Caliency.
Web Development

What Makes a Website SEO-Friendly? A 10-Minute Test

The phrase on every proposal that nobody defines: what Google's documentation says must be built in, what doesn't matter, and how to test it in 10 minutes.

C

Caliency

12 min read
A young web developer in a bright studio office checks a page's source code beside its live preview on a wide monitor, with the mobile version of the same page on a phone by the keyboard and hand-drawn page-hierarchy notes stuck to the edge of the screen.

Almost every website proposal carries the same phrase: "SEO-friendly." Nobody writes down what it means, because there is no standard, no certificate and no definition Google endorses. It is something anyone can say about their own work and nobody checks.

This guide makes it checkable. It covers what Google's own documentation says an SEO-friendly site needs to have built in, the checklist items Google explicitly says don't matter, and a test you can run on any site or proposal in about ten minutes.

What is an SEO-friendly website?

An SEO-friendly website is one built so that Google can find, read and correctly understand its pages, and so that users get a good experience on them.

The distinction matters: an SEO-friendly foundation makes a site rankable, not ranked. Rankings depend on how well the content answers a search, on competitors, and on the trust the site earns from elsewhere, and those are ongoing jobs after launch. The foundation is built once during the project, and it is expensive to change afterwards.

Google describes SEO in plain terms in its own guide: helping search engines understand your content, and helping users find your site (Google Search Central, SEO Starter Guide). The content and authority side of SEO is covered in our complete SEO guide; this piece is only about the site itself.

Much of the web is missing even the basic signals

The web itself shows why "SEO-friendly" is a cheap claim. According to HTTP Archive's Web Almanac 2025, built from a crawl of millions of pages, 98.5 percent of mobile pages have a title tag, but only 67.2 percent have a meta description and only 66 percent have a non-empty H1 (HTTP Archive, Web Almanac 2025 – SEO).

How many pages carry the basic SEO signals0%25%50%75%100%Title tag98.5%Viewport meta tag95.2%Served over HTTPS91.5%Meta description67.2%Canonical tag67%Non-empty H166%Structured data (home pages)50%Hreflang19.7%
Source: HTTP Archive, Web Almanac 2025, SEO chapter (published 15 January 2026, updated 9 June 2026), mobile pages from the HTTP Archive crawl. Structured data is measured on home pages only; the other figures cover home and inner pages. Presence is not correctness: a page can carry a canonical or an H1 that is wrong.

In other words, one page in three is missing at least one of the two most basic signals.

One caveat in the chart deserves attention: presence is not correctness. A page can carry a canonical tag pointing at the wrong address, or an H1 that says nothing about the page. That is why the test below asks not only "is it there?" but "is it right?"

Speed tells the same story: the Core Web Vitals results for 200 popular Turkish sites are in does site speed affect SEO.

What has to be built in during the project

In a bright meeting room, two young colleagues plan a website's page hierarchy as a tree of index cards taped to a glass wall; she places a card while he points at a branch.

Everything below has one thing in common: built in during the project it costs almost nothing, and added after launch it can mean rebuilding part of the site.

The main content should arrive in the server's HTML

Google processes JavaScript-driven pages in three phases: crawling, rendering and indexing. A page waiting to be rendered may stay in that queue for a few seconds, but it can take longer (Google Search Central, JavaScript SEO basics).

The same document says server-side or pre-rendering is still a great idea, because it makes the site faster for users and crawlers and because not all bots can run JavaScript. In practice: if the main text of a page isn't assembled in the browser after the fact, Google doesn't have to wait to see it. How AI crawlers handle this is covered separately in our post on AI crawlers.

URLs should be readable and permanent

Google recommends using readable words rather than long ID numbers where possible, separating words with hyphens, and using words in your audience's language (Google Search Central, URL structure). /services/custom-web-development tells a user and a search engine far more than /page.php?id=482.

The real issue is permanence. Every address that changes after launch needs a redirect, and every missing redirect is a lost page. How that is managed in a rebuild is covered in when to redesign your website.

The mobile version is the site

Google uses the mobile version of a site's content for indexing and ranking, which it calls mobile-first indexing (Google Search Central, mobile-first indexing).

The design consequence is direct: a section that appears on desktop but is dropped on mobile "for space" is largely invisible to Google. The mobile design isn't a shrunken desktop; it is the version Google actually reads.

Speed is decided by build decisions

Google's Core Web Vitals thresholds are explicit: the main content (LCP) should load within 2.5 seconds, interaction responsiveness (INP) should stay at or under 200 milliseconds, and visual shift (CLS) should stay at or under 0.1, measured at the 75th percentile of page loads (web.dev, Web Vitals). Google writes that good Core Web Vitals, along with other page experience aspects, align with what Google's core ranking systems seek to reward (Google Search Central, Core Web Vitals).

The field data shows why this is its own item. In HTTP Archive's August 2026 data, 63.7 percent of US mobile origins pass all three metrics, and across all countries the figure is 53.0 percent (HTTP Archive, Core Web Vitals Technology Report).

Share of mobile sites passing Core Web Vitals, by countryTürkiyeAll countriesUnited States56.3%53.0%63.7%All three passed68.4%65.8%75.0%LCP (loading)78.3%80.1%87.7%INP (responsiveness)
Source: HTTP Archive, Core Web Vitals Technology Report, August 2026 data, all technologies and all rank groups, computed from the report’s own API. Mobile origins tested: Türkiye 218,343, all countries 8,733,157, United States 1,339,066. The field data is the Chrome User Experience Report, which only covers origins with enough traffic from opted-in Chrome users, so small sites are under-represented.

More than a third of US sites fall short of a threshold Google itself sets. Which design decision breaks which metric, and how each is fixed, is covered in our Core Web Vitals guide.

Canonical tags, the sitemap and robots.txt must be set up correctly

These three tell Google which page is the main version, which pages exist, and where it shouldn't go. According to the Web Almanac 2025, 67 percent of pages carry a canonical tag and 85 percent of sites return a successful (200) response for robots.txt (HTTP Archive, Web Almanac 2025 – SEO). The figures look healthy, but the report itself warns that a file existing doesn't mean its rules are correct.

Structured data belongs in from day one

Google uses the structured data it finds on the web to understand the content of a page. It recommends whichever format is easiest to implement and maintain, and says that in most cases that is JSON-LD (Google Search Central, introduction to structured data). How to choose the right schema for company details, services, articles and breadcrumbs is covered in our schema markup guide. If the site is published in more than one language, the hreflang tags that link each version to the others are set up at this stage too; the detail is in multilingual SEO.

How often this infrastructure is missing on popular sites, measured across 200 Turkish domains, is shown in the technical SEO checklist.

What checklists include that Google says doesn't matter

Many "SEO-friendly website" checklists still carry items Google has explicitly called unimportant. Google lists these common misconceptions one by one in its SEO Starter Guide (Google Search Central, SEO Starter Guide):

Common claimWhat Google's documentation says
The meta keywords tag should be filled inGoogle Search doesn't use the keywords meta tag.
Keywords belong in the domain and URLOn their own they have hardly any effect beyond appearing in breadcrumbs.
A page needs a minimum word countThere is no magical word count target, minimum or maximum.
Headings must run H1, H2, H3 in orderGreat for screen readers; from Google Search's perspective the order doesn't matter.
Duplicate content triggers a penaltyInefficient, but not something that causes a manual action; copying others' content is a different story.
E-E-A-T is a ranking factorIt isn't.

This table is useful when reading a proposal. If "meta keywords optimization" or a "word count target" appears as a separate line item, the proposal is out of date.

The heading-order row is easy to misread: unimportant to Google is not the same as unimportant for accessibility. For sites selling into Europe it isn't optional any more; our web accessibility guide explains why.

The 10-minute test: is a site actually SEO-friendly?

A young business owner sits by the window of a bright café, watching a website load on his phone; the laptop on the table shows a performance report with green and orange circular score gauges.

You can run this test on your own site, on a competitor's, or on the site of the agency that sent you a proposal. No paid tools are needed.

  1. Look at the page source. Open the page and view the source with Ctrl+U (Cmd+Option+U on a Mac). Can you read the page's main text there? Is there a <title>, a meta name="description", a rel="canonical" and a single <h1>? If the source shows only an empty container and long lines of script, the content is being built in the browser.
  2. Run PageSpeed Insights on mobile. The top section of a PageSpeed Insights result shows real-user data: does the Core Web Vitals assessment say passed or failed? On new or low-traffic sites that section can be empty, in which case only the lab measurement below it is available.
  3. Open the Rich Results Test. Google's Rich Results Test shows which structured data the home page and an inner page carry, and whether it contains errors.
  4. Open the site on a phone. Are the desktop sections all there on mobile, or have some been removed? Content that isn't on mobile largely doesn't exist for Google either.
  5. Look at the address bar. Are the URLs made of readable words, and is the padlock (HTTPS) there? If you own the site, the URL Inspection tool in Search Console also shows which version of a page Google sees.

If you'd rather run most of these at once, our free SEO checker reports the same items, plus Google's real-user speed data, in a few seconds.

A site that fails any one of these five steps hasn't earned the "SEO-friendly" label. If you have a proposal from an agency, run the test on the agency's own site first; there is no reason to believe it will do on your site what it hasn't done on its own. A fuller audit is described in how to run an SEO audit without a tool.

Does AI search need anything extra?

According to Google, no. Google says there are no additional requirements to appear in AI Overviews or AI Mode, nor other special optimizations necessary. To be shown as a supporting link in those answers, a page must be indexed and eligible to be shown in Google Search with a snippet (Google Search Central, AI features).

That means an SEO-friendly foundation is also the foundation for AI search. The work built on top of it, becoming the source those answers quote, is a separate discipline; our GEO service page and our guide to Google AI Overviews cover that side.

In short

An SEO-friendly website isn't a label; it is a foundation you can check: HTML delivered by the server, readable and permanent URLs, a complete mobile version, speed that clears Google's thresholds, correct canonical tags and sitemap, and structured data from day one.

Those things are cheap to build in and expensive to add later. Items like meta keywords and word-count targets, according to Google's own documentation, shouldn't be on the list at all.

At Caliency, these items are part of the build, whether the site is custom-coded in Next.js or built on WordPress; they aren't sold as a separate "SEO package." You can see what's included and the fixed prices on our custom web development page. If you aren't sure your current site passes the test, send us the address and we'll go through which items are missing with you. If the site is live and the problem is visibility rather than the foundation, our SEO services page covers that work. We measured how often these items are actually met on Türkiye's largest corporate sites: the BIST 100 website study.

Got a project in mind?

We'll scope it and give you a fixed price on one call.

Start a project →