Website Check: The Ten Areas of a Technical SEO Audit Explained

The ten areas at a glance

The order is deliberate. Whatever fails in the first area makes all the others secondary: a blocked page with a perfect title still does not appear.

1. Indexability and technical setup

What is checked: whether the home page responds normally, whether http redirects to https and the variants with and without "www" redirect to one main address, the robots.txt file, the sitemap, error pages, redirect chains, the preferred address of each page (canonical) and the references between language versions (hreflang).

Typical finding: robots.txt responds with a server error. Google writes: "For the first 12 hours, Google stops crawling the site but keeps trying to fetch the robots.txt file." After that, Google uses the last good version for up to 30 days (Google Search Central: robots.txt specification, accessed 31 October 2026). If the file is simply missing (status 404), Google treats the site as if there were no crawl restrictions.

Action: serve the file so that it responds with status 200. For redirects, send every old address straight to the target. Google's crawlers follow "up to 10 redirect hops" by default (Google Search Central: HTTP status codes, accessed 31 October 2026), and Google uses permanent redirects (301, 308) as a signal that the target should be canonical (Google Search Central: Redirects, accessed 31 October 2026). Google calls the canonical tag "a strong signal", not an obligation (Google Search Central: Canonical). For language versions: "If two pages don't both point to each other, the tags will be ignored." (Google Search Central: Localized versions)

2. On-page

What is checked: page title, meta description, main heading (H1), the heading structure, the language declaration, the mobile setup (viewport) and the preview image for shared links.

Typical finding: many pages carry the same title, often the company name. In search results they look identical. Google recommends titles that are "descriptive and concise" and "distinct text that describes the content of the page". According to Google, there is no fixed length limit, but long titles are "truncated in Google Search results as needed" (Google Search Central: Title links, accessed 31 October 2026).

Action: give every page its own title that names the service and, where it fits, the location first. The same goes for the meta description: Google only uses it sometimes, and "Identical or similar descriptions on every page of a site aren't helpful" (Google Search Central: Snippets, accessed 31 October 2026).

3. Content

What is checked: pages with very little text, pages with duplicate text and an outdated year in the copyright notice.

Typical finding: service pages with three sentences and a contact button. There is no minimum word count. Google asks whether you write to a word count because Google supposedly prefers one, and answers: "(No, we don't.)" (Google Search Central: Helpful content, accessed 31 October 2026). A note about thin text is a reason to take a look, not a rule.

Action: measure the page against Google's self-assessment questions, for example: "Does the content provide a substantial, complete, or comprehensive description of the topic?" Answer first, then details, prices or process, examples. Merge duplicate pages or point them to the original with a canonical tag.

4. Internal linking

What is checked: pages without an incoming link, pages that can only be reached after many clicks, meaningless link texts and links that pass through redirects.

Typical finding: a service page is in the sitemap, but no menu item and no text links to it. Google's rule: "Every page you care about should have a link from at least one other page on your site." Google lists link texts such as "Click here" or "Read more" as too generic (Google Search Central: Link best practices, accessed 31 October 2026).

Action: link from thematically related pages, with a link text that names the target. This also helps people using screen readers: under WCAG 2.2, success criterion 2.4.4, the purpose of a link should be clear from the link text or its context (W3C: WCAG 2.2, accessed 31 October 2026).

5. Performance

What is checked: the size of the page code, the server response time, image formats and whether images have fixed dimensions.

Typical finding: images without width and height. The text jumps down as soon as an image loads. Google measures such jumps with the CLS metric and names as the target "a CLS score of less than 0.1", together with LCP within 2.5 seconds and INP below 200 milliseconds (Google Search Central: Core Web Vitals, accessed 31 October 2026). According to Google, good values, along with other page experience aspects, align "with what our core ranking systems seek to reward"; that is not a guaranteed ranking.

Action: set width and height on every image, serve images as WebP or AVIF in suitable sizes. HELFERLAIN SEO does not yet measure the Core Web Vitals themselves, because PageSpeed is not connected yet. Use the tools Google names on the linked page.

6. Structured data

What is checked: whether the home page contains machine-readable company details (Organization or LocalBusiness), whether these details are error-free, navigation paths (breadcrumb), reviews and markup for FAQ or how-to content.

Typical finding: a syntax error in the JSON-LD block, such as a missing comma. The whole block is then unreadable. Google recommends JSON-LD as the format and warns against marking up information that is not visible to the user (Google Search Central: Structured data, accessed 31 October 2026). According to Google, Organization markup on the home page helps it understand the organisation's administrative details and disambiguate it in search results (Google Search Central: Organization).

Action: fix errors and check with the Rich Results Test. Two things not to plan anew: FAQ markup, because "The FAQ rich result feature is no longer shown in Google Search results", and how-to markup, whose documentation Google removed in September 2023 because that rich result is no longer shown (Google Search Central: Documentation updates, accessed 31 October 2026). And your own reviews: pages with LocalBusiness or Organization markup where the reviewed entity controls the reviews about itself are "ineligible" for star reviews (Google Search Central: Review snippet, accessed 31 October 2026).

7. Images and accessibility

What is checked: images without a text description (alt text) and form fields without a label.

Typical finding: team photos and project images without alt text. People using screen readers do not learn what is shown. Google uses alt text "along with computer vision algorithms and the contents of the page" to understand images, and warns against filling it with keywords (Google Search Central: Image SEO, accessed 31 October 2026).

Action: describe content images briefly, mark decorative images with an empty alt text. WCAG 2.2, success criterion 1.1.1, requires a text alternative for non-text content that "serves the equivalent purpose". Give every form field a visible, associated label.

8. Local SEO

What is checked: how many different phone numbers appear on the website and whether the address is stored as machine-readable business information (LocalBusiness).

Typical finding: three spellings of the same number and an old number in the footer. Google's Business Profile Help says: "Businesses with complete and accurate info are more likely to show up in local search results." According to Google, "Local results are mainly based on relevance, distance, and popularity" (Google Business Profile Help, accessed 31 October 2026).

Action: decide which numbers are valid and write them the same way everywhere: website, Business Profile, directories. For the markup, Google advises: "Use the most specific LocalBusiness sub-type possible" (Google Search Central: Local business, accessed 31 October 2026).

9. Trust

What is checked: whether the home page links to a legal notice (in German: Impressum) and a privacy policy. This is an inventory, not legal advice; a technical audit does not check whether the content is legally sufficient.

Typical finding: the legal notice can only be reached via the contact page. In Germany and Austria this matters legally. Germany's § 5 DDG requires providers of digital services offered commercially, usually for a fee, to keep certain information permanently available, easily recognisable and directly accessible (gesetze-im-internet.de: § 5 DDG, accessed 31 October 2026); Austria's § 5 ECG has a similar rule (RIS: § 5 ECG, accessed 31 October 2026). At EU level, Article 5 of the e-Commerce Directive obliges Member States to ensure that service providers make their name, address and contact details, including an email address, "easily, directly and permanently accessible" (EUR-Lex: Directive 2000/31/EC, accessed 31 October 2026). Elsewhere, a page called "legal notice" is less common, but visitors still look for who runs the site and how to reach them. On experience, expertise, authoritativeness and trust (E-E-A-T), Google writes: "Of these aspects, trust is most important." (Google Search Central: Helpful content) Google's self-assessment also asks: "Is it self-evident to your visitors who authored your content?"

Action: put links to contact details, privacy policy and, where required, the legal notice in the footer of every page, and have the content checked by a lawyer in your country.

10. AI visibility

What is checked: whether 13 AI and search crawlers are allowed to read the website according to robots.txt, outdated identifiers, content that only appears with JavaScript, an llms.txt, and whether the company name is spelled consistently.

Typical finding: the main content only appears in the page code once JavaScript runs. Google itself recommends server-side rendering because it makes pages faster and "not all bots can run JavaScript" (Google Search Central: JavaScript SEO, accessed 31 October 2026). Whether their crawlers execute JavaScript is not stated by OpenAI or Anthropic in their crawler documentation (OpenAI: Crawlers; Anthropic Help Center, both accessed 31 October 2026). Only what is in the page code without JavaScript is certain to be read.

Action: serve the main content server-side or pre-rendered. Which crawlers you allow is a policy decision; the trade-offs are covered in AI crawlers in robots.txt. An llms.txt is optional and has no proven effect, see llms.txt: what the file can do.

What a technical audit does not answer

A technical audit shows what stops search engines from reading and understanding a site. It says nothing about where a page ranks today, how often a term is searched or how many other websites link to it. Nor is indexing guaranteed: even a sitemap "doesn't guarantee that all the items in your sitemap will be crawled and indexed", Google writes (Google Search Central: Sitemaps, accessed 31 October 2026). And about agencies that promise first place: "If they guarantee you that their changes will give you first place in search results, find someone else." (Google Search Central: Do you need an SEO?, accessed 31 October 2026)

Where Helferlain stands

HELFERLAIN SEO checks up to 200 pages per analysis in exactly these ten areas and explains for every finding why it matters and what to do. Not included are load time measurements, search volume, rankings and backlinks. How to turn this into a report for the client meeting is explained in SEO audit in the first meeting. The free website check gives an extract for up to 15 pages without signing up; everything else is on the SEO audit page.