How it works

How does CitationOne work?

CitationOne analyzes your article and answers one question: will AI models - ChatGPT, Perplexity and Google AI Overview - cite your content instead of the competition. It breaks the article into 10 quality dimensions, benchmarks against top 10 SERP and generates prioritized recommendations with ready-to-paste fixes.

Full report in ~3 minutes
10 dimensions + E-E-A-T
Top 10 SERP benchmark
Before/After with ready fixes

3 steps to your report

STEP 1

Paste article URL

Enter the page address and keyword. The tool automatically fetches the content - no manual copy-paste. Choose: Full mode (with SERP benchmark, ~3 min) or Content-only (faster, ~30-60 sec, no competitor data).

STEP 2

AI analyzes in parallel

10 parallel AI calls simultaneously: 9 quality dimensions + E-E-A-T, plus algorithmic editorial effort score. In Full mode the tool crawls top 10 SERP, extracts entities from each article and builds a competition benchmark.

STEP 3

Get your report

Report with prioritized Before/After recommendations - exact quote from your article and a ready-made fix with estimated score impact. Export to PDF for your client or Markdown for further work - one click.

Main scores

One score instead of guesswork

CQS (0-100) is an aggregated score from 10 dimensions - shows how far your content is from the standard AI likes to cite. AI Citability (0-10) measures the probability that a language model will choose your page over the competition.

  • -Each of 10 dimensions with separate 0-10 score
  • -Status at a glance: OK / WARNING / CRITICAL
  • -Your score compared with top 10 SERP

CQS

84/100
OK

Citability

6.4/10
WARNING

CQS components

Intent Alignment×0.25
8.4
E-E-A-T×0.20
7.1
Cost of Retrieval×0.20
9
Info Density×0.15
6.8
TF-IDF×0.10
7.5
Semantic Roles×0.10
5.2

10 dimensions + E-E-A-T

What exactly does the tool analyze?

Each dimension is a separate AI call with a dedicated prompt - separate 0-10 score, separate problems and separate recommendations.

CSI-A

Intent Alignment

Checks if the article answers exactly the question the user asked - not a similar one, but exactly that one.

D1

Information Density

Measures how many facts are in the article. Generalities and empty sentences lower the score - concrete data and numbers raise it.

D2

Knowledge Graph

AI sees articles as a network of facts: Entity - Attribute - Value. The more complete the network, the higher the chance of citation.

D3

BLUF

AI models favor articles that give the answer at the beginning of each section. Not at the end, not after an intro - right at the start.

D4

Chunk Optimization

AI systems split articles into chunks before analysis. Each chunk should make sense without reading the whole article - be autonomous.

D5

Cost of Retrieval

The harder it is for AI to find information in the text, the less likely it is to cite it. Headings, lists and tables reduce this cost.

D6

TF-IDF

Checks if the article uses domain terminology. Lack of specialized terms signals AI: "this author does not know the topic deeply".

D7

Semantic Roles

AI absorbs knowledge better when the article topic is an active subject in sentences - not a passive object described by others.

D8

AI Overview Coverage

Each query is effectively several sub-queries at once. We check how many your article covers - because AI uses exactly those sub-queries for synthesis.

D9

Editorial Effort

Measures visible editorial effort: article length, images, video, tables and FAQ schema. AI models prefer polished content - not quick drafts.

E-E-A-T

E-E-A-T

Google and AI models trust content backed by a real expert with experience. The report measures specific trust signals.

SERP Benchmark

See exactly where you fall behind

The tool automatically fetches and scores 10 articles currently ranking for your keyword. You see each competitor's CQS and the exact gap to close - no manual data collection.

  • -CQS per article - tabular comparison
  • -Identify SERP leaders and weak spots
  • -Competitor content formats: tables, FAQ, lists

CQS - top 10 SERP comparison

Your article
72
Competitor #1
88
Competitor #2
81
Competitor #3
76
Competitor #4
65
Competitor #5
58

↑ 16 pts to leader position - report shows exactly what to change

Before/After Recommendations

Not “what to fix” - but “how to fix it”

Each recommendation contains an exact quote from the article and a ready version after the fix. Zero interpretation: you paste, you don't rewrite. Each with priority and estimated CQS impact.

Before

AI optimization is important because language models select content based on their own criteria.

After

CitationOne measures 10 AI citation dimensions - with estimated impact on CQS and a specific fix to paste.

CRITICALHIGHMEDIUMpriority + estimated CQS impact per recommendation
G
AI Overview - synthesis decomposition
3/5covered
what is an AI Search auditcovered
how does Google AI Overview workcovered
what metrics does AI audit measuregap
SEO vs AI Search differencegap
how to improve citability scorecovered

2 gaps - report tells you what to write to cover missing sub-queries

AI Overview Coverage

Google AI Overview is a synthesis of many sub-queries

AI Overview doesn't cite one article - it synthesizes answers to a dozen related sub-queries at once. The tool decomposes this synthesis and shows which sub-queries your content covers, and which gaps cause AI to skip you.

  • -AI Overview synthesis decomposed into sub-queries
  • -Coverage map: which sub-queries you handle
  • -Recommendations to fill gaps with specific content

Knowledge Graph (EAV)

AI sees entities - not just keywords

Language models build knowledge representation from facts - entities and their attributes. The report maps these relationships and shows which facts set you apart from competition (Unique), which are must-have (Root) and which you're missing.

  • -Full entity table with Unique/Root/Rare classification
  • -Coverage map: covered / gap / unique
  • -Interactive knowledge graph in the app
UniqueRootRare
EntityAttributeValueType
AI Searchalgorithmlanguage modelsRoot
BLUFdefinitionBottom Line Up FrontUnique
ChatGPTtypelanguage modelRoot
CQSrange0-100Unique
Google AIOsourceSERP groundingRare

AI Search Audit Report

example-article.com · keyword: AI audit

PDFMD
Content Quality Score (0-100) with dimension breakdown
AI Citability Score (0-10)
Radar chart of 10 dimensions
Top 10 SERP benchmark - comparison table
Before/After recommendations with priorities
Knowledge graph and EAV entity table
AI Overview Coverage and sub-query analysis
Executive summary ready to send

Report export

Report ready to send to your client

PDF designed for presentation - all report sections in readable format without additional editing. Markdown for project documentation or automated processing.

PDF

Ready to send

Markdown

For documentation

Quick Wins

Instant fixes - right after the audit

Up to 7 ready-made fixes generated algorithmically (no extra AI calls). Each with a source badge and a link to the dimension that detected it. You know what to fix before reading the full report.

EffortAdd a comparison table - competitors use it in 7/10 articles
TitleShorten title to 60 characters (currently 78)
EEATAdd last updated date and author bio
Fan-OutCover sub-query "how to measure results"
SchemaAdd FAQPage schema - you have a FAQ section
TF-IDFAdd missing terms: "conversion", "retention"

Schema.org

Structured data audit

Algorithmic analysis of schema.org JSON-LD (0 AI calls). Detects ~14 schema types, checks field completeness and flags missing required schemas with priority.

Article, FAQPage, Product, HowTo, Review, BreadcrumbList, WebPage, Organization, Person, AggregateRating - each with required and recommended fields. Status: present / incomplete / missing. Google Rich Result eligibility.

Schema.org Audit
ArticleIncompletedateModified, author.url
FAQPageMissingentire schema
BreadcrumbListOK
WebPageMissingentire schema
OrganizationOK
IG Score68/ 100
Unique claims72%
Unique EAV61%
Unique TF-IDF terms55%
Format advantage80%

Information Gain

How much unique value does your content bring?

Information Gain measures content uniqueness vs SERP competition. Does not affect CQS - purely informational metric. Extracts 10-20 factual claims and compares with competitor content (token overlap).

IG Score (0-100) consists of 4 components: claim uniqueness (40%), unique EAV triples (30%), unique TF-IDF terms (20%) and format advantage (10%). Each claim marked with a unique or common badge.

Check if your content
is ready for AI Search.

First audit takes ~3 minutes. Report ready to send to your client right after completion.

Run audit

No subscription · VAT invoice · Bank transfer