How search works
A plain-English introduction to SEO: how Google finds, understands and ranks pages, and which parts you can actually control.
What SEO is
Search engine optimisation, or SEO, is the work of making your pages easy to find, easy to understand and worth recommending. It is not a trick or a secret setting. When it works, Google shows your page to people who are looking for exactly what you offer, and you do not pay for each visit.
Search engines do three jobs: they find pages, they understand and store them, and they decide which ones to show for each search. Almost every SEO task helps with one of those three jobs.
1. Crawling: finding your pages
Google runs programs called crawlers. Its main one is Googlebot. A crawler starts from pages it already knows and follows links to discover new ones, a bit like someone clicking every link on every page.
That has two practical consequences:
- A page nobody links to is hard to find. Link to your important pages from your menu, your homepage or related articles.
- Your sitemap is a map you hand to Google. A sitemap is a file, usually at
/sitemap.xml, that lists the pages you want found. It does not force anything into Google, but it helps.
Your robots.txt file, at /robots.txt, tells crawlers which parts of the site they may visit. A single wrong line there can hide your whole site, which is why the audit checks it.
2. Indexing: understanding and storing pages
After crawling a page, Google reads it: the title, headings, text, links, images (through their alt text) and any structured data. If the page is useful and allowed, Google stores it in its index, a giant library of pages it can show.
A page can be left out of the index on purpose or by accident:
- a noindex tag tells Google not to show the page,
- a canonical tag says "the main version of this page lives at another address",
- very thin or duplicated pages may be skipped because they add nothing new.
If a page is not in the index, it cannot rank, no matter how good it is.
3. Ranking: choosing the best answer
When someone searches, Google picks the pages that best answer that search and orders them. It uses hundreds of signals, but they fall into four groups you can reason about:
- Relevance. Does the page actually answer this search? Words in the title and headings help Google see what a page is about, but the content has to deliver.
- Quality and trust. Is the content accurate, helpful and written by someone who knows the subject? Google calls this E-E-A-T: experience, expertise, authoritativeness and trust.
- Authority. Do other respected sites link to you? A link from another site works like a recommendation. These links are called backlinks.
- Page experience. Does the page load fast, work on a phone, use HTTPS and avoid annoying pop-ups?
No single signal wins on its own. A fast page with nothing useful on it will not rank, and a brilliant article on a site nobody links to may struggle.
The results page is more than ten links
The page Google shows after a search is called the SERP, short for search engine results page. Today it often includes much more than the classic list of ten blue links:
- ads at the top and bottom,
- an AI Overview, a generated answer with source links,
- a featured snippet, a quoted answer from one page,
- People Also Ask, a box of related questions,
- a local pack, the map with three nearby businesses,
- videos, images, shopping results and news.
Each of these features takes attention away from the plain results below it. That is why IMVASA shows which features appear for every keyword. A keyword where an AI Overview answers the question fully may send fewer clicks than its search volume suggests.
Keywords and search intent
A keyword is what someone types into the search box. Search intent is why they typed it. There are four common intents:
| Intent | What the person wants | Example | Page that fits |
|---|---|---|---|
| Informational | To learn something | how to make cold brew | A guide or article |
| Commercial | To compare before buying | best burr grinder | A review or comparison |
| Transactional | To buy or sign up | buy single origin beans | A product or pricing page |
| Navigational | To reach a specific site | brewlab login | Your own page for that thing |
Matching intent matters more than almost anything else. A product page will not rank for "how to make cold brew", however well optimised it is, because people searching that want instructions, not a shop.
What you control, and what you do not
- You control your content, your site's structure, technical health, speed, internal links and structured data. The site audit covers these.
- You influence backlinks, brand mentions and reviews. You cannot force them, but good content and outreach earn them.
- You do not control Google's algorithm updates, what competitors publish, or how many people search. You can only watch and adapt, which is what the tracking tools are for.
How long SEO takes
SEO is slow compared with ads, and that is normal:
- Fixing a technical blocker, such as an accidental noindex, can show results within days of Google recrawling the page.
- A new page on an established site often takes a few weeks to a few months to settle into its position.
- A brand new site usually needs three to six months before search traffic becomes meaningful.
The upside is that rankings compound. A page that earns its place keeps sending visitors without you paying for each one.
Where AI search fits
More and more questions are now answered by AI: Google's AI Overviews, ChatGPT, Perplexity, Claude and Copilot. These answers still rely on web pages, and they cite their sources. How AI search works explains how they choose which pages to cite and what you can do about it.
Next steps
- Run the site audit to see how your site does on crawling, indexing and page experience.
- Learn to pick keywords with Keyword research.
- Follow Your first 30 days.