Onsite SEO: What Moves the Needle

Onsite SEO is the practice of optimising the content, structure, and technical elements within your own website to improve its visibility in search engines. Unlike link building or brand mentions, it is entirely within your control, which makes it both the most actionable part of SEO and the most frequently done wrong.

Most sites do not have an onsite SEO problem. They have a prioritisation problem. The signals that matter most, clear content structure, genuine topical depth, fast and crawlable pages, are not complicated. They are just consistently undervalued in favour of tactics that feel more sophisticated than they are.

Why Onsite SEO Is Where Most Strategies Stall

I have reviewed a lot of SEO audits over the years, running agencies across multiple disciplines, and the pattern is almost always the same. The audit is thorough. The recommendations are technically sound. And then nothing happens, because the list of issues is 47 items long and no one has decided which three actually matter.

Onsite SEO suffers from a prioritisation failure more than a knowledge failure. The information is widely available. The problem is that most teams treat an audit like a to-do list rather than a strategic document. They optimise title tags on pages that get no traffic, fix alt text on images that are not indexed, and spend three weeks on schema markup while their most important landing pages have duplicate H1s and no internal links pointing to them.

If you want a framework for how onsite SEO fits within a broader search strategy, the complete SEO strategy hub covers the full picture, from technical foundations to content and authority building. This article focuses specifically on the onsite elements and how to sequence them for commercial impact.

Title Tags: The Most Undervalued Real Estate in Search

Your title tag is not just an SEO signal. It is the headline your prospective customer reads before deciding whether to click. Most sites write title tags for the algorithm and forget there is a human making the decision.

The fundamentals are not complicated. Keep titles under 60 characters so they do not truncate in results. Front-load the primary keyword because that is what users are scanning for. Include something that differentiates the result, whether that is a number, a qualifier, or a specific angle. And write a different title tag for every page. That last point sounds obvious until you audit a 500-page site and find 80 pages using variations of the same template.

Google rewrites title tags when it thinks it knows better. That is frustrating, but it is also diagnostic. If Google is consistently rewriting your titles, it is telling you that your tags do not match what users are searching for or what the page actually contains. Rather than fighting the rewrite, read it as feedback.

Meta descriptions do not directly influence rankings, but they influence clicks, and clicks influence rankings indirectly. A well-written meta description functions like ad copy. It should give the reader a reason to choose your result over the four others on the page. If you are writing meta descriptions that summarise the page rather than sell it, you are leaving click-through rate on the table.

Content Structure: How Google Reads What You Write

Search engines do not read content the way humans do. They parse structure. Heading hierarchy, paragraph length, sentence clarity, and the logical flow of information all contribute to how well a page is understood and indexed. Good content structure is not about gaming the algorithm. It is about communicating clearly, which happens to be what the algorithm rewards.

One H1 per page, containing the primary keyword or a close variant. H2s for main sections. H3s for subsections within those. This is not an SEO trick. It is how documents are supposed to be structured, and search engines have learned to expect it. Violating the hierarchy, using H2s as design elements or skipping levels, creates ambiguity about what the page is actually about.

Paragraph length matters more than most content teams acknowledge. Long, unbroken paragraphs reduce readability and increase bounce rate. Short, punchy paragraphs improve both. I am not suggesting you write like a tabloid. I am suggesting that if your paragraphs run to eight sentences, you are making the reader work harder than they need to, and they will leave before they convert.

Featured snippets are worth pursuing deliberately. Google extracts snippet content from pages that answer a question clearly and concisely, typically in 40 to 60 words, directly after the question is posed. If you structure your content with the question as a heading and a clean, direct answer in the first paragraph beneath it, you are building for snippet eligibility. That is not a separate tactic. It is just good writing discipline, which is something the team at Copyblogger has argued for years.

Keyword Placement: Signal Without Stuffing

Keyword placement is one of those areas where the advice has swung between two unhelpful extremes. The first extreme was keyword density, stuffing a target phrase into every paragraph until the content read like a legal contract written by a robot. The second extreme is the current fashion for “write naturally and do not worry about keywords,” which sounds liberating but often produces content that ranks for nothing because it fails to signal clearly what the page is about.

The sensible middle ground: include your primary keyword in the title tag, the H1, the opening paragraph, at least one H2, and the meta description. After that, write naturally. Use synonyms, related terms, and the language your audience actually uses. Understanding how keywords connect to commercial intent is more useful than chasing exact-match frequency.

Semantic relevance has become more important as search engines have become more sophisticated. A page about “project management software” that also covers topics like task assignment, deadline tracking, and team collaboration will outperform a page that just repeats “project management software” twelve times. The depth of coverage signals expertise. The keyword signals the topic. You need both.

One thing I have noticed consistently when reviewing content strategies for clients, across industries from financial services to ecommerce, is that the pages performing best are rarely the ones most optimised in the traditional sense. They are the ones that most completely answer the question. Completeness is the signal. Keywords are just the label on the tin.

Internal Linking: The Lever Most Sites Ignore

Internal linking is the most underused onsite SEO tactic I have encountered across client work. It is free, it is within your control, it works immediately, and almost every site does it poorly. That combination should make it the first thing you fix, not the last.

Internal links do three things simultaneously. They pass PageRank between pages, helping your most important pages accumulate more authority. They signal topical relationships to search engines, helping Google understand which pages belong to which subject clusters. And they improve crawl efficiency, ensuring that new or updated content gets discovered and indexed faster.

The practical application is straightforward. Every time you publish a new page, identify five to ten existing pages on your site that are relevant to the new content and add links from those pages to the new one. Use descriptive anchor text that reflects what the destination page is about, not generic phrases like “click here” or “read more.” And audit your highest-traffic pages regularly to ensure they are linking out to your most commercially important pages, because that is where you want to concentrate authority.

Orphan pages, pages with no internal links pointing to them, are a persistent problem on sites that have been publishing content for years without a linking strategy. They exist in the index but receive no authority and are often crawled infrequently. A quarterly internal link audit to identify and resolve orphan pages is one of the highest-return maintenance tasks in onsite SEO.

Page Experience and Core Web Vitals: The Floor, Not the Ceiling

Core Web Vitals became a Google ranking signal, and the industry responded in the way it usually does: by treating a floor requirement as a ceiling achievement. Teams celebrated passing the threshold as if they had done something exceptional, when in reality they had just stopped doing something harmful.

The three metrics worth understanding are Largest Contentful Paint, which measures how quickly the main content loads; Cumulative Layout Shift, which measures visual stability as the page renders; and Interaction to Next Paint, which replaced First Input Delay as the measure of responsiveness. Google publishes thresholds for each. Passing them is table stakes. The business case for going beyond them is not about rankings. It is about conversion.

I spent a significant portion of my agency years managing performance marketing budgets across retail and financial services clients. The pages that converted worst were almost always the slowest. Not because users consciously noticed the load time, but because every additional second of delay increased the probability they would leave before the page finished rendering. The relationship between page experience and conversion rate is well documented, and it is a stronger commercial argument for performance investment than any ranking benefit.

Mobile experience deserves specific attention. Google indexes the mobile version of your pages first. If your mobile experience is materially worse than desktop, your rankings reflect the mobile version regardless of how good the desktop experience is. Test your most important pages on actual mobile devices, not just browser emulators, and prioritise fixing anything that creates friction before a user can access the content they came for.

URL Structure and Site Architecture: Getting the Foundations Right

URL structure is one of those onsite elements that is easy to get right at the start and expensive to fix later. A clean URL tells both users and search engines what a page is about. A messy URL, full of parameters, session IDs, or auto-generated strings, tells them nothing and can create indexation problems at scale.

The principles are simple. Use lowercase letters and hyphens to separate words. Include the primary keyword where it fits naturally. Keep URLs as short as the content hierarchy allows. Avoid dynamic parameters in URLs that are intended to be indexed. And when you change a URL, implement a 301 redirect from the old address to the new one, without exception.

Site architecture is the broader question of how pages relate to each other hierarchically. A flat architecture, where important pages are never more than a few clicks from the homepage, is preferable to a deep architecture where content is buried five or six levels down. The flatter the structure, the more efficiently PageRank flows to every page, and the more easily crawlers can reach new content.

Siloing, the practice of grouping related content into topical clusters with clear internal linking between them, is one of the more durable architectural strategies in SEO. It mirrors how Google understands topical authority, and it makes the site easier to handle for users at the same time. The topic cluster model is not new, but it is consistently underimplemented. Most sites have the content. They just have not connected it deliberately.

Schema Markup: Structured Data That Earns Rich Results

Schema markup is machine-readable code that tells search engines explicitly what your content represents. A recipe page with schema tells Google it is a recipe, with a cooking time, ingredient list, and calorie count. A product page with schema tells Google the price, availability, and review rating. Without schema, Google has to infer these things from the content. With schema, you remove the ambiguity.

The commercial case for schema is rich results. Pages with appropriate structured data are eligible for enhanced search result formats, star ratings, FAQ dropdowns, event listings, and product information displayed directly in the results page. These formats increase visibility and, more importantly, click-through rate, because they give users more information before they click.

The most universally applicable schema types are Article for editorial content, FAQ for question-and-answer sections, Product for ecommerce pages, LocalBusiness for location-based businesses, and BreadcrumbList for site navigation. Implement the types relevant to your content, validate them using Google’s Rich Results Test, and do not implement schema for content types you do not actually have. Misleading structured data is a policy violation and can result in manual action.

One thing worth flagging: schema is not a ranking factor in the traditional sense. It does not make your page rank higher for a given query. What it does is make your result more visible and more informative when you already rank. That distinction matters for how you prioritise it relative to other onsite work.

Image Optimisation: The Onsite SEO Task Most Teams Rush

Images are a meaningful source of both performance drag and missed SEO opportunity. Uncompressed images slow page load times. Missing alt text leaves accessibility and indexation value on the table. Generic file names like “image-1.jpg” tell search engines nothing about the content of the image.

The optimisation checklist is short. Compress images before uploading, using formats like WebP where browser support allows. Write descriptive alt text that explains what the image shows, including the primary keyword where it fits naturally and honestly. Name image files descriptively before uploading, using hyphens to separate words. And use lazy loading for images below the fold so they do not block initial page render.

Image search is a traffic channel that most content teams undervalue. For categories like recipes, travel, fashion, interior design, and product discovery, a meaningful percentage of search traffic enters through image results rather than web results. If your images are properly optimised and contextually relevant, you are eligible for that traffic. If they are not, you are invisible in that channel entirely.

Content Freshness: When to Update and When to Leave Well Alone

Freshness is a ranking signal for queries where recency matters. News, current events, product comparisons, and anything with a year in the search query all favour recently updated content. For evergreen topics, freshness matters less, but it still matters enough to be worth managing deliberately.

The default assumption in most content teams is that publishing new content is always the right move. I would push back on that. In my experience managing content strategies for clients with large existing archives, updating and consolidating existing content consistently outperforms publishing new content at the same investment level. A page that already has some authority, some inbound links, and some search history is a better starting point than a blank page. Improving it is faster and the returns compound on an existing base.

The signal for when to update is simple: check rankings and traffic trends quarterly. If a page that was previously ranking well has dropped without any obvious technical issue, the content is likely outdated relative to what is now ranking. Update the substance, not just the date stamp. Google is sophisticated enough to detect cosmetic freshness signals, and gaming the date without improving the content does not work reliably.

Content consolidation, merging multiple thin pages on related topics into a single authoritative page, is one of the highest-impact onsite SEO moves available to sites with large archives. It concentrates authority, eliminates cannibalisation, and typically produces a page that ranks better than any of the individual pages did. It also reduces the crawl budget wasted on low-value content, which matters at scale.

For a broader view of how onsite SEO connects to technical performance, link strategy, and content planning, the SEO strategy hub brings those threads together in one place. Onsite work does not exist in isolation, and the decisions you make here affect everything else in the search channel.

The Onsite SEO Audit: How to Prioritise When Everything Feels Urgent

I mentioned earlier that most sites have a prioritisation problem, not a knowledge problem. The audit is where that problem becomes most visible. A competent technical audit will surface dozens of issues. The question is which ones are costing you the most, and the answer is rarely obvious from the audit output alone.

A useful prioritisation framework has two axes: impact and effort. High-impact, low-effort fixes go first. These are typically things like fixing broken internal links, adding missing title tags, resolving duplicate content on high-traffic pages, and improving the internal linking structure around your most commercially important pages. None of these require development resources or significant time investment.

High-impact, high-effort fixes come next. These include site speed improvements that require engineering work, content consolidation projects that involve redirects and rewrites, and structural changes to URL hierarchies. These need planning and resourcing, but they belong on the roadmap with a clear owner and deadline.

Low-impact fixes, regardless of effort, go to the bottom of the list or get dropped entirely. This is where most teams waste time. The 47-item audit becomes a 12-item action plan when you apply this filter honestly. The discipline of cutting the list is harder than building it, but it is where the real strategic thinking happens.

The best onsite SEO thinking is not complicated in hindsight. Clear structure, complete content, fast pages, deliberate linking. These are not new ideas. They are just consistently underexecuted because teams chase novelty when the fundamentals would serve them better. I have seen this pattern repeat across agencies, across industries, and across budgets of every size. The sites that rank well are rarely the ones doing the most sophisticated things. They are the ones doing the basics with unusual consistency.

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