Most explanations of SEO skip the part that actually matters: how Google decides which page wins. After 14 years of doing this work, I find that when clients understand the underlying mechanics, they make better decisions, spend money more wisely, and set more realistic expectations. So here is how it actually works.

The Three Jobs of a Search Engine

Before any ranking happens, a search engine does three things in sequence: crawl, index, and rank. These are not interchangeable. A page can be crawled but not indexed. An indexed page can still rank nowhere. Understanding where a problem lives in this chain is half the diagnostic work in SEO consulting.

  • Crawling — Googlebot follows links across the web, discovering URLs. How often it crawls a given page depends on the page’s perceived importance, your crawl budget, and how frequently the content changes. Pages that are not linked from anywhere may never be crawled at all.
  • Indexing — After crawling, Google decides whether to add the page to its index. Pages that are thin, duplicate, blocked by robots.txt or meta noindex, or simply not linked from anywhere useful are often excluded entirely. Being indexed is not guaranteed.
  • Ranking — For a given query, Google scores every eligible indexed page against hundreds of signals and returns a ranked list. This is where most people focus, but crawl and index problems upstream make ranking efforts irrelevant.

What Signals Actually Drive Rankings

Google has confirmed several core ranking factors over the years, and practitioners have validated others through observation and testing. I group them into four buckets.

Relevance Signals

The page has to be about the query. That sounds obvious, but relevance is more nuanced than keyword density. Google uses semantic understanding — it knows that a page about ‘how to fix a leaking faucet’ is also about plumbing, even if the word ‘plumbing’ never appears. Title tags, headings, and the structure of the content all contribute to how well Google understands what a page covers. Internal links from related pages on the same domain reinforce topical relevance further. Google rewards sites that go deep on a topic, not sites that mention every possible phrase once.

Authority Signals

Links from other websites still carry significant weight. A link is a vote of trust. A link from a credible, relevant domain carries far more weight than a link from a directory nobody reads. What has changed over the past decade is that manipulative link schemes — paid links, private blog networks, link farms, irrelevant guest posts at scale — now create liability rather than benefit. Google has gotten much better at identifying artificial link profiles, and the penalties are real. Genuine editorial links built through real work — creating content worth citing, building relationships, earning press coverage — are what move the needle sustainably.

Experience and Helpfulness Signals

Google has invested heavily in evaluating whether content actually helps the person who searched. User behavior signals (click-through rate, time on page, return visits), content depth, author credibility signals, and what Google calls E-E-A-T (Experience, Expertise, Authoritativeness, Trustworthiness) all factor into how content is evaluated. A page that gets clicks, keeps visitors engaged, and genuinely answers the question is telling Google something real about its quality. A page clearly written to target a keyword phrase without actually serving the searcher will underperform over time.

Technical and Page-Experience Signals

Core Web Vitals (load speed, interactivity, visual stability), mobile usability, HTTPS, and crawlability are table stakes in modern SEO. They are not ranking accelerators on their own — but failing them creates a hard ceiling. I have seen well-written, well-linked pages held back entirely by a slow server or a JavaScript rendering issue that prevented Googlebot from seeing the content at all. Technical hygiene is not exciting, but it is necessary.

Where Schema Markup Fits In

Structured data — schema markup — is not a direct ranking factor in the traditional sense, but it is a precision tool for communication with search engines. It tells Google exactly what your content represents: a product, a review, a person, a local business, an event, an FAQ. Done well, it unlocks rich results in the SERP — star ratings, FAQ dropdowns, event listings, breadcrumb trails — and makes disambiguation easier. It is where a meaningful share of my consulting work lives. Most agencies treat schema as a checkbox. Done with real specificity and strategic intent, it changes how a page appears in results entirely, which directly affects click-through rate. The team at Salterra has built entire service lines around advanced custom schema implementation.

The Role of Content

Content is the thing you are optimizing. There is no SEO without it. But more content is not the answer — the right content for the right queries, written with genuine depth and a clear point of view, is what earns rankings. Google’s Helpful Content system has been explicit about this: content written primarily for search engines rather than people will be downgraded over time. Every page should answer one thing well rather than trying to capture everything vaguely. A page that is the best available answer to a specific question will outperform a page that shallowly addresses ten questions.

A Concrete Example: Local Professional Services

Say a CPA firm in Scottsdale wants to rank for ‘business tax accountant Scottsdale.’ Here is what actually has to happen. Their page needs to be crawlable and properly indexed. It needs to be clearly and specifically about business tax accounting in Scottsdale — title tag, heading, body copy, and ideally LocalBusiness schema markup that confirms the location and service type. The domain needs some authority — links from the local chamber of commerce, a local business journal, relevant accounting directories. The page needs to load quickly on mobile. And it needs to be more genuinely useful to someone searching that phrase than every competing page already ranked. That last requirement is the hard part, and it is the part no keyword density tool solves for you.

What SEO Actually Does

SEO is the ongoing discipline of making your site easy to crawl, easy to understand, more authoritative in its topical area, and more genuinely valuable to users than competing pages. It is not a one-time project and it is not magic. Sites that rank well over the long term do so because they have built real topical relevance and domain authority, maintained technical hygiene, and earned genuine editorial links. That combination takes time — usually six to twelve months before compounding returns are clearly visible — and it requires strategy, not just a checklist of tactics.

For more on the planning and prioritization side, see my posts on building an SEO strategy and the realistic timeline in how long SEO takes. If you want to understand what this work looks like in practice, the what is SEO post covers the foundational definition.

If you want to talk through how these mechanics apply to your specific site and situation, the SEO consulting page is a good place to start.