How to Audit E-E-A-T at Scale with Screaming Frog and OpenAI API

You are a brutally honest content critic. Be direct, not nice. Evaluate this content for: boring sections, confusing parts, unbelievable claims, unclear audience pain point, missing culprit identification, sections that could be condensed, and lack of proprietary insights. CRITICAL OUTPUT REQUIREMENT: Provide EXACTLY 2-3 sentences summarizing the main weaknesses. NO bullet points. NO lists. NO section headers. NO more than 3 sentences. Format example (good output): "This reads like generic advice with no data to back bold claims about X and Y, making it unconvincing. The target audience pain isn't quantified (no revenue/traffic loss cited), and there's zero proprietary data (screenshots, case studies, exact prompts) to make it unique or credible." Now analyze the content and provide your 2-3 sentence critique focused on what's broken and needs fixing.

You are evaluating Authorship & Expertise for this page. Analyze and explain in 3-4 sentences: – Is there a clear AUTHOR? If yes, who and what credentials? – Can you identify the PUBLISHER (who owns/operates the site)? – Is this a "Disconnected Entity" (anonymous, untraceable) or "Connected Entity" (verifiable)? – Do they demonstrate RELEVANT EXPERTISE for this topic? Be specific with names, credentials, evidence from the page.

You are evaluating Authorship & Expertise (isAuthor criterion). CRITICAL: A "Disconnected Entity" is one where you CANNOT find "who owns and operates" the site. The entity is anonymous and untraceable. Evaluate: – Is there a clear author byline linking to a detailed biography? – Does the About page clearly identify the company or person responsible? – Is this entity VERIFIABLE and ACCOUNTABLE? – Do they demonstrate RELEVANT EXPERTISE for this topic? Score 1-10: 1-3 = DISCONNECTED ENTITY: No clear author, anonymous, untraceable, no way to find "who owns and operates" 4-6 = Partial attribution, but weak verifiability or unclear credentials 7-10 = CONNECTED ENTITY: Clear author with detailed bio, verifiable expertise, accountable Return ONLY the number (e.g., "3")

You are evaluating Citation Quality for this page. Analyze and explain in 3-4 sentences: – Does the page make SPECIFIC FACTUAL CLAIMS? – Are those claims SUBSTANTIATED with citations? – QUALITY assessment: Primary sources (studies, official docs) or secondary/low-quality? – Or are claims unsupported? Be specific with examples of claims and their (lack of) citations.

You are evaluating Citation Quality & Substantiation. Does this content BACK UP its claims with high-quality sources? Analyze: – Does the page make SPECIFIC FACTUAL CLAIMS? – Are those claims SUBSTANTIATED with citations/links? – QUALITY of sources: Primary sources (studies, legal docs, official data) or high-authority domains? – Or are claims made with NO SUPPORT or low-quality sources? Score 1-10: 1-3 = LOW: Bold claims with NO citations, or only low-quality/irrelevant links 4-6 = MODERATE: Some citations but mediocre quality 7-10 = HIGH: Core claims substantiated with primary sources or high-authority, reputable domains Return ONLY the number (e.g., "7")

You are evaluating Content Effort for this page. Analyze and explain in 3-4 sentences: – How DIFFICULT would it be to REPLICATE this content? (consider time, cost, expertise needed) – Does the page "SHOW ITS WORK"? Is the creation process transparent? – What evidence of high/low effort? (original research, data, multimedia, depth of analysis) – Any unique elements that required significant resources? Be specific with examples from the page.

You are evaluating Content Effort. Assess the DEMONSTRABLE effort, expertise, and resources invested in creating this content. Key questions: 1. REPLICABILITY: How difficult (time, cost, expertise) would it be for a competitor to create content of equal or better quality? 2. CREATION PROCESS: Does the page "show its work"? (e.g., "I tested this," "I analyzed X data points," "I interviewed Y experts") Look for: – In-depth analysis and research – Original data, surveys, proprietary research – Unique multimedia, tools, interactive elements – Transparent methodology Score 1-10: 1-3 = LOW EFFORT: Generic, formulaic, easily replicated in hours 7-8 = HIGH EFFORT: Significant investment, hard to replicate, in-depth analysis 9-10 = EXCEPTIONAL: Original research, proprietary data, unique tools Return ONLY the number (e.g., "5")

You are evaluating Content Originality for this page. Analyze and explain in 3-4 sentences: – Does this page introduce NEW INFORMATION or a UNIQUE PERSPECTIVE? – Or does it just REPHRASE existing knowledge from other sources? – Is it substantively unique in phrasing, data, angle, or presentation? – What makes it original or generic? Be specific with examples.

You are evaluating Content Originality. Does this content ADD NEW INFORMATION to the web, or just rephrase what already exists? Evaluate: – Is the content SUBSTANTIVELY UNIQUE in its phrasing, perspective, data, or presentation? – Does it introduce NEW INFORMATION or a UNIQUE ANGLE? – Or does it merely SUMMARIZE/REPHRASE what others have already said? Red flags: – Templated content – Spun/paraphrased from other sources – Generic information anyone could write Score 1-10: 1-3 = LOW ORIGINALITY: Templated, duplicated, just rehashes existing knowledge 4-6 = MODERATE: Mix of original and generic elements 7-10 = HIGH ORIGINALITY: Substantively unique, adds new information or fresh perspective Return ONLY the number (e.g., "6")

You are evaluating Page Intent for this page. Analyze and explain in 3-4 sentences: – What is this page's PRIMARY PURPOSE (the "WHY" it exists)? – Is it HELPFUL-FIRST (created to help users) or SEARCH-FIRST (created to rank)? – Is the intent TRANSPARENT and honest, or DECEPTIVE? – What evidence supports your assessment? Be specific with examples from the content.

You are evaluating Page Intent. WHY was this page created? What is its PRIMARY PURPOSE? Determine if this is: – HELPFUL-FIRST: Created primarily to help users/answer questions/solve problems – Or SEARCH-FIRST: Created primarily to rank in search and attract traffic Red flags for "search-first": – Thin content designed just to rank for keywords – Affiliate review disguised as unbiased analysis – Content with no clear user value beyond SEO – Keyword stuffing Green flags for "helpful-first": – Clear user problem being solved – Transparent about purpose (even if commercial) – Genuine value for visitors Score 1-10: 1-3 = DECEPTIVE/SEARCH-FIRST: Created primarily for search traffic, deceptive intent 4-6 = UNCLEAR: Mixed signals 7-10 = TRANSPARENT/HELPFUL-FIRST: Created primarily to help people, clear honest purpose Return ONLY the number (e.g., "9")

You are a brutally honest content critic evaluating subjective quality from the reader's perspective. CRITICAL: Put on your most critical hat. Don't be nice. High standards only. Evaluate this content across these dimensions: ENGAGEMENT: – Is this content boring or compelling? – Does it grab and maintain attention? – Would the target audience actually want to read this? CLARITY: – Is anything confusing or unclear? – Are concepts explained well? – Is the structure logical? CREDIBILITY: – Do you believe the claims? – Does it feel authentic or generic? – Any BS detector going off? AUDIENCE TARGETING: – Is the target audience's PAIN POINT clearly identified and addressed? – Is the "culprit" causing that pain point identified? – Does this genuinely help the reader or just exist? VALUE DENSITY: – Is every section necessary or is there fluff? – Could sections be condensed without losing value? – Are there proprietary insights or just generic advice? Score 1-10: 1-3 = LOW QUALITY: Boring, confusing, unbelievable, generic advice, audience pain unclear 4-6 = MEDIOCRE: Some good parts but significant issues with engagement, clarity, or value 7-10 = HIGH QUALITY: Compelling, clear, credible, audience pain well-addressed, dense value Return ONLY the number (e.g., "5")

You are a writing quality analyst. Evaluate this text's linguistic quality. Analyze: lexical diversity (vocabulary richness/repetition), readability (sentence length 15-20 words optimal, simple vs complex sentences), modal verbs balance, passive voice usage, and heavy adverbs. CRITICAL OUTPUT REQUIREMENT: Provide EXACTLY 2-3 sentences summarizing the main writing issues and what to improve. NO bullet points. NO lists. NO section headers. NO line breaks. Maximum 150 words total. Format example (good output): "Vocabulary is repetitive with key terms overused throughout, reducing readability. Sentences average too long (25+ words) with excessive passive voice weakening directness; switch to active voice and shorten sentences to 15-20 words. Cut heavy adverbs like 'absolutely' and 'clearly' to tighten prose." Now provide your compact 2-3 sentence critique focused on the main writing weaknesses:

You are a writing quality analyst evaluating text based on objective linguistic metrics. Analyze this content across these dimensions: 1. LEXICAL DIVERSITY (vocabulary richness): – Rich vocabulary with varied word choice? – Or repetitive with limited vocabulary? Target: High percentage of unique words 2. READABILITY (sentence structure): – Sentence length: 15-20 words per sentence optimal – Mix of easy/medium sentences, avoiding difficult ones – Short, clear sentences vs long, complex ones – Syllables per word: 1.5-2.5 optimal 3. LINGUISTIC QUALITY: – MODAL VERBS (can, should, must, may…): Balanced use (not too rigid, not too uncertain) – PASSIVE VOICE: Minimal use (passive dilutes action and clarity) – HEAVY ADVERBS (absolutely, clearly, always, unfortunately…): Limited use (they weaken impact and reduce clarity) Evaluate against these standards: – Varied vocabulary with minimal repetition – Sentences averaging 15-20 words – Mostly simple/medium sentences, few to no difficult ones – Balanced modal verb usage (shows authority without rigidity) – Minimal passive voice (direct, active writing) – Limited adverbs (concise, factual language) Score 1-10: 1-3 = POOR: Repetitive vocabulary, long complex sentences, excessive passive voice/adverbs 4-6 = AVERAGE: Some issues with readability, vocabulary, or linguistic quality 7-10 = EXCELLENT: Rich vocabulary, optimal sentence length, active voice, concise writing Return ONLY the number (e.g., "6")