On-Page SEO: What Moves Rankings in 2025

On-page SEO is the practice of optimising individual web pages so search engines can understand what they’re about and rank them accordingly. It covers everything from how you structure your content and use keywords to how you signal relevance through title tags, headings, internal links, and page experience. Done well, it’s the foundation that makes every other SEO effort, including link building and technical work, produce better returns.

Most of the advice you’ll find on this topic is either surface-level or a decade out of date. This article focuses on what the evidence and experience actually support in 2025, not what was true when keyword density was considered a strategy.

Key Takeaways

  • On-page SEO works best when it serves the reader first and the search engine second. Pages that genuinely answer a query tend to rank better than pages engineered to rank.
  • Title tags and H1s remain among the highest-leverage on-page elements. Getting them wrong costs you visibility before a user ever lands on your page.
  • Content depth matters more than content length. A 600-word page that fully resolves a query will outperform a 3,000-word page that circles the topic without landing.
  • Internal linking is one of the most underused on-page levers. It distributes authority, clarifies site structure, and tells Google what your most important pages are.
  • On-page signals compound over time. A page optimised once and left alone will drift. Treat your most important pages as assets that need periodic attention.

Why On-Page SEO Still Matters When Everyone Talks About Authority

There’s a version of the SEO conversation that goes something like this: links are everything, domain authority is what really counts, and on-page work is just table stakes. I’ve heard that argument in agency meetings for years. It’s partially true and mostly misleading.

Yes, a strong backlink profile helps. Yes, domain authority creates a floor from which it’s easier to rank. But I’ve watched well-linked pages sit on page three because the content didn’t match what the searcher actually wanted, and I’ve watched lean, well-structured pages from newer domains rank on page one because they resolved the query cleanly. On-page work is where you make your case to Google. Links are the votes, but the page is what they’re voting for.

If you’re building out a broader search strategy, the on-page layer sits at the centre of it. The complete SEO strategy hub on The Marketing Juice covers how on-page work connects to technical SEO, content planning, and link acquisition. Worth reading alongside this piece if you’re thinking about the full picture.

Title Tags: The Most Underestimated Element on the Page

The title tag is what appears in search results as the blue clickable headline. It’s also one of the clearest signals you can send Google about what your page is covering. Despite this, most pages still get title tags wrong in one of three ways: they’re too vague, they’re keyword-stuffed to the point of being unreadable, or they’ve been written for the brand rather than the searcher.

A good title tag does two things simultaneously. It tells Google what the page is about, and it gives the person scanning results a reason to click. Those two goals are not in conflict. A title that’s specific and useful tends to perform better on both dimensions.

Keep title tags under 60 characters to avoid truncation in search results. Front-load the primary keyword where it reads naturally. Don’t repeat the same phrase twice. And write for a person, not an algorithm. If you wouldn’t click on it, don’t publish it.

One thing worth knowing: Google rewrites title tags more often than most people realise. If your title tag doesn’t align well with the page content, Google may substitute something from your H1 or elsewhere on the page. The solution isn’t to fight that, it’s to make sure your title tag and H1 are both strong and consistent.

Heading Structure: Organisation That Helps Both Users and Crawlers

Headings are structural signals. They tell Google how your content is organised and which sections address which aspects of a topic. They also make content scannable for readers, which matters because most people don’t read web pages linearly. They scan, find what they need, and read that section.

Your H1 should match the intent of the page and include the primary keyword where it fits naturally. One H1 per page. After that, use H2s for main sections and H3s for subsections within those. Don’t use headings just to break up visual space. Each one should signal a genuine shift in the content.

When I was running iProspect and we were auditing client pages as part of growth work, heading structure was one of the first things we checked. Not because it was glamorous, but because it was often broken in ways that made everything else harder. Pages where the H1 was the company tagline and the actual topic was buried three sections down. Pages with five H1s because a developer hadn’t cleaned up a template. These aren’t edge cases. They’re common, and they cost rankings.

Keyword Placement: Where It Matters and Where It Doesn’t

The era of keyword density as a metric is over. Google’s ability to understand language has developed to the point where repeating a phrase seventeen times on a page signals nothing positive. What matters now is whether your keywords appear in the right places and whether the surrounding content actually supports them.

The places that still carry weight: the title tag, the H1, the first 100 words of the page, at least one H2, the meta description (for click-through rather than ranking), and the URL slug. Beyond that, write naturally. Use synonyms and related terms because that’s what good writing looks like, not because you’re gaming a system.

URL slugs deserve a specific mention because they’re often neglected. A clean, keyword-relevant slug, something like /onpage-seo/ rather than /page?id=4827, is a small signal but a consistent one. It also makes links easier to read when they appear in other contexts, which matters for click-through.

Content Quality: What Google Is Actually Trying to Measure

Google has been explicit about what it’s trying to reward: content that demonstrates experience, expertise, authority, and trustworthiness. The shorthand is E-E-A-T, and while it’s not a direct ranking factor in the mechanical sense, it describes the characteristics of pages that tend to rank well for competitive queries.

What that means practically: your content needs to show that a real person with relevant knowledge wrote it, that it covers the topic with enough depth to be useful, and that it doesn’t leave the reader needing to go somewhere else to get the full answer. That last point is important. If someone lands on your page and immediately bounces back to search results to find a better answer, that’s a signal Google notices.

I judged the Effie Awards for a period, which involved reading hundreds of marketing effectiveness cases. The work that won wasn’t necessarily the most creative or the most technically sophisticated. It was the work that most clearly connected what was done to what changed in the market. Content is the same. The pages that rank aren’t always the longest or the most elaborate. They’re the ones that most directly resolve what the searcher came to find.

Depth and length are not the same thing. A page can be 4,000 words and still be thin if it’s padded with repetition, generic observations, and content that doesn’t advance the reader’s understanding. Write what needs to be written to fully cover the topic. Stop there.

Meta Descriptions: Not a Ranking Factor, Still Worth Writing Well

Meta descriptions don’t directly influence rankings. Google confirmed this a long time ago, and there’s no credible evidence to suggest that’s changed. What they do influence is click-through rate, and click-through rate has a relationship with how much traffic a page generates from a given position.

Write meta descriptions as a pitch to the person scanning search results. What will they find on this page? Why is this result worth clicking over the others? Keep it under 155 characters to avoid truncation. Make it specific to the page, not a generic description of your site. And don’t keyword-stuff it. Nobody clicks on a list of search terms.

Like title tags, Google sometimes rewrites meta descriptions, particularly when the search query doesn’t match what you’ve written. The best defence is a meta description that closely reflects the page content and naturally includes the primary query. If Google keeps rewriting yours, that’s a signal worth investigating rather than ignoring.

Internal Linking: The On-Page Lever Most Teams Ignore

Internal links do three things that matter for SEO. They distribute page authority across your site, they help Google understand the relationship between your pages, and they give users a path deeper into your content. Most sites use internal links as an afterthought, adding them when they remember to rather than treating them as part of the page architecture.

The anchor text you use in internal links is a relevance signal. Linking to a page about on-page SEO with anchor text that says “click here” tells Google nothing. Linking with anchor text like “on-page optimisation signals” tells Google something specific about the destination page. Use descriptive, keyword-relevant anchor text for internal links, but vary it naturally rather than repeating the exact same phrase every time.

One pattern I’ve seen consistently across site audits: important pages that rank on page two or three despite strong content, simply because they have almost no internal links pointing to them. The site has authority, the page has good content, but it’s effectively invisible to the internal link graph. Adding a handful of relevant internal links from high-traffic pages can shift rankings in a matter of weeks. It’s one of the highest-effort-to-reward ratios in on-page work.

For a broader view of how internal linking fits into site architecture and content planning, the SEO strategy hub covers the structural side in more detail.

Image Optimisation: Small Details That Add Up

Images contribute to on-page SEO in ways that are easy to overlook. Alt text is the most discussed element, and it matters both for accessibility and for giving Google additional context about the page content. Write alt text that describes the image accurately and, where it fits naturally, includes a relevant keyword. Don’t stuff keywords into alt text for images where they don’t belong. That’s not optimisation, it’s noise.

File size is a page speed issue as much as an SEO issue. Uncompressed images are one of the most common causes of slow load times, and slow load times affect both rankings and conversion rates. Compress images before uploading. Use modern formats like WebP where your CMS supports them. Name image files descriptively rather than leaving them as IMG_4827.jpg.

These feel like small details. They are small details. But they compound. A page that handles every small detail well tends to outperform a page that handles only the big things. That’s true in SEO and it’s true in most of marketing.

Page Experience Signals: Where Technical and On-Page Overlap

Core Web Vitals, Google’s set of page experience metrics, sit at the intersection of technical SEO and on-page work. They measure loading performance (Largest Contentful Paint), interactivity (Interaction to Next Paint), and visual stability (Cumulative Layout Shift). Poor scores on these metrics don’t automatically tank rankings, but they create friction that affects both user experience and how Google evaluates the page.

From an on-page perspective, the most common culprits are render-blocking scripts, unoptimised images, and layout elements that shift as the page loads. These aren’t purely developer problems. They’re often the result of how content is structured and what assets are loaded on the page. A content team that adds a heavy third-party embed to every article without considering the load impact is making an on-page decision with technical consequences.

The practical advice here is to check your Core Web Vitals in Google Search Console regularly and treat poor scores as a content and development problem to solve together, not just a ticket to hand to the dev team and forget about.

Schema Markup: Telling Google Exactly What Your Content Is

Schema markup is structured data that you add to a page to help Google understand what type of content it contains. It doesn’t guarantee rich results in search, but it increases the likelihood of them. FAQ schema can produce expandable Q&A results. Article schema helps Google categorise editorial content. Product schema surfaces pricing and availability. Review schema can generate star ratings in results.

The value of rich results is primarily click-through. A result with a star rating or expandable FAQ takes up more visual space and gives the searcher more information before they click, which tends to increase the proportion of clicks that result in a visit. For competitive queries, that difference in click-through rate can be meaningful.

Implement schema that accurately reflects your content. Don’t add FAQ schema if the page doesn’t have a genuine FAQ section. Don’t add review schema if the ratings aren’t real. Google’s guidelines on structured data are clear, and pages that use schema misleadingly risk manual actions. Use it honestly and it’s a useful tool. Use it as a trick and it tends to backfire.

Content Freshness: When Updating Beats Creating

One of the most consistent findings across sites I’ve worked with is that updating existing pages often produces better ranking improvements than publishing new ones. This is particularly true for informational content where the topic hasn’t changed fundamentally but the specific details have evolved.

Google has a freshness signal that rewards pages that are kept current for queries where recency matters. That doesn’t mean changing a date in the footer and calling it an update. It means reviewing the content, adding what’s missing, removing what’s outdated, and ensuring the page still fully resolves the query it’s targeting.

The pages most worth updating are those sitting in positions four through fifteen for valuable queries. They’re close enough to rank well but not quite there. Often the gap is a content quality issue that an update can close. A proper audit of those pages, looking at what the top-ranking pages cover that yours doesn’t, is one of the more productive uses of an SEO team’s time. Moz has a useful framework for thinking through SEO auditing that’s worth reviewing if you’re approaching this systematically.

Early in my agency career, before I understood this properly, we spent a lot of time creating new content for clients while their existing pages drifted down the rankings through neglect. The new content rarely performed as well as the existing pages had at their peak. The lesson took a while to land, but it did: treat your best pages as assets, not as finished work.

The On-Page Audit: What to Check and in What Order

If you’re approaching on-page SEO systematically rather than page by page, an audit is the right starting point. The goal is to identify where the gaps are before deciding where to invest time.

Start with the pages that matter most commercially. For most businesses, that means the pages targeting queries with genuine purchase or conversion intent, not the blog posts that get traffic but don’t convert. Check title tags and H1s for relevance and keyword alignment. Check that the content depth matches what’s ranking for the same query. Check internal links pointing to the page. Check Core Web Vitals. Check whether schema is implemented and accurate.

That’s a lot to cover, which is why prioritisation matters. A page ranking on page two for a high-value query deserves more attention than a page ranking on page one for a low-volume informational query. Moz’s thinking on adapting SEO strategy to commercial context is a useful reference for how to frame those prioritisation decisions, particularly in B2B settings where the funnel is longer and the query landscape is more complex.

The output of an on-page audit shouldn’t be a list of everything that’s wrong. It should be a ranked list of what to fix first, based on the likely impact on rankings and commercial outcomes. If your audit produces 200 recommendations with no prioritisation, it’s not an audit. It’s a list.

How On-Page SEO Connects to the Rest of Your Search Strategy

On-page SEO doesn’t operate in isolation. It’s one layer of a broader system that includes technical foundations, content strategy, link acquisition, and how you track and interpret performance data. Getting the on-page layer right makes everything else more efficient. A technically sound site with strong backlinks but weak on-page work will underperform relative to its potential. A site with excellent on-page work but no links or technical issues will hit a ceiling it can’t break through.

The most effective SEO programmes I’ve seen treat these layers as interdependent rather than sequential. You don’t finish technical SEO and then start on-page work and then start link building. You work across all three simultaneously, with the relative emphasis shifting based on where the biggest gaps are at any given time.

If you’re thinking about how on-page work fits into a full search strategy, the articles in the Complete SEO Strategy hub cover the connected pieces in detail, from how Google determines rankings to how to track positioning changes without drawing the wrong conclusions from the data.

About the Author

Keith Lacy is a marketing strategist and former agency CEO with 20+ years of experience across agency leadership, performance marketing, and commercial strategy. He writes The Marketing Juice to cut through the noise and share what works.

Frequently Asked Questions

What is on-page SEO and what does it include?
On-page SEO refers to the optimisations you make directly on a web page to improve its visibility in search results. This includes title tags, H1 and heading structure, keyword placement, content quality and depth, meta descriptions, internal linking, image alt text, URL structure, schema markup, and page experience signals like Core Web Vitals. It’s distinct from off-page SEO, which covers external signals like backlinks, and technical SEO, which covers site infrastructure.
How important is keyword density for on-page SEO?
Keyword density is not a meaningful metric for modern on-page SEO. Google’s ability to understand language means that repeating a keyword a specific number of times does not improve rankings. What matters is whether your primary keyword appears in the right places, including the title tag, H1, first paragraph, and at least one H2, and whether the surrounding content genuinely covers the topic. Write naturally and use related terms as they arise, not as a tactic.
Do meta descriptions affect search rankings?
Meta descriptions are not a direct ranking factor. Google confirmed this years ago and there is no credible evidence that has changed. However, meta descriptions influence click-through rate, which affects how much traffic a page generates from its position in search results. A well-written meta description that accurately reflects the page content and gives searchers a reason to click is worth writing carefully, even if it doesn’t move rankings directly.
How often should you update existing pages for SEO?
There is no fixed schedule that applies universally. The most productive approach is to review your highest-value pages, those targeting queries with commercial intent or those sitting just outside the top three positions, at least once or twice a year. For topics where the information changes frequently, more regular updates are warranted. The trigger for an update should be a genuine gap between what your page covers and what the top-ranking pages cover, not just a desire to change a date.
What is the difference between on-page SEO and technical SEO?
On-page SEO covers the content and structural elements of individual pages, including titles, headings, copy, internal links, and schema. Technical SEO covers the infrastructure of the site as a whole, including crawlability, indexation, site speed, mobile usability, and structured data implementation at a site level. The two overlap in areas like Core Web Vitals and schema markup, but the distinction is roughly content versus infrastructure. Both matter, and neither substitutes for the other.

Similar Posts