I’ve sat in enough client kick-off calls to know the first question people ask when it comes to getting ranked in search engines is almost always “what keywords should we target?”
This is the wrong first question.
Before keywords, before content, before any of it, your site has to survive three separate stages: Google has to find it, understand it, and decide it deserves a spot above everyone else’s. And with AI changing the game, taking that step back is even more important than ever.
I’ve spent ten years running SEO and content for businesses scaling from nothing to millions in revenue, and the operators who get this right are the ones who understand the three stages well enough to diagnose their own problems instead of guessing at title tags and keywords.
The three things Google does with every page on the internet
Every result you’ve ever seen in Google went through the same three-stage process:
- Crawling — discovering the page exists
- Indexing — working out what it’s about
- Ranking — deciding where it belongs against everything else
A problem at any one stage kills your chances, regardless of how good the other two are. A beautifully written page nobody’s linked to won’t get crawled. A crawled page stuffed with duplicate content won’t get indexed. An indexed page with no authority behind it won’t rank. Here’s each stage on its own.
Stage 1: Crawling — can Google even find you?
Google uses automated bots, Googlebot being the main one, that move around the web following links from page to page, continuously, forever.
Think of it like a delivery driver working from a list of addresses. If your page isn’t on the list and nothing links to it, the driver never shows up. It doesn’t matter what’s inside.
How it actually works:
- Googlebot starts from a set of known URLs, previous crawls plus whatever you’ve submitted via a sitemap
- It visits those pages, finds the links on them, and follows those links to discover new ones
- It repeats this endlessly
The bit that catches people out: if nothing on your site links to a page, Google may never find it at all. A brilliant piece of content sitting with zero internal links pointing to it is invisible. Not poorly ranked. Invisible.
What actually controls crawling
Your sitemap. The file at yoursite.com/sitemap.xml lists your important URLs. Submit it through Google Search Console so Google has a map instead of having to stumble across everything by accident.
Your robots.txt file. This tells crawlers what they’re allowed to access. Go check yoursite.com/robots.txt right now. If it says “Disallow: /” you’ve accidentally blocked your entire site from Google, and I’ve seen this happen on live client sites more than once.
Internal linking. Pages buried three or four clicks deep with nothing pointing to them take far longer to get discovered, if they’re discovered at all. Every page you actually care about needs something linking to it. This is one of the fundamentals we cover in our complete guide to website design for growing businesses.
Site structure. A logically organised site is easy to crawl efficiently. A sprawl of orphaned pages with no clear hierarchy wastes the crawler’s attention, and on larger sites that attention is genuinely limited, what’s known as crawl budget.
Server response time. A slow or unstable server frustrates crawlers, who have a limited amount of time budgeted per site. Slow enough, and they give up before finishing.
Crawling problems that quietly kill rankings
- Blocked by robots.txt — the “Disallow: /” issue above
- Noindex tags left on pages that should be public
- Orphaned pages with no internal links pointing to them
- Server errors that stop crawling outright
- Slow response times that cause crawlers to abandon the attempt
Check Google Search Console’s Coverage report if you want to see exactly which of these, if any, apply to your site.
Stage 2: Indexing — does Google understand what you’ve made?
Once a page is crawled, Google analyses it and stores what it’s learned in a vast database: the index. This is where “discovered” becomes “understood.”
What Google actually looks at
Everything. Text content, image alt text and file names, title tags and meta descriptions, heading hierarchy, internal and external links plus their anchor text, structured data, page speed, mobile-friendliness, and the broader topic and entity relationships the content sits within.
How “understanding” has changed
Google in 2010 matched keywords through fairly simple pattern recognition. Google in 2026 is doing something closer to reading comprehension: working out searcher intent, whether a page comprehensively covers a topic, how entities relate to one another, and the semantic meaning behind a query rather than just the literal words in it.
Search “apple” and Google works out whether you mean the fruit or the company based on your history, location, and a dozen other contextual signals, without you specifying either.
E-E-A-T: the framework Google uses to judge quality
You’ll see this acronym everywhere in SEO writing and rarely see it explained practically. Here’s what each part actually means for a business website:
- Experience — first-hand knowledge shows. A product review from someone who’s actually used the thing outranks a generic description written by someone who hasn’t.
- Expertise — relevant knowledge or credentials. For medical or financial content especially, Google is increasingly asking: is the person behind this actually qualified to say it?
- Authoritativeness — do other credible sites in your space link to and reference you? Being cited elsewhere is one of the strongest signals you can send.
- Trustworthiness — HTTPS security, visible and accurate contact information, claims that are backed up rather than asserted.
None of this is about gaming a system. It’s Google trying, imperfectly, to measure whether you actually are what you claim to be.
The honest way to score well here is to genuinely be credible: publish real case studies (take a look at how this played out for Hyform Performance and Fundingwise), show real work, keep your contact details visible and accurate. It’s also worth remembering this only works if the brand behind it is actually clear and consistent — see our piece on why your logo isn’t your brand for more on that distinction.
What stops a page being indexed at all
- Noindex tags explicitly telling Google not to
- Robots.txt blocks applied during the indexing phase
- Duplicate content that substantially overlaps with another page
- Thin content offering little unique value
- Technical errors, particularly JavaScript rendering issues that hide content from crawlers
- Canonical tags pointing to a different page as the “real” one
Search “site:yoursite.com” in Google to see exactly what’s currently indexed from your domain. It’s a two-second check that tells you more than most audits.
Stage 3: Ranking — why some pages win and others don’t
Once a page is indexed, Google has to decide, in milliseconds, where it sits against every other indexed page competing for the same search.
This is where the widely quoted “200+ ranking factors” figure comes from. Nobody outside Google has the full list, and it shifts constantly. But the major categories are well understood.
Content relevance and quality, still the biggest lever
Does the page match what the searcher actually wants? An informational search wants a comprehensive answer. A transactional search wants a product page. A commercial search wants comparison or review content. Serve the wrong content type for the intent, and quality won’t save you, you’re simply the wrong answer to the question being asked.
Within the right content type, Google is weighing depth against surface-level coverage, originality against rehashed information, and genuine usefulness against content padded out to hit a word count. Keywords still matter, in titles, headers, and body copy, alongside natural variations and synonyms, but stuffing them in unnaturally does more harm than good. If you want a structured approach to this, our guide on keyword research that actually drives results walks through it in more depth.
Technical performance
Page speed is a confirmed ranking factor. Mobile-friendliness matters more than desktop, because Google indexes mobile-first. Core Web Vitals, the metrics measuring loading speed, interactivity, and visual stability, all feed in directly. HTTPS is a baseline requirement at this point, not a nice-to-have. Getting the foundations right here usually comes down to how the site itself was built, something we go into in our website design guide.
Backlinks, still doing more work than most people credit
A backlink is effectively a vote of confidence from another site. One link from a genuinely relevant, authoritative publication outweighs a hundred from spam directories, and Google’s gotten considerably better at telling the difference.
What actually counts as a good backlink: it comes from an authoritative, topically relevant site, it was earned editorially rather than paid for or exchanged, it sits in a relevant context, and it comes from a domain genuinely different from your last twenty links rather than the same source repeated. If you’re tracking this yourself, our roundup of the 10 best SEO tools covers the platforms worth using for backlink monitoring alongside everything else.
User experience signals
Google watches how people behave once they’ve clicked through. Click-through rate tells it whether searchers are choosing your result over the alternatives. Dwell time and bounce rate tell it whether what they found matched what they were promised. Write a misleading title to inflate your click-through rate, and the immediate bounce that follows tells Google exactly what it needs to know.
What’s changed about search in 2026
AI Overviews are now a fixture, not a novelty. Google generates summarised answers at the top of many searches, pulling from multiple sources. Traditional organic results have been pushed further down the page as a result. The practical shift: being cited as a source within an AI answer now matters alongside your standard ranking position, which means comprehensive, clearly structured, genuinely authoritative content matters more than ever, not less.
Zero-click searches are the norm for a lot of queries. Featured snippets, knowledge panels, and AI overviews increasingly answer the question directly on the results page. This means optimising for featured snippet placement, leaning into branded search visibility, and accepting that some content is doing its job by building authority even when it doesn’t generate a click.
This is especially true for smaller, local businesses (see our guide to local SEO for small businesses for how to win visibility here without competing head-on for the biggest terms).
Entity-based search has replaced pure keyword matching. Google understands that “Tim Cook” is the CEO of “Apple,” a company headquartered in “Cupertino,” competing with named others, and it uses those relationships to inform results.
Structured data, clear author information, and sensible internal linking between related concepts all help Google place you correctly within that web.
Why Google works the way it does
Google’s revenue comes from advertising, and that shapes the whole system whether it’s acknowledged or not. Good organic results keep people using Google. Ads pay the bills. Both need to coexist, which is why user experience genuinely helps your rankings, established brands get a certain benefit of the doubt, and commercial searches are typically the most ad-heavy and hardest to win organically.
The practical takeaway: for highly commercial keywords with heavy ad competition, paid search may simply be the faster route in. SEO does its best work on informational queries and longer-tail terms where the ad landscape is thinner and intent is easier to serve directly. This is also why SEO rarely works well in isolation — it needs to sit inside a wider marketing system rather than standing alone.
10 SEO signals worth your attention, ranked
Based on what Google states publicly and what shows up consistently in performance data:
- High-quality, genuinely relevant content
- Backlinks from authoritative, relevant sites
- Technical performance, speed and mobile experience specifically
- User experience signals
- Domain authority built up over time
- Proper keyword targeting, used naturally
- Content comprehensiveness
- E-E-A-T signals
- Structured data and semantic markup
- Freshness, particularly for time-sensitive queries
Notably absent from that list: meta keywords (ignored since 2009), social shares (indirect influence at best), exact-match domains (no longer a shortcut), and keyword density (a stuffed page gets penalised, not rewarded).
What do modern search results contain
Ten blue links stopped being the whole picture years ago. A single search results page can now contain:
- paid ads,
- a local pack,
- a featured snippet,
- a “People Also Ask” box,
- a knowledge panel,
- image and video results,
- and an AI overview,
Today, all of these get listed before you reach a standard organic listing.
The goal isn’t just “rank first for your target keyword” because that first listing is often 4 or five items down the page today. You need to get cited in the AI summary, featured in the people also ask box, have multi-media content (YouTube content is good for this), and ideally get highlighted in the featured snippet.
This means adapting your content, and improving off-page SEO as well as on-page SEO.
SEO you control vs what you don’t
| Category | Examples |
|---|---|
| Full control | Content quality, technical performance, site structure, keyword targeting, internal linking, page speed, structured data |
| Partial control | Earning backlinks (you can’t force them), user behaviour signals (you can only build the conditions for good ones), branded search volume |
| No control | Algorithm updates, competitor activity, market shifts, SERP feature changes |
The strategy follows from that table: master the first column, build the conditions for the second, and stop losing sleep over the third.
SEO misconceptions
“Google’s algorithm has it in for my site.”
It’s automated. It doesn’t hold a grudge. If you’re not ranking, there’s a specific technical or quality reason, and it’s usually findable.
“I need to resubmit my site every time I make a change.”
No. Google recrawls on its own schedule. Update the content and let it work.
“More pages means better rankings.”
Only if every page earns its place. Ten genuinely useful pages will consistently beat a hundred thin ones.
“I can trick the algorithm if I’m clever enough.”
Might work for a while. The correction, when it comes, tends to cost far more than the short-term gain was worth.
“Get the technical SEO perfect and rankings follow automatically.”
Technical SEO is what allows your content to be crawled and indexed properly in the first place. It’s the entry ticket, not the result. Content quality, authority, and links are what actually determine where you land.
What to do when Google updates its algorithm?
Google runs three or four major core updates a year plus thousands of smaller tweaks. You could spend hours re-optimising content and pages. But, my honest take here is you shouldn’t need to.
Instead aim to make your content is genuinely useful and the experience around it as solid as possible. Google wants to reward the best, most authoritative content. So, that’s what you should aim to create.
If an update does knock your rankings, it’s usually surfacing an underlying quality issue that was already there, not creating a new problem out of nowhere. Fix what the update revealed rather than trying to reverse-engineer the update itself.
Your search engine ranking checklist
Can Google crawl you?
- Sitemap submitted through Search Console
- robots.txt checked and not blocking anything important
- Server stable and fast
- Internal linking connects every page that matters
Can Google index you?
- No stray noindex tags on important pages
- Content genuinely unique, not thin or duplicated
- Canonical tags pointing where they should
- No unresolved duplicate content issues
Are you set up to rank?
- Content built around real search intent, not just a keyword
- Keywords used naturally, not stuffed
- Backlinks being earned from relevant, credible sites
- Site fast and genuinely usable on mobile
- E-E-A-T signals visible: real case studies, real credentials, real contact details
Creating content that ranks in search engines: final thoughts
Once you understand that Google has to crawl a page before it can index it, and index it before it can rank it, most SEO advice stops sounding like a list of arbitrary rules and starts making sense as a set of consequences.
Weak internal linking isn’t a minor issue, it’s a crawling problem. Thin content isn’t just “not great,” it’s an indexing problem. No backlinks isn’t a technicality, it’s a ranking and authority problem. Once you can place a symptom at the right stage, you know exactly what to fix, instead of guessing at title tags and meta descriptions and hoping you stumbled onto a winning formula.
If you want a second pair of eyes on your content and SEO strategy, have questions around optimising for AI visibility, or need some help improving your technical SEO book in a chat with us, we’d love to help.
We offer a free 30-minute consultation, no commitment.
