How SEO Works Today (From Crawl to Conversion)

Google can fetch a page and still leave it out of its index. We saw that on CMX in an October 2026 Search Console review. Several newer articles loaded normally and allowed indexing, yet Search Console reported them as “Crawled – currently not indexed.” The familiar technical checks ruled out some problems. They didn’t make Google include the pages.

Search works in stages, and each one can fail for a different reason. Before rewriting a page that gets no clicks, find out whether Google indexed it. Then look at the searches where it appears and what readers can do when they arrive. That’s the path this guide follows.

How Google Finds and Ranks Webpages

SEO guides often shorten the process to crawl, index, and rank. That shorthand is useful, but Google describes its own process as crawling, indexing, and serving search results. Before crawling, a URL also has to be discovered.

Finding webpages

Google learns about URLs from places such as links on pages it already knows and XML sitemaps. A sitemap can help Google discover the URLs you consider important, but submitting one does not guarantee that every listed page will be crawled or indexed. An article linked naturally from a relevant hub has an additional discovery and context path that an isolated URL listed only in a sitemap lacks.

Reading the page

After Google discovers a URL, Googlebot may fetch it. Google can also render the page with a browser style system so it can process content that depends on JavaScript. A successful 200 response is necessary for a normal indexable page, but a successful fetch does not mean the page will automatically enter the index.

Deciding which version to keep

During indexing, Google analyzes the page and its important signals, including the main text, titles, images, links, and canonical information. It also groups substantially similar pages and selects a representative canonical URL. Google explicitly says indexing is not guaranteed, even when a page is crawlable and technically eligible.

The CMX pages mentioned above shared the same publishing system as other pages Google did index. Successful crawling and a self-referencing canonical tag weren’t enough to settle the indexing decision. That is why a technical audit needs to say what it confirmed and what remains outside the site’s control.

Choosing the search results

When someone searches, Google looks through its index for pages that may answer the query and applies multiple ranking systems and signals to choose what to show. Relevance, links, freshness where appropriate, language, location, device, and other signals can all affect the result. There is no single universal checklist that guarantees a position.

Google’s guide to how Search works is the best starting point for the technical pipeline. It is also useful when diagnosing a page because it separates a crawling problem from an indexing problem and an indexing problem from a ranking problem.

When Your Own Pages Compete for the Same Search

Search intent is the reason behind a query. People asking how indexing works need an explanation. People searching for an SEO agency usually want to know what the service includes and whether the provider can help them. Those searches might contain some of the same words, but the useful destination is different.

That’s why this guide and CMX’s service pages have different jobs. An article can explain the search process without pretending everyone reading it is ready to buy something. A service page should make the offering and the route to contact clear. If an older blog post already answers a new keyword idea well, improve that page before creating another one.

Once you’ve chosen the page, check that Google can discover and read it. Give readers a clear answer to the question they searched for. Link to other pages when they genuinely help with the next question. Keyword research is useful for finding those questions, though a search-volume number alone can’t tell you which answer belongs on which URL.

Keeping those responsibilities straight matters on an established site. Several articles can end up chasing the same query after years of publishing. Read them together before making another. Sometimes an older page needs a better explanation. Sometimes two near-duplicates should become one useful resource, with the old address handled properly.

What Influences Rankings After a Page Is Indexed

Once a page is in the index, ranking becomes a matching and selection problem. Google uses multiple ranking systems, and the weight of a signal can change with the query. A breaking news search, a local service search, and a technical tutorial do not need the same evidence.

For practical SEO work, four questions are more useful than chasing a list of hundreds of supposed ranking factors. Does the page answer the searcher’s task? Is the information trustworthy and specific enough to use? Does the wider web and the site itself give the page meaningful context through links and references? Can the user access and use the page without unnecessary friction?

Links remain important because they help discovery and can act as signals about relationships and reputation, but they are not a substitute for relevance. Page experience matters, but a fast page with a weak answer does not become the best result simply because it loads quickly. Google’s ranking systems guide is useful precisely because it describes multiple systems instead of reducing ranking to one metric.

Site Structure and Internal Linking

Internal linking does more than move visitors from one article to another. It gives search engines discovery paths and helps explain how pages relate to each other. A useful structure usually has clear hub pages for broad topics, narrower supporting pages for specific questions, and service or product pages for commercial intent.

For example, a site may have one broad guide explaining how SEO works, another article specifically about backlinks, and another about generative engine optimization. Those pages can reinforce each other if their jobs are distinct and the links describe why a reader would continue. If every article links to every other article with vague anchor text, the number of links goes up while the structure becomes less meaningful.

This is one reason we review page ownership before adding more content. A page that should function as the main SEO explainer needs stronger contextual links to its supporting articles, while those articles should point back when the broader explanation helps. Our guide to website structure for SEO goes deeper into that architecture.

What Useful Search Content Looks Like

A useful page keeps going when the easy advice runs out. A WordPress migration guide should tell readers what to check before switching DNS, how to verify redirects, and what a rollback would involve if the new site fails. Those details matter because someone could lose traffic or leave visitors on a broken site without them.

Detail still needs a purpose. If that migration guide wanders into a long explanation of DNS history, it may be better to link to a focused reference. Define the terms readers need, keep the decisions and warnings close to the steps they affect, and cut background that makes the task harder to follow.

Evidence matters most where a reader has reason to question the claim. Product specifications should point to current documentation. A statistic should identify its source and date. An SEO claim should distinguish between what Google documents, what an industry study observed, and what is simply a reasonable working hypothesis.

Where AI Overviews Fit Into Search

Google’s AI Overviews and AI Mode can combine information from multiple searches before presenting an answer. Google’s documentation calls part of that process query fan-out: searching related aspects of a question to gather useful material. A focused guide may help answer one part even when it isn’t the broadest page on the subject.

Google’s AI features guidance makes the practical requirements clear. Pages still need to be indexed and eligible for Search with a snippet to appear as supporting links. There isn’t a special “AI SEO” file or schema that guarantees an appearance. A citation in one generated answer also tells you little about whether anyone visited the page afterward.

Keep the useful evidence in the article and make its subject clear. CMX’s SEO After AI guide examines what AI search changes for content and measurement. This guide sticks to the underlying search process.

What the Numbers Can Tell You

Search Console can show when Google displays a page and whether someone clicks the result. It cannot tell you whether the contact form reached the business or whether the person who called was looking for that service. You need information from the website and the business to understand what happened afterward.

Follow the evidence in the order it becomes available. If the page isn’t indexed, begin with URL Inspection. Once Google shows it, look at the queries. Check clicks in the context of position and the other results on the page. Only then investigate what visitors do after arriving.

Our website ranking guide digs further into these checks. For a business site, keep the results tied to real inquiries and completed work. Traffic alone can’t tell you whether SEO is paying for itself.

Start With One Important URL

Pick a service page or a guide that matters to the business and inspect its status in Search Console. If it’s indexed, read the searches that bring it up. Open the result on your phone and see whether the page answers those questions. Record what you change so you can interpret the next few weeks of data.

That habit will tell you more than another generic SEO audit with a hundred warnings. When the evidence points to technical fixes or a larger content problem, CMX’s digital marketing services cover the work needed to put those changes into practice.