writing
- Keyword appears naturally in title, URL slug, first 100 words, at least one H2, and meta description
- Semantic and NLP terms present naturally throughout: related concepts, entities, and vocabulary that belong to the topic
- Every relevant entity for the topic (tools, organisations, named concepts) mentioned where it fits naturally
Title Tag
- Title written manually in Rank Math (never leave this field blank)
- Between 51 and 60 characters
- Primary keyword appears near the front
- Written for the reader first so it promises a specific, useful outcome
- H1 and title tag closely aligned in topic and language
Meta Description
- Meta description written manually (never leave this field blank)
- Under 160 characters, most important content in the first 120
- Primary keyword included naturally, once
- States what the reader gets, not just what the page contains
- Active voice used throughout with passive voice below 10% of sentences
Headings
- Exactly one H1 on the page, which is the post title WordPress generates automatically
- H2 tags used for all main sections, written as real questions or clear statements
- H3 tags used only for genuine subsections within H2 sections
- No heading levels skipped
- Primary keyword or a close semantic variation appears in at least one H2
- No heading tags used for pull quotes, decorative text, or visual styling
Content Quality and E-E-A-T
- Opening paragraph answers the primary query within the first 100 words
- Every factual claim links to a named, credible source with a year
- No statistics without attribution and no invented or unverifiable data anywhere in the article
- Paragraphs kept to two to four sentences with one idea each
- Topic covered completely so a reader would not need to search elsewhere for the core answer
- Author bio is present, real, and links to a credible external profile
- Active voice used throughout with passive voice kept below 10% of sentences
Internal Links
- Links to the pillar page from this cluster article (or links to all cluster articles from this pillar)
- Links to at least one related cluster article with descriptive anchor text
- All anchor text is descriptive: “WordPress duplicate content fix” not “click here”
- After publishing: add a link to this new page from two or three relevant existing articles within the same session
Images
- All images named descriptively before uploading, using hyphens, lowercase letters, and relevant terms
- Alt text written manually for featured images and key content images
- All images converted to WebP format
- Width and height attributes present on all images, verified through Rank Math Image SEO settings or page source inspection
Technical and Publish Steps
- URL slug is short, readable, and contains the primary keyword
- Schema configured in Rank Math: Article schema applies automatically and FAQPage schema added manually for FAQ sections
- Rank Math red flags reviewed and addressed, using the score as editorial guidance rather than a target to chase
- After publishing: URL submitted immediately for indexing in Google Search Console
Frequently Asked Questions
Q. What is the single most important on-page SEO factor for WordPress in 2026?
Search intent alignment. You can get every technical element right — title tag length, heading structure, keyword placement, meta description — and still rank poorly if your content format does not match what searchers expect to find. A how-to guide ranks for how-to queries. A comparison page ranks for comparison queries. Get the format right first. Then optimise the elements within it. Everything else is secondary to whether the page genuinely satisfies the query it targets.
Q. Do LSI keywords and NLP keywords help rankings in 2026?
Google does not use Latent Semantic Indexing as a ranking algorithm. John Mueller confirmed this explicitly. What Google does use are advanced NLP models — BERT and MUM — that evaluate the meaning and semantic context of your entire page, not just individual keywords. The practical implication is identical to what LSI advocates recommend: cover the topic completely using the natural vocabulary of the subject, including related concepts, entities, and questions. When you do that genuinely, the semantic signals take care of themselves. When you do it artificially by inserting “LSI keywords” from a tool, it reads as unnatural and rarely improves performance.
Q. Why does Google keep rewriting my title tags?
Google rewrites title tags when it judges that its version better matches the search query that triggered the display, or when the original title is too long, too short, misleading, or heavily keyword-focused. The most effective ways to reduce rewrites: keep titles between 51 and 60 characters, align your H1 closely with your title tag, include a specific number in both where possible, and write titles that clearly describe the page’s actual content rather than optimising for a keyword cluster. When your title accurately represents the page and aligns with the H1, Google has less reason to generate an alternative.
Q. How do I check on-page SEO issues across my existing WordPress posts?
Start with Rank Math’s SEO score in each post editor — it flags missing title tags, meta descriptions, focus keywords, and alt text. For a site-wide audit, run Screaming Frog (free up to 500 URLs) to generate a complete report of duplicate titles, missing meta descriptions, missing H1 tags, and broken internal links across all pages simultaneously. In Google Search Console, the Performance report shows which pages earn impressions but low clicks — these are the strongest candidates for title tag and meta description improvement because the ranking is already there, but the click is being lost.
Q. How long should a WordPress article be to rank well?
Long enough to cover the topic completely, short enough to not pad it with filler. For competitive how-to guides and technical topics, 2,500 to 5,000 words is typical among top-ranking pages — but word count is not a ranking factor. The correct length is whatever prevents a reader from needing to search for something else after finishing your article. If a reader would, the article is too short. If you are repeating yourself or covering tangential topics to reach a word count target, it is too long. Write the right length for the topic, then stop.
Q. What is the difference between on-page SEO and technical SEO?
Technical SEO covers the infrastructure layer — ensuring Google can crawl, render, and index your pages correctly. It includes Core Web Vitals, sitemaps, crawl errors, robots.txt, HTTPS configuration, and duplicate content. On-page SEO covers the content layer — what you say, how you structure it, how you signal relevance to Google, and how you demonstrate credibility through sourcing and writing quality. Both layers are essential. Technical SEO without solid on-page SEO means Google can find your pages but has no strong reason to rank them. On-page SEO without technical SEO means you have written good content that Google cannot reliably access or evaluate. Read our complete Technical SEO for WordPress guide to make sure the infrastructure layer is solid before investing in content optimisation.
Kia has worked in SEO and digital marketing for over a decade, building and optimising websites across different industries. He founded Technexies to share what actually works in modern search written from direct professional experience rather than theory. All content on Technexies is researched, written, and reviewed by Kia personally.

