Most content audits are inventory projects. They tell you how many posts you have, which ones get traffic, and what keywords they rank for. That’s useful data. It’s not the same as knowing whether your content is working.
A piece can rank, get traffic, and accomplish nothing. It can be well-written, well-optimized, and strategically invisible. It can pass every technical SEO check and still give readers no reason to trust the source, give AI systems no specific claim to cite, and give the brand no authority it didn’t already have before the piece was published.
The Content Authority Audit is a different kind of diagnostic. It doesn’t measure whether content is technically correct. It measures whether it earns the trust of readers and the citation of AI systems — the two audiences that now determine whether a piece of content actually does anything in the world.
What content authority actually means
Authority is not traffic. Traffic tells you whether people arrived at the page. Authority is what they concluded when they got there — whether they decided this source knows what it’s talking about, is worth coming back to, is worth sharing, is worth citing.
That conclusion is made by three audiences simultaneously: human readers deciding whether to trust and engage, search engines modeling whether the site is a reliable expert source on a topic, and AI systems deciding whether this content is specific and credible enough to cite in a synthesized answer.
All three audiences make the same underlying judgment, and they make it based on the same signals. A piece with genuine information gain — something new, specific, and verifiable — earns trust from all three. A piece that summarizes what everyone else has already said earns trust from none of them, regardless of how well-structured it is.
The Content Authority Audit makes those signals explicit and scoreable.
The eight dimensions
The audit scores any piece of content across eight dimensions, each on a 0–4 scale, for a maximum of 32 points.
1. Information Gain
Does this piece add something that doesn’t already exist? Not “is it accurate” — accuracy is the floor, not the ceiling. The question is whether a reader who has already read several pieces on this topic would find something here they couldn’t get elsewhere.
Information gain comes from four sources: original data or research, first-person accounts of actually doing the thing, a synthesis that surfaces a non-obvious connection, or a level of specificity that generic coverage doesn’t reach. A piece with none of these is content the AI can generate on demand. There’s no reason to click through to find it.
Score 0: Everything in this piece exists in nearly identical form elsewhere.
Score 4: This is the piece to cite — it contains something no one else has.
2. Author Authority & Attribution
Is there a specific, credentialed, verifiable person behind this content? Not a brand. Not “the team.” A person, with credentials specific to the topic, documented in a way that a search engine can model as an entity and a reader can evaluate as a source.
“Jacob Clifton, who spent fourteen years as a staff writer at Television Without Pity and has written 40 books” is more citable than “The Clifton Creative Team” — not because the first is more impressive but because it’s specific, verifiable, and attributable. AI systems prefer specificity. Anonymous or corporate-attributed content gets paraphrased without credit or ignored.
Score 0: No attribution — published by a brand or no one.
Score 4: Author is a named entity with cross-platform consistency and Person schema with sameAs links.
3. Specificity & Evidence Quality
Are the claims in this piece attributable and verifiable, or are they the kind of thing anyone could say? “Studies show” without a named study is not evidence. “According to the 2024 HubSpot State of Marketing Report” is. The difference between them is the difference between a claim that earns trust and one that quietly erodes it.
AI can generate floating statistics endlessly. It cannot generate accurate attributed ones. Content that relies on specific, verifiable evidence is materially harder to substitute.
Score 0: Claims made without support — assertion only.
Score 4: Primary evidence — original data that exists only here.
4. Structural Clarity & Argument
Does the piece make an argument — a sequence of moves that changes what the reader thinks or can do — or does it list topics? “Five benefits of content audits, three mistakes to avoid, how to get started” is a structure. “Why the standard approach to content audits fails at the moment that matters most, and what to do instead” is an argument. Only arguments earn authority. Topic lists earn pageviews.
Score 0: No structure — wall of text or random assembly.
Score 4: The piece is the best case for this position — it anticipates counter-arguments and earns its conclusion.
5. Brand Voice Distinctiveness
Would a reader who knows your brand recognize this piece without a byline? Voice is not tone. Tone adjusts by context; voice is consistent across every context. AI-generated content defaults to the median internet voice — competent, even, neutral, unrecognizable. If your content could have been written by anyone, it was effectively written by no one.
Score 0: Interchangeable with any competitor.
Score 4: The piece could only have been written by this brand.
6. AEO / GEO Readiness
Is this piece structured to be cited by AI systems — ChatGPT, Perplexity, Claude, Google’s AI Overviews? The AI answer layer now sits above most organic results for informational queries. Content that earns positions there needs to be answer-first, explicitly structured as Q&A, attributed to a verifiable author entity, and marked up with FAQPage and Article schema. Content that doesn’t meet these criteria gets paraphrased anonymously or ignored.
Score 0: Buries the answer, no schema, no Q&A structure.
Score 4: Full schema stack, verified author entity, answer-first throughout — validated in Google’s Rich Results Test.
7. Internal Architecture & Linking
Is this piece wired into the rest of the content library, or is it an island? A page with no internal links receives no link equity from the rest of the site and sends no signals about its relationship to related content. Topical authority is built at the cluster level, not the page level.
Score 0: Orphaned — no internal links in or out.
Score 4: Fully wired — cluster attribution, pillar link, contextual spoke links, bidirectional.
8. Reader Utility & Actionability
Does a reader who finishes this piece know something they didn’t know, or can do something they couldn’t do? Thought leadership that makes no practical claims, explainers that cover the topic without enabling the reader to do anything with the information — these score low here. Utility is not a bullet list of takeaways. It’s whether the piece was designed to change what the reader can do.
Score 0: Reader finishes having gained nothing actionable.
Score 4: Piece becomes a resource readers return to — a tool, framework, or reference with increasing value on reuse.
What the scores tell you
The overall score (0–32) places content in one of four bands:
0–12: Critical. Content is producing active harm — eroding brand credibility, occupying URLs that should be doing work, signaling to search systems that the site is not a reliable source. Triage required.
13–19: Weak. Content is technically acceptable but editorially invisible. It ranks for nothing competitive, earns no citations, and gives readers no reason to trust the source.
20–25: Solid. Content is earning its place. Specific gaps exist but the foundation is sound. Targeted improvements in the weakest dimensions will produce measurable gains.
26–32: Strong. Content is operating at citation-worthy level — it gives readers something they can’t get from AI-generated summaries and gives AI systems a specific, credible source to attribute.
The overall score is useful for tracking direction over time. The individual dimension scores are where the real diagnostic value lives. The single lowest dimension is always the first priority. A piece that scores 28/32 but 0/4 on Author Authority has a fundamental problem no amount of improvement elsewhere can fix — AI systems cannot cite an anonymous piece regardless of its quality.
How to use it on a library, not just a piece
Score 10–15 pieces from across your content library — different topic clusters, different content types, different authors if you have them. Average the score for each dimension across the full sample. Any dimension averaging below 2.0 is a systemic problem, not a piece-level one. Systemic problems require policy changes and strategy shifts, not revision passes.
Common systemic patterns:
- Dimension 1 (Information Gain) below 2.0 — the content strategy has no information gain requirement. The library is producing substitutable content by design.
- Dimension 2 (Author Authority) below 2.0 — authors are not being credited specifically. This is a publishing policy problem, not a writer problem.
- Dimension 4 (Structural Clarity) below 2.0 — briefs specify topics, not arguments. The editorial direction is missing.
- Dimension 6 (AEO Readiness) below 2.0 — schema markup is not part of the publishing workflow. The strategy doesn’t account for the AI citation layer.
The 20/32 floor
Content below 20/32 has a net-negative effect on the site’s authority. It occupies URLs, consumes crawl budget, sends weak topical signals, and trains both readers and search systems to expect less from this source. There is no neutral publishing position. Publishing below the floor is worse than not publishing, because it puts your name on something that isn’t worth the name.
This is why the audit matters more than a word count check, a keyword density check, or a readability score. Those metrics evaluate content against an abstract standard. This audit evaluates content against the audience, the competitive landscape, and the citation behavior of the AI systems that now mediate discovery.
Download the scoring rubric
The full Content Authority Audit — complete 0–4 criteria for each dimension, diagnostic signals for each score level, action items for low and high scores, a printable score sheet, and the full interpretation guide — is available as a free download.
It’s a working tool, not a lead magnet. Print it, use it on the next piece that’s about to publish, and use the score to decide whether it’s ready.
→ Download the Content Authority Audit PDF
If you want to run this audit on your existing content library and translate the findings into a strategy that fixes the systemic problems, that’s what a content authority audit engagement is for. The rubric is the diagnostic. The engagement is the prescription.
This rubric pairs with two free, interactive tools in the Diagnostics Cluster — the Search Visibility Diagnostic and the 10 Content Authority Checks — at diagnostics.cliftoncreative.agency/10-dx. If you’re not sure this audit is the right starting point, The Diagnostics You’re Avoiding walks through all ten and helps you find the one that matches the problem you actually have.
This post is part of the Clifton Creative guide to E-E-A-T and author authority. For the companion piece on building the editorial infrastructure that makes content like this possible, see Your Content Team Doesn’t Have an Editorial Standard. It Has Preferences.