Which SEO basics do websites actually miss? Eight steps, checked against our audits

Short answer: Most SEO basics come down to text you type: one clear main heading, a title and description that fit, headings that match the questions people ask, and pages that link to each other. In our recent audits, the typed basics fail far more often than the structural ones. About 41.3% of homepages miss our title check and 85.0% have no FAQ section we can find, while only 9.6% have fewer than three internal links.

This guide sets out eight SEO basics. For each one we explain what it is, show what our own audit data says about how common the problem is, and how to check your own site. Where our audit doesn’t measure something, we say so.

Where do these figures come from?

SearchScore has analysed 1,000,000+ sites, but the figures in this guide come from a smaller, current set: 47,542 websites checked with our current SEO checks between 28 October and 20 November 2026, using each site’s latest audit. Two figures (the answer-first and hreflang ones) come from our AI search checks over the same weeks, on 71,750 sites.

Three limits to keep in mind:

1. Does your homepage use the words people actually search for?

The basic: make your homepage title and main heading (H1) match the exact search your customers type, turn your H2s into the questions they ask, and answer the common ones in an FAQ.

What our data shows:

Check (our rule) Sites that fail
Homepage title missing, under 30 or over 65 characters 41.3%
…title over 65 characters 17.0%
…title under 30 characters 23.3%
No FAQ section found 85.0%
No page we read has two or more question headings 73.6%
Homepage doesn’t open with a short paragraph straight after the main heading 39.8%

Separately, our Q3 2026 SAVI report found that 96.6% of websites in its 20,000-site sample publish no FAQ schema. That measures markup, not a visible FAQ section, so the two figures aren’t comparable. Google stopped showing FAQ rich results from 7 May 2026 (see schema markup in 2026), so an FAQ is worth writing for readers first, whether or not you mark it up.

Why it matters: Google says title links are “truncated in Google Search results as needed, typically to fit the device width”, and that it may use heading elements such as the H1 when it writes a title link (Google Search Central). A short, plain title that says what you do, in the words a customer would use, does two jobs at once.

How to check your own site:

  1. Write down the questions customers ask before choosing you. Search Console’s queries report, sales calls and support emails are better sources than a keyword tool on its own.
  2. Compare them with your homepage title, H1 and H2s. Our guide to question headings shows how to rewrite statement headings as questions.
  3. Put a direct answer in the first paragraph under each heading. The answer-first method explains how.
  4. The free audit checks your homepage title length, whether it finds an FAQ section, question headings on the pages it reads, and whether the homepage opens with an answer.

2. Are the technical basics in place: one H1, real links, and pages that render without JavaScript?

The basic: one H1 per page, menus built from real links rather than JavaScript buttons, and pages prerendered so crawlers get the content without running scripts.

What our data shows:

Check (our rule) Sites that fail
Homepage doesn’t have exactly one visible H1 38.7%
Homepage HTML has fewer than three internal links 9.6%

The internal-links check is a stand-in, not a direct test of menus. A menu built from JavaScript buttons usually leaves the raw HTML with few or no ordinary links, which is what this check picks up. It can’t tell you why the links are missing.

On the H1, be precise about what Google says. Google’s SEO starter guide says “There’s also no magical, ideal amount of headings a given page should have”, and lists the number and order of headings among things not to focus on. Our check counts one clear H1 as good practice, because it makes the page’s subject unambiguous to readers, crawlers and the title link. It isn’t a Google rule, and a page with two H1s isn’t penalised for it.

On links, Google is firmer: “Generally, Google can only crawl your link if it’s an <a> HTML element” with an href attribute (Google Search Central). Google also says “Not all bots can run JavaScript”. Our technical SEO for AI search guide covers what that means for AI crawlers.

Speed: our audit doesn’t measure response time or time to first byte, so we have no figure for it. Use PageSpeed Insights or the Core Web Vitals report in Search Console.

How to check your own site:

  1. View the page source (not the inspector) and search for <h1. Count how many a visitor can actually see.
  2. In the same source, check that your main menu items are <a href="..."> links.
  3. Run curl -s https://yoursite.com/ | grep "a sentence from your page". If the sentence isn’t there, crawlers that don’t run JavaScript won’t see it either.
  4. The free audit counts visible H1s and internal links on your homepage and compares the raw HTML with what a browser renders.

3. Do several of your pages share the same title or content?

The basic: find pages that share a title or meta description, and give each its own, or merge pages that are really the same.

What our data shows:

Check (our rule) Sites that fail
Homepage meta description missing, under 80 or over 165 characters 44.7%
…no meta description at all 19.3%
Two or more pages we read return identical HTML before JavaScript runs 9.5% of the sites where we could compare pages

What we don’t measure: our audit doesn’t yet compare titles or meta descriptions across pages, which is the problem this step is about. The duplicate-pages figure above is a different problem: separate URLs that serve the same raw HTML, which usually means a JavaScript app sends one shell for every address. It’s worth fixing, but don’t read it as “shared titles”.

Why it matters: Google says to “Avoid repeated or boilerplate text in <title> elements” (title links), and that “Identical or similar descriptions on every page of a site aren’t helpful when individual pages appear in search results” (snippets). Duplicate content on its own isn’t a penalty, though. Google’s starter guide says it’s “not a violation of our spam policies, but it can be a bad user experience.”

How to check your own site: run a site crawler and sort by title and by meta description to find repeats, and check Search Console’s page indexing report for “Duplicate” exclusions. If traffic dropped after a site change, start with our four quiet technical checks, which cover canonicals and redirects.

4. Do you have one page for each comparison people search for?

The basic: where people search for “X vs Y” or “alternatives to X”, build one page per comparison, to a consistent spec.

What our data shows: nothing. Our audit doesn’t detect comparison pages, so we have no figure.

How to do it well: a comparison page earns trust when it helps the reader choose, even when the answer is a competitor. Our guide to comparison, alternatives and pricing pages covers verdict-first structure, dated facts and declaring that you made the page, with templates you can copy.

5. Do you have hub pages that link related content together?

The basic: group related articles under hub pages, and link related posts to each other.

What our data shows:

Check (our rule) Sites that fail
No content hub found (a blog, guides, resources or similar section) 27.7%
No page we read has three or more links inside its main content 33.6%

How to check your own site: open your three most important articles. Does each link to the hub it belongs to, and to at least two related pages, inside the text rather than only in the menu? The free audit checks for a content hub and for links within the content of the pages it reads.

6. Do you publish data nobody else has?

The basic: publish your own numbers, from your own work, that other sites can’t copy from each other.

What our data shows: our audit doesn’t judge whether data is original. That needs a human.

How to do it well: say how you measured, how many cases, and when. Our guide on customer proof for AI search covers writing results in a form other people can check and repeat. It’s the approach we’ve tried to follow in this guide: every figure comes with its sample and its limits.

7. Have you translated the pages that already work?

The basic: use Search Console’s country data to find where you already get interest, translate the pages that perform, add hreflang, and have a fluent speaker review the result.

What our data shows:

Check Result
Sites with hreflang tags 17.5%
Of those, tags that fail our check (no x-default, or an invalid language code) 28.3%

Our check is stricter than Google here. Google says x-default is “recommended”, not required (localised versions), so a site can fail our check and still be set up acceptably. The rule Google does state firmly is that “Each language version must list itself as well as all other language versions”, and that “If two pages don’t both point to each other, the tags will be ignored.”

What we don’t measure: translation quality, whether you picked the right countries, or Search Console data. Only the tags.

How to check your own site: open the page source of one page and one of its translations, and confirm each lists itself and the other. The free audit flags hreflang tags with no x-default or an invalid language code.

8. Are you publishing on a schedule?

The basic: keep publishing, steadily.

What our data shows: our audit doesn’t measure how often you publish. It does check whether your pages carry dates, which is part of reading your findings, but one audit can’t see a schedule.

How to do it well: pick a pace you can keep, and update the pages that matter rather than only adding new ones. A dated “last updated” line tells readers, and crawlers, that someone still looks after the page.

Which SEO basics are most often missed?

Ranking the checks we can measure, from the most often failed:

Check (our rule) Sites that fail
No FAQ section found 85.0%
No page with two or more question headings 73.6%
Meta description missing or outside 80 to 165 characters 44.7%
Title missing or outside 30 to 65 characters 41.3%
Homepage doesn’t open with a short answer paragraph 39.8%
Not exactly one visible H1 38.7%
No links inside the main content of any page we read 33.6%
No content hub found 27.7%
Fewer than three internal links on the homepage 9.6%
Identical raw HTML on separate pages 9.5%

The pattern is the point of this guide. The structural checks, which a theme or platform usually handles, fail least. The ones that need someone to write something specific (a title, a description, a question, an answer) fail most. That’s our reading of the table, not something it proves: we haven’t measured why sites miss them. These fixes are mostly typing, and nobody else can type them for you.

How do I check these SEO basics on my own site?

The free SearchScore audit checks your homepage title and meta description, visible H1s, FAQ section, question headings, answer-first opening, internal links, content hub, hreflang tags and raw-HTML duplicates, and lists what to fix. It’s a readiness check of your own site: it doesn’t report your Google rankings, and it doesn’t ask any AI assistant about you. It also doesn’t cover speed, titles repeated across pages, comparison pages, original data, translation quality or publishing frequency, so use Search Console and a crawler for those.

Sources

SearchScore data checked 20 November 2026. Google sources checked 20 November 2026.

  1. Google Search Central, Influencing your title links in search results
  2. Google Search Central, Control your snippets in search results
  3. Google Search Central, SEO starter guide
  4. Google Search Central, Make your links crawlable
  5. Google Search Central, Tell Google about localized versions of your page
  6. SearchScore, Q3 2026 SAVI report (FAQ schema figure)
  7. SearchScore, Technical SEO for AI search (JavaScript and AI crawlers)

Frequently asked questions

Does Google require exactly one H1 per page?

No. Google's SEO starter guide says there's no ideal number of headings and lists the number and order of headings among things not to focus on. Our audit flags a homepage that doesn't have exactly one visible H1 because one clear main heading makes it obvious what the page is about, and Google may use a heading when it writes the title link. Treat it as tidiness, not a ranking rule.

What length should a title tag be?

Google sets no limit, but it truncates title links to fit the screen. Our audit passes a homepage title of 30 to 65 characters. Shorter titles often say too little, and longer ones get cut off. More important than length is that each page has its own title describing that page.

Does the free SearchScore audit check page speed?

No. The free audit doesn't measure response time or time to first byte. For speed, use Google PageSpeed Insights or the Core Web Vitals report in Search Console.

Does the audit find duplicate titles across my pages?

Not yet. It checks the homepage title and meta description, and it checks whether several of the pages it samples return identical HTML before JavaScript runs. For titles and descriptions repeated across many pages, use a site crawler or the page indexing reports in Search Console.

Will fixing these basics get my business mentioned by AI assistants?

Nobody can promise that. These fixes make a site easier for search engines and AI crawlers to read and understand. Whether an assistant names your business also depends on what other websites say about you and how each assistant picks its sources. The free audit measures how ready your site is, not how often AI mentions you.

Where do your figures come from?

From websites in our audit corpus that were checked with our current SEO checks between 28 October and 20 November 2026. Most were scanned by us, some were requested by their owners. They aren't a random sample of the web, and each figure is the share of sites that failed our check, using our own thresholds.