What the audit checks
Three scored dimensions, a look at the market around your site, and a Fix Pack for everything it finds. Every check has a published weight, so any score can be rebuilt by hand from the checklist in the report.
How an audit runs
Discover
Reads the homepage,
robots.txtand the sitemap, then builds an inventory of every URL from the sitemap, navigation, footer and in-page links, tagging each by type. It also requests a nonsense URL to confirm the site returns a real 404.Verify in real HTML
Titles, canonicals, robots tags, every JSON-LD block, the heading outline, images and links are captured from each page's actual code rather than a text summary, because text converters routinely strip schema and head tags.
Market check
Maps the queries a buyer would use to the page that should rank for each, checks who actually ranks, and asks AI search a set of buyer questions to see how often you're cited.
Score
Each check is marked Pass, Partial, Fail or not applicable, then weighted into a score out of 10 for each dimension.
Fix Pack
Every finding gets a fix number, a priority and an owner, and the fix itself: copy, schema, code or step-by-step instructions.
Report and check
Builds the report and code bundle, then re-checks its own work: every schema block parsed, every title and description re-counted, every claim in new copy traced back to the site.
Full or quick
A full audit reads every content page and fixes everything. A quick audit reads the homepage plus up to five key pages, reports the top ten findings, and writes fixes for critical and high-priority items only. It runs a full audit unless you ask otherwise.
SEO: technical and on-page
Whether search engines can reach your pages, understand each one, and find a page for everything you sell.
| Check | Weight |
|---|---|
| Every indexable page returns 200, has a self-referencing canonical, and has no stray noindex | 3 |
| Unique titles, 30–60 characters, primary term near the front | 3 |
| Unique meta descriptions, 70–160 characters, specific and with a call to action | 2 |
| Exactly one descriptive H1; logical H2/H3s containing search terms | 2 |
| Clean, readable URLs; real 404s; no redirect chains | 1 |
| Sitemap declared, complete, with truthful lastmod | 1 |
| Every page reachable from the nav or body links; descriptive anchor text | 2 |
| Commercial page for every core offer and priority query | 3 |
| Alt text on content images; modern formats; sensible weights | 1 |
| Open Graph and Twitter tags with page-specific images | 1 |
| Server-rendered primary content; performance measured, or flagged as unmeasured | 1 |
GEO: AI citation readiness
Whether AI search tools have enough clear, verifiable information about you to cite you, and whether they actually do.
| Check | Weight |
|---|---|
Organisation entity: schema with @id, logo, contact, address, sameAs, founder, founding date | 3 |
| Real, named experts with credentials, profile links and Person schema | 3 |
| Verifiable proof: case studies with results, attributed testimonials, client links | 3 |
| Fact-dense copy (prices, timelines, specs, numbers) stated plainly | 2 |
| Original expertise content (guides, data, point of view) | 3 |
| Consistent brand name across the site, schema and social profiles | 1 |
| AI crawlers allowed; content in the HTML; llms.txt present | 1 |
| GEO probe citation rate (50% or more Pass, 1–49% Partial, 0 Fail) | 3 |
The citation probe asks 8 to 12 natural buyer questions through web search and records how many answers cite or mention you, which sources are cited instead, and what those sources have that your site doesn't.
AEO: answer extraction
Whether your answers are shaped so that answer boxes, voice assistants and AI summaries can lift them cleanly.
| Check | Weight |
|---|---|
| Direct 40–60 word answers under question-style headings | 3 |
| FAQPage schema that matches visible FAQs on every page that has them | 2 |
| Lists and real HTML tables for steps and comparisons | 2 |
| Clear definitions ("X is…") for core terms | 1 |
| Long-tail and pre-purchase questions covered | 2 |
| Local answer data: hours, area served, phone, map | 1 |
| Local-language answers where the market uses them | 1 |
How the scores work
Each check scores 1 for Pass, 0.5 for Partial and 0 for Fail. Checks that don't apply are left out rather than counted against you. A dimension's score is the weighted points earned divided by the weighted points available, times ten.
| Score | Status |
|---|---|
| 8.0 – 10 | Strong |
| 6.0 – 7.9 | On Track |
| 4.0 – 5.9 | Needs Work |
| Below 4.0 | Critical |
The checklist results are printed in the report, so a score can always be traced back to the checks behind it. It's a structured judgement, not a Google metric.
The market check
- Intent map. The 10 to 20 queries a buyer would type, each mapped to the page that should rank for it. A query with no page is a content-gap finding.
- Search reality. The top queries run through web search, recording who ranks and whether lists, tables, local results or video win.
- Citation probe. How often AI search cites you, and who it cites instead.
- Local signals. For a local business: consistent name, address and phone, opening hours, a Google Business Profile link, and whether the site serves the language the market searches in.
How findings become fixes
Every finding gets a fix number (F1, F2…), a priority and an owner. Critical means it blocks indexing or leaves a core offer unable to rank. High is a major trust, entity or content gap. A quick win takes under thirty minutes.
Each fix then arrives in one of six forms:
| Fix type | What you get |
|---|---|
| New page or rewrite | A spec table with title, description and H1 counted to the character, full body copy with every heading tagged, FAQs, a call to action, internal links, and the page's JSON-LD. |
| Metadata or copy edit | The current text quoted beside the replacement, with character counts. |
| Schema | Complete JSON-LD, linked through a single organisation ID. |
| Code or config | The current snippet, the replacement, and the file it lives in. |
| Check or process | Numbered steps in the named tool, such as Search Console or Google Business Profile. |
| Content plan | One complete article plus briefs for the next eight to ten. |
The rule that keeps it honest
New copy and schema use only facts published on your site or given by you. Anything
unknown is left as a [PLACEHOLDER] and listed for you to fill in. It doesn't
invent statistics, results, reviews, dates or credentials, and it won't mark up your own
testimonials as review stars, which Google treats as ineligible.