Google's algorithm is not a single system — it is a layered set of processes that work together to crawl the web, index content, and rank pages for billions of queries each day. Understanding how these processes work is not optional for agencies that deliver SEO for clients. Every tactical decision — what content to create, which links to acquire, what technical fixes to prioritise — should be grounded in how Google actually evaluates and ranks pages, not in simplifications or outdated mental models.
Step 1 — Crawling
Before Google can rank a page, it must find and crawl it. Googlebot is a web crawler that discovers pages by following links across the internet. It starts from a set of known pages and continuously follows the outbound links on those pages to discover new ones. This is why internal linking and external links pointing to a site both matter — they are the pathways through which Googlebot navigates.
Crawl budget is the number of URLs Googlebot will crawl on a site within a given time period. For large sites, managing crawl budget matters — if Googlebot wastes budget crawling low-value pages (duplicate content, faceted navigation URLs, soft 404 pages), important pages may be crawled less frequently. For most small and medium sites, crawl budget is not a limiting factor.
Pages that cannot be crawled cannot be indexed. Common crawl barriers include: pages blocked in robots.txt, pages behind login walls, pages with noindex tags, and JavaScript-rendered content that Googlebot does not execute. Technical SEO work often focuses on removing these barriers for pages that should be indexed.
Step 2 — Indexing
Once Googlebot crawls a page, the content is processed and added to Google's index — a database of all the web content Google has seen and assessed. Indexing is not automatic or instant. Google decides whether to index a page based on its quality and uniqueness. Pages with thin, duplicated, or low-quality content may be crawled but not indexed, or may be indexed and later removed from the index following a quality update.
Content that needs to be indexed should be rendered in HTML that is accessible without JavaScript execution. While Googlebot can render JavaScript, it is a secondary process that runs later than the initial HTML crawl. Critical content — headings, body text, meta tags — should be present in the initial HTML response from the server. This is why PHP-rendered sites (like this one) have an inherent indexing advantage over client-side rendered JavaScript applications for SEO purposes.
Step 3 — Ranking
When a user submits a query, Google's ranking systems retrieve all indexed pages that are relevant to the query and score them against hundreds of signals. The result is a ranked list. The signals can be grouped into several categories:
Content relevance signals assess how well the content matches the query. This includes keyword presence in the title, headings, and body text, but has expanded significantly beyond simple keyword matching. Google uses natural language processing to understand semantic relationships between terms, so a page about "emergency plumbing" can rank for "burst pipe repair" even if that exact phrase does not appear on the page.
Authority signals assess the trustworthiness and credibility of the page and domain. Inbound links from relevant, authoritative sites remain the strongest authority signals. Google uses a version of PageRank — the original link authority algorithm — as a core component of ranking. The quality, relevance, and anchor text of inbound links all factor into the authority assessment.
User experience signals include Core Web Vitals (LCP, INP, CLS), mobile usability, and HTTPS security. These are confirmed ranking signals that affect how well a page can rank even when content and authority are strong. A page with excellent content and strong links that loads slowly on mobile will underperform against a faster competitor.
E-E-A-T: Google's Quality Framework
E-E-A-T stands for Experience, Expertise, Authoritativeness, and Trustworthiness. It is the framework used by Google's quality raters to evaluate the overall quality of a page and a site. It is not a direct algorithmic signal — it is a human evaluation framework — but the underlying signals it describes feed directly into algorithmic ranking decisions.
Experience was added to the original E-A-T framework to reflect the value of first-hand, lived experience. A review of a plumbing tool written by someone who has used it extensively is more valuable than a review written by someone who has not. Google's helpful content system is partially designed to reward this kind of direct experience over content that merely describes what others have said.
Expertise means demonstrable knowledge of the subject matter. For medical, legal, and financial content (YMYL topics), Google expects formal credentials. For other topics, demonstrated knowledge evidenced through the quality and depth of the content is sufficient.
Authoritativeness refers to the recognition of the site or author as an authority by others in the field. This is most directly built through external references — links, mentions, citations — from other authoritative sources. A site that is widely linked to and cited by others in its industry is more authoritative than one that is not, regardless of its self-description.
Trustworthiness is the foundation. It includes factual accuracy, transparent ownership, clear authorship, accessible contact information, and security (HTTPS). Trustworthiness signals apply at the page level and the domain level.
Core Updates and the Helpful Content System
Google's broad core updates are algorithm changes that update how the ranking systems weight signals across many pages simultaneously. They do not target specific sites or tactics — they recalibrate the overall weighting of signals. Sites that see ranking changes after a core update should assess whether their content quality, E-E-A-T signals, and link profiles genuinely deserve the position they hold for their target queries.
The helpful content system is a site-wide signal that assesses whether a site's primary purpose is to help users or to rank in search. Sites with significant proportions of content written primarily for SEO purposes — formulaic, thin, not reflecting genuine expertise or experience — receive a site-wide quality demotion. The system rewards content that serves a genuine audience need, demonstrates first-hand experience, and would be valuable even if it never ranked in Google.
For agencies delivering SEO that holds through algorithm updates, our white label SEO service focuses on sustainable, quality-first strategies. See also our SEO content writing service for content that meets these quality standards, or contact us to discuss your agency's requirements.