Jump to section▾
Reference
How to read your audit
Every score on your audit means something specific. This page explains what — without jargon, without hand-waving.
How scoring works
Every check produces a score from 0 to 100 and a status color.
Status colors
Passing. The article is doing what it should.
Concerns. Something to look at, not necessarily a crisis.
Failing. Real risk to ranking or reader trust. Fix this one first.
Couldn't run. Usually because data wasn't available (no brand samples, no SERP results). Doesn't count toward your overall score.
The Overall Score is a weighted average across all eight checks. Heavier checks (fact-grounding, helpful content) move it more than lighter ones (technical SEO).
The eight checks
Search Intent Match
What it measures
Whether your article answers the kind of question someone was actually searching for.
Why it matters
Google ranks articles that match search intent. If someone searches "best espresso machines under $500" (commercial intent — they want recommendations to buy), and your article is a generic explainer of how espresso machines work (informational intent), you won't rank no matter how good the writing is.
What it checks for
- •Informational — "how does X work," "what is X"
- •Commercial — "best X," "X vs Y," "top X for Y"
- •Transactional — "buy X," "X coupon code," "X near me"
- •Navigational — searching for a specific brand or site
- •If the article's content type (explainer, listicle, comparison, how-to, sales page) doesn't match the query intent, you'll see a mismatch flag.
Fix it by
Rewriting the article structure to match what the search intent actually wants — add product comparisons for commercial queries, add concrete steps for how-to queries, etc.
Fact Grounding
What it measures
Whether the specific factual claims in your article are supported by authoritative sources currently on the web.
Why it matters
Google's quality raters look at factual accuracy. Wrong facts erode reader trust and can get pages demoted. Worse, if your article tells readers something incorrect, you can hurt them.
What it checks for
- •Extracts up to 5 fact-checkable claims (specific numbers, standards, dates, technical assertions)
- •Verifies each against current authoritative sources
- •Supported — authoritative sources confirm this claim
- •Unsupported — no authoritative source confirms or denies it (claim presented as established but isn't)
- •Contradicted — authoritative sources directly contradict the claim
- •Unverifiable — can't be checked from public sources (often a personal experience or proprietary metric)
Fix it by
For contradicted claims, correct them or remove. For unsupported claims, cite the source you're drawing from, or reframe as your own opinion/methodology rather than established fact.
Helpful Content
What it measures
Whether your article is substantive or fluff. Modeled on Google's Helpful Content classifier — the algorithm system that demoted thousands of AI-generated content farms starting in 2023.
Why it matters
Thin, generic, AI-pattern content gets demoted in search regardless of how many keywords it has. Helpful content stays up.
What it checks for
- •Signal density — substance per word, vs throat-clearing and filler
- •Originality of insight — your own POV vs rehashing the top 10 SERP results
- •Specificity — concrete examples, real numbers, named tools, vs vague generalities
- •AI-tell patterns — repetitive transitions ("moreover," "furthermore"), formulaic intro-body-conclusion structure, throwaway openers ("in today's fast-paced world…")
- •Reader-task focus — does this actually help a reader accomplish something?
- •You'll see specific filler quotes pulled from your article ("examples of filler") so you can find and fix them.
Fix it by
Rewriting flagged filler passages with specific facts, examples, or actionable steps. Cut throat-clearing.
E-E-A-T
What it measures
E-E-A-T stands for Experience, Expertise, Authoritativeness, Trustworthiness — Google's quality framework for evaluating content, especially in "your money or your life" categories (health, finance, legal, anything that could affect someone's wellbeing).
Why it matters
Google's quality raters score E-E-A-T explicitly. Pages with weak E-E-A-T get demoted or ignored. This is the single most common failure mode for AI-assisted content.
What it checks for
- •Experience — first-person markers ("I tested," "we measured," "in our shop"), original photos referenced, dates of personal experience
- •Expertise — appropriate technical depth, accurate domain vocabulary, named credentials
- •Authoritativeness — citations to authoritative external sources, clear positioning of who's writing and why they know
- •Trustworthiness — named author, author bio, transparent methodology, balanced views, disclosure of stakes (affiliate links, sponsorships)
Fix it by
Add a named author byline. Add an author bio. Cite authoritative external sources (peer-reviewed papers, government sites, recognized industry references). Add first-person experience markers where relevant. Make your methodology transparent.
Most common pattern we see
Articles score red on E-E-A-T because they have no author, no first-hand markers, and only internal links. This is fixable in an afternoon.
Originality
What it measures
Whether the article adds a distinct angle, insight, and point of view — or reads like commodity content rehashing what already exists on the topic.
Why it matters
Google has no reason to rank your article above existing pages unless it adds something readers cannot get elsewhere. Formulaic B2B explainers and AI slop score poorly here.
What it checks for
- •Original insight vs rehashed industry consensus
- •Distinct framing, POV, or first-hand perspective
- •Avoidance of formulaic structure and cliché talking points
- •Inferred target query when search intent is available
- •High score — clear differentiation — something only this article says this way
- •Low score — commodity content indistinguishable from generic articles on the topic
Fix it by
Add original data, original POV, or original examples. Cover an angle competitors miss. Aim for genuine differentiation, not paraphrase.
Internal Linking Quality
What it measures
How well your article links to other relevant pages on your own site.
Why it matters
Internal links pass authority between pages, help Google understand site structure, and keep readers on your site longer. Generic anchors and missed links are wasted SEO equity.
What it checks for
- •Link count — too few internal links signals an orphan article
- •Anchor descriptiveness — "click here" and "read more" are bad anchors; descriptive anchors are good
- •Contextual relevance — does each link actually relate to the topic at hand?
- •Missed opportunities — are there obvious pillar pages on your site that you should have linked but didn't?
Fix it by
Add 3–5 contextual internal links per article with descriptive anchor text pointing to your most authoritative existing pages on related topics.
Brand Voice Fit
What it measures
Whether the article sounds like the rest of your site.
Why it matters
Brand voice consistency is a trust signal for readers and a recognizable pattern for repeat visitors. AI-written content often defaults to a generic explainer voice that doesn't match the brand.
What it checks for
- •Tone — formal vs casual, authoritative vs conversational
- •Sentence rhythm — complexity and length patterns
- •Vocabulary register — does the article use the same kind of language as your other writing?
- •Signature phrases — distinctive moves your brand makes that this article uses (or doesn't)
Fix it by
Rewrite the article matching your brand voice, or feed your brand voice samples to your AI writer up front.
Important caveat
High consistency doesn't mean good content — it just means consistent. Low consistency is the more diagnostic finding.
Technical SEO
What it measures
The structural and metadata elements that make your page indexable and well-presented in search results.
Why it matters
These are the easiest wins. They're fully under your control and a few minutes of work each.
What it checks for
- •Title tag — present, 30–60 characters
- •Meta description — present, 120–160 characters
- •H1 tag — present, exactly one, matches title intent
- •Heading hierarchy — H2s under H1, H3s under H2, no skipped levels
- •Image alt text — every image has descriptive alt text
- •Schema.org Article markup — present and valid
- •Canonical URL — set correctly
- •OpenGraph tags — og:title, og:description, og:image for social sharing
- •Word count — at least 600 (penalty below this)
Fix it by
The report lists exactly what's missing and what to set it to. Usually a 20-minute fix per article.
Other terms you'll see
- SERP
- Search Engine Results Page. The page Google shows after a search. We use this to compare your article to what's currently ranking for the same query.
- Cosine Similarity
- A mathematical measure of how similar two pieces of text are, from 0 (completely different) to 1 (identical). We use it in the originality check.
- Embedding
- A numeric representation of text that captures meaning, not just word matches. Two pages with different wording but the same meaning will have similar embeddings. This is what makes the originality check more sophisticated than simple plagiarism detection.
- Pillar Page
- A foundational article on your site that covers a topic in depth and serves as the hub other related articles link to. Good internal linking points to your pillars.
- Quality Raters
- Human contractors Google uses to evaluate search results. Their guidelines (published as the "Search Quality Rater Guidelines") shape how Google's algorithm learns to score pages. E-E-A-T comes directly from this document.
- Helpful Content Update
- A Google algorithm update first launched in August 2022 (with ongoing refinements) that demotes content created primarily to rank in search rather than to help readers. AI-generated content farms have been particularly affected.
- YMYL (Your Money or Your Life)
- Google's term for topics that could materially affect a reader's wellbeing — health, finance, legal, safety. Google holds YMYL content to a higher E-E-A-T standard.
- Indexed
- A page is "indexed" when Google has added it to its searchable database. Pages can be crawled but not indexed — typically when Google's quality classifier decides the content isn't worth surfacing.