Google uses hundreds of signals to decide which pages to show and in what order, but they are not equally important and many popular “factors” have little or no effect. This guide explains the ranking factors that matter in 2026, grouped the way Google’s own systems work: relevance, quality, links, page experience, technical access, behaviour and local signals. It also covers what has changed with AI Overviews and which myths you can safely ignore.

A note on evidence before we start. Google publishes a guide to its ranking systems and a set of spam policies, and its engineers have described parts of the process in conference talks and court testimony. Everything else, including most “200 factor” lists, is informed opinion. Where something is confirmed, this guide says so; where it is a well-supported theory, it is labelled as one.

How Google ranks pages in brief

Ranking happens after crawling and indexing. When someone searches, Google pulls thousands of candidate pages from its index, then a series of systems score and re-order them. Some systems interpret the query (such as BERT and neural matching), some judge quality and trust, some count links, some demote spam, and some reorder results using how people interact with them.

Three points follow from this:

There is no single score. A page can be excellent on one signal and still lose because it misses the intent of the query.
Weights change by query. Freshness matters a great deal for news and almost nothing for a recipe. Trust signals matter far more for health and money topics.
Google reworks these systems constantly, with several broad core updates a year, so a ranking is never permanent.

1. Relevance: matching the query and its intent

The first job of any page is to answer the query it ranks for. Google looks at the words on the page, but modern language systems mean it understands meaning rather than counting exact phrases.

Search intent

Search intent is the most important relevance factor. If the results for a query are all product category pages, a blog post will struggle to rank there, however good it is. Before writing or optimising, search the term yourself and note what type of page, format and depth Google already rewards.

Keywords and topic coverage

Using the main phrase in the title tag, the H1, early in the copy and in a few subheadings still helps Google confirm what a page is about. Beyond that, coverage matters more than repetition: a page that answers the related questions people have, uses the natural vocabulary of the subject and explains things clearly tends to beat one that repeats a keyword. Keyword density targets are not something Google uses.

Passages and specific answers

Google can rank a specific section of a long page for a narrow question, a capability it calls passage ranking. Clear subheadings, with a direct answer straight after each one, make it easier for Google to find and use those sections. The same structure helps AI features quote you.

Freshness

For searches where recent information matters, such as news, prices, software versions or annual statistics, Google’s freshness systems favour up-to-date pages. Changing a date without improving the content does not help. Real content freshness means updating facts, examples and screenshots when they go out of date.

2. Content quality and helpfulness

Since 2022 Google has pushed hard on rewarding content written for people rather than for search engines. The helpful content system was folded into the core ranking systems in March 2024, so these signals now apply all the time, not only during specific updates.

Helpful, people-first content

Google’s guidance on helpful content asks questions such as: does the content provide original information or analysis, does it fully answer the question, would a reader leave feeling satisfied, and was it created mainly to attract search visits? Pages that summarise what others have already said, with nothing added, are increasingly filtered out.

Originality and information gain

Content that adds something new tends to perform better: first-hand experience, original research, real examples, your own photos and data. SEOs often call this information gain. Google holds a patent on the idea, but how far it is used in ranking is not confirmed, so treat it as a sensible principle rather than a proven factor.

E-E-A-T

E-E-A-T stands for experience, expertise, authoritativeness and trust. It comes from Google’s quality rater guidelines, which human raters use to assess search results. Google has said E-E-A-T is not a single ranking factor, but that its systems use a mix of signals that line up with it. In practice that means:

Clear author bylines with real credentials, linked to an about page.
Accurate, sourced claims, especially on YMYL topics such as health, finance and legal matters.
A business that is easy to verify: address, contact details, reviews and mentions elsewhere on the web.
A reputation that holds up when someone searches for your name.

Topical authority

Sites that cover a subject in depth, with many well-linked pages on related questions, tend to rank more easily for new pages on that subject. This is usually called topical authority. Google has not named it as a factor, but the effect is widely observed and fits with how it describes rewarding sites with expertise in a topic.

Thin and unhelpful content elsewhere on the site

Quality is judged partly at site level. Large numbers of thin pages, near-duplicates or outdated posts can hold back the stronger pages on the same domain. Improving, merging or removing weak pages is often one of the fastest ways to lift a whole site.

3. Links and authority

Links remain one of the strongest ranking signals. Google’s link analysis systems, including PageRank, still use links to understand which pages are important and what they are about, although Google has said links matter less than they once did compared with other signals.

Backlinks from other sites

Relevance: a link from a respected site in your industry is worth far more than many from unrelated directories.
Authority of the linking page: links from pages that themselves attract links pass more value.
Number of linking sites: links from many different referring domains are a stronger signal than many links from one site.
Anchor text: the clickable words help Google understand the target page, but exact-match anchor text used at scale looks manipulative.
Editorial placement: a link inside the body of a relevant article counts for more than one in a footer or sidebar.

Link spam

Buying links, large-scale guest posting for links, private blog networks and link exchanges all break Google’s rules on link schemes. Google’s spam detection, including SpamBrain, mostly neutralises these links rather than penalising the site, which means the money is simply wasted. Serious cases can still lead to manual actions.

Internal links

Internal linking is the ranking lever you control completely. Links between your own pages spread link equity, show Google which pages matter most and help it understand how topics connect. Pages with no internal links pointing to them are hard for Google to find and rarely rank well.

4. Page experience and Core Web Vitals

Google rewards pages that are pleasant to use, but it is clear that page experience is a lighter signal than relevance and quality. A fast page with weak content will not beat a slower page that answers the question far better. Where competing pages are similar, experience can tip the balance.

Core Web Vitals

Core Web Vitals measure real-user experience in Chrome, assessed at the 75th percentile of visits:

Metric	What it measures	Good
Largest Contentful Paint	Loading: when the main content appears	2.5 seconds or less
Interaction to Next Paint	Responsiveness to clicks and taps	200 milliseconds or less
Cumulative Layout Shift	Visual stability while loading	0.1 or less

INP replaced First Input Delay as a Core Web Vital in March 2024.

Mobile experience

Google completed its move to mobile-first indexing in 2023, so it ranks pages based on the mobile version. Content, links and structured data that only appear on desktop may as well not exist. Text should be readable without zooming, and buttons easy to tap.

HTTPS and intrusive interstitials

HTTPS has been a lightweight ranking signal since 2014 and is now simply expected. Pop-ups that cover the main content on mobile, known as intrusive interstitials, can count against a page. Cookie banners and legally required notices are treated differently from promotional pop-ups.

5. Technical foundations: can Google access the page?

Technical SEO is less a ranking factor than a precondition. If Google cannot crawl, render and index a page, none of the other factors matter.

Crawlability: pages must not be blocked in robots.txt, and important pages should be reachable through internal links and listed in an XML sitemap.
Indexability: check that key pages are not set to noindex by mistake and that they return a 200 status code.
Canonicalisation: where similar URLs exist, canonical tags tell Google which version to rank, so signals are not split by duplicate content.
Rendering: content loaded by JavaScript must be visible to Google. JavaScript SEO problems are a common reason pages index without their main text.
Clean URLs and site structure: a logical URL structure and shallow click depth help both crawling and understanding.
Structured data: schema markup does not directly raise rankings, but it can earn rich results that improve click-through rates and helps search engines understand your entities.

Google Search Console shows which pages are indexed and why others are not. It is the first place to look when a page does not rank at all.

6. User behaviour signals

For years Google played down the role of clicks. Testimony in the US Department of Justice antitrust case, and the 2024 leak of internal API documents, made clear that click and interaction data does feed into ranking, most notably through a system called NavBoost. Google has said the leaked documents lack context and should not be read as a list of active factors, so the details remain uncertain.

What is reasonably safe to say:

Pages that consistently satisfy searchers, so people do not return to the results to try another page, tend to hold their positions.
An attractive, accurate title and description that earn the click help, but misleading titles that disappoint visitors work against you.
Single metrics such as bounce rate or time on page from your own analytics are not used directly. Google does not see your Google Analytics data for ranking.

The practical lesson is the same as for content quality: answer the question quickly and well, and make the next step obvious. User signals are hard to fake at scale and tend to follow genuine quality.

7. Brand and entity signals

Well-known brands rank more easily, and this is not only because they have more links. Searches that include your brand name, mentions across the web, a Knowledge Panel, consistent business information and reviews all help Google recognise your business as a real, trusted entity. Growth in branded search is often a sign that wider marketing is supporting SEO.

This is also one of the clearest links between SEO and AI search: AI assistants are more likely to mention businesses that are described consistently in many trustworthy places.

8. Local ranking factors

For searches with local intent, such as “plumber near me” or “dentist in Leeds”, Google uses a separate set of signals for the map pack. Google’s own help pages name three main factors: relevance, distance and prominence.

Relevance: how well your Google Business Profile matches the search, including your primary category, services and description.
Distance: how far the searcher, or the location in the query, is from your business. You cannot change this, which is why rankings vary across a city.
Prominence: how well known the business is, based on reviews (number, rating and recency), links, mentions and your organic rankings.

Consistent name, address and phone details across directories (NAP consistency), location pages on your website and local links all support these signals. See the full guide to local ranking factors for more detail.

9. Spam policies and demotions

Some factors only work in one direction: breaking the rules can push you down, but following them earns nothing extra. Google’s spam policies cover cloaking, hidden text, keyword stuffing, doorway pages, scraped content and link spam. In March 2024 Google added three more:

Scaled content abuse: producing large amounts of unoriginal content mainly to rank, whether written by people, AI or both.
Site reputation abuse: third-party pages published on a trusted site to exploit its ranking strength.
Expired domain abuse: buying old domains to rank low-value content on their past reputation.

Using AI to help write content is not against the rules in itself. Google judges the result, not the tool: content made mainly to manipulate rankings is the problem.

10. Ranking factors and AI search

AI Overviews and AI Mode draw on Google’s normal index and ranking systems. Google has said there is no special optimisation required to appear in them beyond the same fundamentals: crawlable, indexable pages with helpful content. In practice, pages that are cited tend to:

Already rank reasonably well for related queries.
Give clear, self-contained answers under descriptive subheadings.
Include specific facts, figures and first-hand detail that are worth quoting.
Come from sites with a consistent reputation on the topic.

Other assistants such as ChatGPT and Perplexity use their own retrieval, often based on Bing or their own crawlers, but reward similar qualities. The practice of optimising for them is often called generative engine optimisation.

Ranking factor myths to ignore

Meta keywords: Google has ignored the meta keywords tag for ranking since at least 2009.
Keyword density and LSI keywords: there is no ideal percentage, and LSI keywords are not something Google uses.
Word count: longer is not better in itself. The right length is whatever fully answers the query.
Domain age: Google has said the age of a domain is not a ranking factor in itself.
Social shares: likes and shares are not direct ranking signals, although the visibility they bring can lead to links and searches.
Submitting your site to Google repeatedly: requesting indexing does not improve rankings.
Exact-match domains: a keyword in the domain name gives little or no advantage today.

How to prioritise ranking factors for your site

Every site has a different weak point. A sensible order of work for most businesses is:

Fix access problems first. Check indexing in Google Search Console and fix anything that stops important pages being crawled or indexed.
Match intent. Make sure each important page targets one clear search intent and is the right type of page for it.
Improve the content. Add original detail, answer related questions and update anything out of date. Merge or remove thin pages.
Strengthen internal links. Link from strong pages to the pages you want to rank, using descriptive anchor text.
Build trust. Add author details, reviews, case studies and clear business information.
Earn links. Create resources, data and expert commentary that other sites want to reference.
Tidy page experience. Fix Core Web Vitals failures and mobile usability issues, especially on key templates.

A structured SEO audit covers all of these in one pass and shows which will make the biggest difference.

Frequently asked questions

How many ranking factors does Google use?

Google has long said it uses “hundreds” of signals, combined across many separate systems. No public list is complete, and the weight of each signal changes from query to query.

What is the most important Google ranking factor?

There is no single most important factor, but relevance to the search intent comes first: a page that does not answer the query will not rank. After that, content quality and links from relevant, trusted sites have the biggest influence for most searches.

Are Core Web Vitals a ranking factor?

Yes, as part of Google’s page experience signals, but a relatively light one. They mostly matter when competing pages are otherwise similar in quality and relevance.

Is E-E-A-T a ranking factor?

Not directly. E-E-A-T is a framework from Google’s quality rater guidelines. Google uses many signals that reflect experience, expertise, authority and trust, so improving these qualities tends to help rankings over time.

How long does it take for changes to affect rankings?

Small on-page changes can be reflected within days of Google recrawling the page. Site-wide quality improvements and new links usually take weeks to months, and some effects only show after a core update.

For definitions of every term used here, see the SEO Encyclopedia. Further reading: Google’s guide to its ranking systems.

Want to know which ranking factors are holding your site back? Get a free SEO check.