10 AI Prompts to Generate a Complete SEO Audit Report for Any Website

A complete SEO audit evaluates technical infrastructure, on-page content relevance, backlink equity, and search intent alignment to uncover why a website isn’t ranking at its full potential. While traditional audits often require 20 to 40 hours of manual data collation across spreadsheets, modern large language models can analyze your raw crawl exports, Google Search Console metrics, and page HTML in minutes when guided by structured prompts.

This guide provides 10 battle-tested AI prompts engineered to generate a client-ready, comprehensive SEO audit report. Each prompt processes real diagnostic inputs, prevents hallucinated metrics, and outputs structured recommendations prioritized by technical impact and ease of execution.

How to Structure AI Prompts for Website SEO Audits

Large language models excel at synthesizing complex analytical data, identifying anomalous patterns, and generating clear technical recommendations. However, they lack direct real-time access to search engine ranking databases unless supplied with raw diagnostic exports. To get actionable, production-grade audit results rather than generic marketing advice, adhere to these operational principles:

• Supply the raw diagnostic data: Feed the model extracted exports from tools like Google Search Console, Screaming Frog, PageSpeed Insights, Ahrefs, or Semrush. Never ask an AI model to guess search volume or crawl stats from memory.
• Enforce strict factual boundaries: Include a explicit negative constraint forbidding the model from inventing traffic figures, domain ratings, or ranking positions.
• Define structured output formats: Request data in Markdown tables, prioritized matrices, or JSON-LD blocks to simplify copying findings directly into audit slide decks or client deliverables.
• Isolate diagnostic layers: Run dedicated prompts for technical crawlability, on-page architecture, performance, and internal linking rather than asking for a monolithic audit in a single request.

The Master SEO Audit Prompt Framework

When customizing audits for specific site architectures (such as e-commerce platforms, multi-lingual SaaS directories, or local service sites), use this master prompt architecture to maintain precision:

Act as a Principal Technical SEO Consultant and Web Auditor.
Task: Conduct a rigorous audit of [SPECIFIC AUDIT FOCUS AREA] for [WEBSITE URL / NICHE].

Context and Input Data:
[PASTE CRAWL EXPORTS, HTML SNIPPETS, GSC CSV DATA, OR URL METRICS HERE]

Audit Requirements:
1. Ground every finding strictly in the provided data. Do not extrapolate unverified metrics or fabricate ranking statistics.
2. Group identified issues into three severity tiers: Critical (Blocking indexing or causing severe rank drops), Moderate (Suboptimal architecture or lost efficiency), and Low (Polishing and minor enhancement).
3. Provide concrete remediation instructions for each finding, including code snippets or rewrite templates where applicable.
4. Conclude with an Impact vs. Effort score (High/Medium/Low) for development prioritization.

Output Format: Present the analysis in a structured Markdown report featuring an Executive Overview table followed by detailed diagnostic breakdowns.

10 Essential Prompts for a Complete SEO Audit

1. Technical Crawlability and Indexation Audit

Technical crawlability determines whether search engine spiders can efficiently discover, parse, and index your priority URLs without exhausting crawl budget.

Act as an enterprise Technical SEO Architect. Perform a crawlability, indexation, and status code audit based on the crawl data provided below.

Target Website / URL: [INSERT URL OR DOMAIN]
Crawl Data & Directives:
[PASTE ROBOTS.TXT CONTENT, XML SITEMAP URLS, AND STATUS CODE SUMMARY OR SCREAMING FROG EXPORT]

Audit Tasks:
1. Robots.txt Analysis: Identify any disallow rules blocking critical assets (CSS, JS, media) or indexing inadvertently blocked URLs. Check sitemap declarations.
2. HTTP Status Code Triage: Audit URLs returning 3xx redirect chains, 4xx broken links, and 5xx server errors. Quantify the chain length and specify final destination URLs.
3. Canonical Tag Configuration: Evaluate canonical implementation. Identify self-referential canonical mismatches, missing tags, cross-domain conflicts, or conflicting HTTP header canonicals.
4. Meta Robots & X-Robots Directives: Detect conflicting indexing signals (e.g., page allowed in robots.txt but tagged with "noindex, follow").
5. Remediation Table: Deliver a Markdown table with columns: [Issue Identified, Affected URL Pattern, Severity (Critical/Warning/Notice), Recommended Fix, Developer Action Item].

Real-world example: Auditing an e-commerce catalog migration that resulted in unexpected traffic drops due to trailing slash redirect chains and misplaced noindex directives in staging headers.

How to use it: Export your Screaming Frog or Sitebulb crawl summary (focusing on response codes, directives, and canonicals), copy the top 50–100 non-200 or irregular URLs, and paste them into the input section.

2. On-Page Architecture and Metadata Evaluation

On-page architecture ensures search engines understand document topic relevance while maximizing click-through rates on search engine result pages (SERPs).

Act as a Senior On-Page SEO Strategist. Audit the on-page structure, metadata, and heading hierarchy for the provided page set.

Website Context / Core Topic: [INSERT NICHE OR PRIMARY KEYWORD]
Page Data:
[PASTE LIST OF TARGET URLS WITH CURRENT TITLE TAGS, META DESCRIPTIONS, H1S, AND HEADINGS OUTLINE]

Audit Tasks:
1. Title Tag Optimization: Flag titles exceeding 60 characters (or ~580px width), truncated titles, missing primary brand modifiers, or keyword stuffing. Provide rewritten alternatives adhering to high-CTR best practices.
2. Meta Description Review: Identify missing, duplicate, or truncated descriptions (over 155 characters). Write revised descriptions featuring active voice, primary entities, and clear intent triggers.
3. Heading Architecture (H1-H6): Audit heading nesting logic. Flag multiple H1 tags, skipped levels (e.g., H2 straight to H4), and vague headings that lack semantic topical descriptors.
4. Slug & URL Structure: Flag over-parameterized URLs, excessive folder depths (>3 subfolders), and non-descriptive slugs.
5. Deliverable: Present the audit in a comparative before-and-after table: [URL, Current Title, Optimized Title, Current Meta Description, Optimized Meta Description, Heading Fixes].

Real-world example: Reviewing top revenue-generating SaaS product landing pages where generic titles like “Home – Platform” fail to capture non-branded commercial search queries.

How to use it: Pull your top 20 landing pages from Google Analytics or GSC, export the title and heading strings, and feed them into the prompt to generate polished replacements.

3. Core Web Vitals and Page Speed Triage

Page speed and visual stability directly influence bounce rates, mobile user experience, and search engine ranking eligibility under Google’s page experience system.

Act as a Web Performance Engineer and Core Web Vitals Specialist. Analyze the provided Lighthouse / PageSpeed Insights diagnostic data and formulate an actionable engineering punch list.

Page URL & Device Type: [INSERT URL AND MOBILE/DESKTOP]
Diagnostic Metrics & Lighthouse Output:
[PASTE PAGESPEED AUDIT METRICS: LCP, INP, CLS, TTFB, TBT, AND UNUSED JAVASCRIPT/CSS OPPORTUNITIES]

Audit Tasks:
1. Largest Contentful Paint (LCP): Identify the LCP candidate element (hero image, video poster, H1 block). Prescribe remedies: priority resource hints (fetchpriority="high"), CDN caching, next-gen image conversion (WebP/AVIF), or server response time reductions.
2. Interaction to Next Paint (INP): Pinpoint main-thread blocking JavaScript, excessive script execution, long event listeners, or third-party tag manager bloat.
3. Cumulative Layout Shift (CLS): Locate layout shifts caused by unsized images/embeds, dynamic ad injections, or late-loading web fonts (FOUT/FOIT).
4. Asset Optimization: Detail specific third-party scripts to defer, async, or remove entirely.
5. Engineering Action Plan: Create a prioritized task list formatted for a front-end sprint board: [Metric Affected, Bottleneck Component, Technical Solution, Expected CWV Impact].

Real-world example: Addressing mobile CLS penalties triggered by dynamically rendered consent banners and slow hero images on high-traffic publisher sites.

How to use it: Run Google PageSpeed Insights on mobile, copy the JSON or text summary of the audit diagnostics and opportunities sections, and paste it directly into the prompt.

4. Content Depth and Topical Authority Analysis

Topical authority requires comprehensive semantic coverage of a subject, satisfying user queries with genuine information gain while demonstrating verifiable E-E-A-T.

Act as an Lead SEO Content Auditor and Semantic Search Specialist. Conduct an exhaustive topical depth, entity coverage, and E-E-A-T audit for the article provided below.

Target Keyword / Core Query: [INSERT PRIMARY TOPIC]
Target Page Content:
[PASTE FULL BODY TEXT OF TARGET ARTICLE]
Top Ranking Competitor Outlines / Summaries:
[PASTE OUTLINES OR EXTRACTS FROM TOP 3 SERP COMPETITORS]

Audit Tasks:
1. Information Gain & Unique Value: Assess whether this page offers novel data, original analysis, expert commentary, or merely parrots competitor content.
2. Semantic Entity Coverage: Identify critical missing entities, subtopics, terminology, and related concepts present in top competitors that are absent from this content.
3. Thin or Redundant Content Flagging: Highlight sections containing generic fluff, repetitive phrasing, or non-actionable filler that dilutes topical relevance.
4. E-E-A-T Assessment: Evaluate demonstrative expertise, author credentials, authoritative citations, primary source references, and objective verification points.
5. Content Upgrade Blueprint: Provide an expansion outline featuring new subheadings, recommended illustrative data tables, and specific paragraphs to rewrite.

Real-world example: Modernizing a ranking blog post that has steadily lost positions over 12 months as newer competitors published more granular tutorials with interactive tooling.

How to use it: Copy your article text alongside competitive headings from the top 3 ranking URLs in your target market to detect semantic content gaps.

5. Search Intent and SERP Feature Optimization

Mismatching search intent causes immediate bounce backs. Aligning page content with SERP features maximizes organic visibility through rich snippets, PAA blocks, and AI answers.

Act as a Search Intent Analyst and SERP Feature Strategist. Evaluate intent alignment and rich snippet capture opportunities for the following keyword and URL mapping.

Target Keyword: [INSERT KEYWORD]
Current SERP Features Observed: [E.G., FEATURED SNIPPET, PEOPLE ALSO ASK, VIDEO PACK, AI OVERVIEW, SHOPPING GRID]
Target Page Content / Summary:
[PASTE PAGE CONTENT OR OUTLINE]

Audit Tasks:
1. Intent Classification: Classify search intent across stages: Informational (know/know simple), Commercial Investigation, Transactional (do), or Navigational (website). Does the page structure match the SERP expectations?
2. Featured Snippet Engineering: Identify the primary definition or step query. Draft a concise 42–58 word paragraph snippet or numbered step list designed to trigger the position-zero snippet.
3. People Also Ask (PAA) Integration: Extract 4–6 high-value questions from current search behavior and draft clear, authoritative 2-3 sentence answers ready for an on-page FAQ module.
4. Intent Disconnect Flags: Highlight any on-page friction where transactional CTAs interrupt an informational researcher prematurely.

Real-world example: Repurposing a product category page that was failing to rank because Google’s SERP had shifted from direct purchasing intent to educational comparison guides.

How to use it: Search your target keyword in an incognito window, observe what Google ranks (calculators, guides, product grids, or videos), and provide those SERP observations in the prompt.

6. Internal Linking and Click-Depth Siloing

A deliberate internal linking architecture passes link equity efficiently throughout your domain and signals topical relationships between parent pillar pages and supporting cluster articles.

Act as an Information Architect and SEO Link Equity Analyst. Analyze the internal linking structure and URL depth for the page dataset provided below.

Site Structure / Pillar Category: [INSERT TOPIC SILO OR DIRECTORY]
Internal Link Export & URL Data:
[PASTE LIST OF PAGES WITH INCOMING INTERNAL LINK COUNTS, OUTGOING LINKS, CLICK DEPTH, AND ANCHOR TEXT EXAMPLES]

Audit Tasks:
1. Orphan & Deeply Nested Pages: Identify priority pages buried deeper than 3 clicks from the homepage or receiving fewer than 3 internal incoming links.
2. Anchor Text Quality: Flag ambiguous anchors (e.g., "click here", "read more", "source", raw URLs). Recommend descriptive, semantically rich anchor text variations incorporating target entities.
3. Link Equity Distribution (PageRank Flow): Identify authority hubs (pages with high external backlinks) that fail to pass link equity downward to high-intent conversion pages.
4. Topic Silo Integrity: Detect stray internal links bridging unrelated silos that dilute contextual topic boundaries.
5. Internal Linking Map: Build a prioritized linking matrix: [Source URL, Destination URL, Recommended Anchor Text Context, Rationale].

Real-world example: Uncovering why a collection of high-value industry whitepapers failed to rank because they were isolated 5 clicks deep without contextual links from the main blog.

How to use it: Run an internal link report in your crawler, sort URLs by inlink count ascending to find isolated content, and provide the data to generate precise anchor suggestions.

7. Backlink Profile Health and Anchor Distribution

A clean backlink profile establishes off-page domain trust, while unnatural anchor patterns or sudden spikes in low-quality directories can trigger algorithmic filtering.

Act as an Off-Page SEO Auditor and Link Profile Risk Analyst. Analyze the backlink profile metrics and anchor text distribution provided below.

Target Domain: [INSERT DOMAIN]
Backlink & Referring Domain Metrics:
[PASTE AHREFS, SEMRUSH, OR MOZ BACKLINK SUMMARY: TOTAL BACKLINKS, REFERRING DOMAINS, TOP ANCHOR TEXT RATIOS, TLD BREAKDOWN]

Audit Tasks:
1. Anchor Text Balance: Calculate the distribution percentage of Brand anchors, Exact Match keywords, Partial Match, Generic ("website", "visit"), and Naked URLs. Flag any exact-match ratios exceeding safe industry benchmarks (>15–20%).
2. Referring Domain Velocity & Quality: Screen for link networks, spam TLD spikes (.xyz, .top, .buzz), syndicated scraper links, or automated comment link schemes.
3. Link Equity Attribution: Identify the top 5 most linked-to pages on the site and verify whether that equity is actively funneled into commercial targets via internal links.
4. Disavow / Risk Recommendation: Provide objective criteria on whether suspicious links warrant a Google Disavow file submission or if algorithmic ignoring is sufficient.

Real-world example: Diagnosing an unexpected stagnation in rankings caused by an aggressive third-party negative SEO link blast using repetitive commercial exact-match anchors.

How to use it: Download the Anchor Text overview CSV and Referring Domains table from your backlink intelligence software and paste the summarized metrics into the prompt.

8. Schema Markup and Structured Data Generation

Structured data communicates explicit page context directly to search engine crawlers, enabling rich snippets, knowledge graph entities, and improved natural language understanding.

Act as a Structured Data Engineer and Semantic Schema Developer. Audit and generate valid JSON-LD markup for the page specified below.

Page Type: [E.G., E-COMMERCE PRODUCT, HOW-TO ARTICLE, SAAS SOFTWARE APPLICATION, LOCAL BUSINESS SERVICE]
Page Details & Business Entities:
[PASTE RELEVANT DETAILS: BUSINESS NAME, AUTHOR BIO, PRICING, PRODUCT SPECS, FAQ QUESTIONS/ANSWERS, RATINGS]

Audit Tasks:
1. Schema Gap Identification: Determine which schema types are missing based on Google's Search Central documentation for this specific page type.
2. Schema Generation: Write complete, syntactically valid JSON-LD code incorporating all recommended nested entities (e.g., Organization, Author with SameAs social references, PrimaryImageOfPage, AggregateRating).
3. Entity Disambiguation: Add Wikipedia/Wikidata @id entity references where appropriate to establish unambiguous topical authority.
4. Rich Result Compatibility: Ensure compliance with Google Rich Results requirements, preventing deprecation warnings or missing field flags.

Real-world example: Implementing advanced SoftwareApplication and FAQPage schema on a B2B SaaS landing page to earn interactive review stars and expand SERP footprint.

How to use it: Provide your raw page content, company info, and customer review aggregates. Run the output code directly through Schema.org validator or Google’s Rich Results Test tool.

9. AI Search Readiness and AEO Optimization

Answer Engine Optimization (AEO) prepares your site to be cited and recommended by generative AI engines including Google AI Overviews, Perplexity, and ChatGPT Search.

Act as an AI Search Optimization (AEO) Specialist and LLM Citability Auditor. Analyze the provided page content to assess how readily generative search engines can extract, cite, and reference this material.

Target Topic / Brand: [INSERT TOPIC OR BRAND NAME]
Page Content:
[PASTE PAGE CONTENT OR PRIMARY PRODUCT/SERVICE OVERVIEW]

Audit Tasks:
1. Information Extraction Clarity: Does the page present definitive answers in structured, clear semantic blocks (tables, definitions, ordered steps) that an LLM can parse into factual embeddings?
2. Entity Grounding: Check whether company claims, statistics, and industry distinctions are supported by named primary sources and clear attribution anchors.
3. Direct Question Answering: Check if the first 150 words of each major section provide an answer directly before expanding into contextual nuance.
4. Semantic Ambiguity Scan: Flag passive phrasing, colloquialisms, or ambiguous pronouns that confuse automated text scrapers and summarization agents.
5. Citability Optimization: Rewrite the core thesis of the page into a concise, cite-worthy quote block engineered for attribution by AI search engines.

Real-world example: Adapting high-value research articles so that Perplexity and Google Gemini cite your company as the authoritative primary source in generative summaries.

How to use it: Feed in your cornerstone informational or commercial guides, examine how the AI identifies factual extraction points, and update introductory paragraphs accordingly.

10. Executive Summary and Prioritized Roadmap

An audit is useless if it sits in an unread 80-page document. This prompt compiles technical findings into an executive-ready action matrix with clear sprint timelines.

Act as a Chief SEO Officer and Technical Project Manager. Synthesize the collected audit findings from all audit areas into an Executive Summary and Prioritized 30-60-90 Day Action Roadmap.

Audit Findings Input:
[PASTE SUMMARY NOTES OR ISSUES DISCOVERED FROM TECHNICAL, ON-PAGE, CWV, CONTENT, AND BACKLINK AUDITS]

Roadmap Deliverables:
1. Executive Summary: Write a 3-paragraph executive overview summarizing the current organic health, core revenue blockers, and projected growth opportunities in plain business language.
2. Impact vs. Effort Prioritization Matrix: Group all identified action items into a 2x2 matrix:
   - Quick Wins: High Impact, Low Effort (Execute immediately)
   - Major Projects: High Impact, High Effort (Plan into engineering roadmap)
   - Fill-Ins: Low Impact, Low Effort (Assign during downtime)
   - Money Pits: Low Impact, High Effort (Deprioritize or discard)
3. 30-60-90 Day Phased Sprint Schedule:
   - Days 1–30 (Foundation): Critical crawlability blockers, security, 4xx/5xx errors, high-priority indexing fixes.
   - Days 31–60 (Optimization): On-page title/meta rewrites, Core Web Vitals remediation, internal linking silos.
   - Days 61–90 (Authority & Expansion): Content gap filling, schema deployment, advanced AEO adaptations, E-E-A-T enhancements.
4. Resource Allocation: Specify the required team member for each task (Developer, SEO Specialist, Content Writer, Designer).

Real-world example: Translating a chaotic list of 150 disparate crawler warnings into an actionable, cross-functional roadmap that engineering leadership will actually approve.

How to use it: Take the action items generated from Prompts 1 through 9, paste them into this prompt, and produce the concluding deliverable for your client presentation.

Comparing Top AI Models for SEO Auditing

Not all foundation models handle audit workloads identically. Selecting the right model for each phase ensures faster processing and fewer hallucinations:

• Claude 3.5 Sonnet: The preferred model for large CSV file analysis, code generation (JSON-LD, robots.txt, htaccess), and nuanced content quality evaluations. Its strong reasoning capabilities minimize hallucinated patterns in data tables.
• ChatGPT Plus (GPT-4o): Highly effective for on-page copywriting rewrites (titles, meta descriptions, FAQ modules), marketing angle ideation, and rapid SERP intent classification.
• Google Gemini 1.5 Pro: Exceptional for massive input contexts (up to 2 million tokens). Ideal for ingesting entire site crawl exports, thousands of log files, or complete competitive landing pages in a single prompt session.

Critical Pitfalls to Avoid in AI-Assisted SEO Audits

While AI accelerates analysis, over-reliance without human verification introduces distinct risks:

• Hallucinating metrics and rankings: Large language models cannot access private Google Search Console dashboards or real-time keyword rankings unless you provide the data. Never ask AI “What does my site rank for?”
• Blindly executing generic recommendations: Automated outputs may suggest disavowing links or rewriting valid canonicals because they lack full company context or architectural awareness. Every technical recommendation must be verified by an experienced practitioner.
• Truncated data inputs: Pasting a 5-row sample from a 10,000-page crawl will yield skewed conclusions. Always provide statistically significant data samples or aggregated summaries.
• Ignoring brand voice: AI-generated title tags and meta descriptions can default to formulaic clichés. Edit every snippet to preserve brand differentiation.

Step-by-Step Workflow to Assemble Your Final Audit Deliverable

To turn these prompts into a professional SEO audit workflow, follow this 4-step sequence:

1. Step 1 – Data Gathering: Run a complete crawl using your preferred crawler (Screaming Frog, Sitebulb). Export search performance data from Google Search Console, Google PageSpeed Insights, and backlink aggregators.
2. Step 2 – Sequential Diagnostic Runs: Process Prompts 1 through 9 individually. Keep a dedicated document or spreadsheet where you record the tables and code snippets generated by each prompt.
3. Step 3 – Verification and Filtering: Review every finding against the live site. Verify that status codes are accurate, test generated JSON-LD schema using validator tools, and confirm that flagged redirect chains exist in production.
4. Step 4 – Executive Synthesis: Feed all confirmed action items into Prompt 10 to build your 30-60-90 day roadmap and executive summary. Package the final output into a presentation deck or PDF deliverable.

Frequently Asked Questions

Can AI tools crawl a live website directly for an SEO audit?

Most standard AI chat interfaces cannot actively crawl complex, JavaScript-rendered websites in real time. While some models feature basic web browsing, they lack the specialized capabilities of a dedicated crawler like Screaming Frog. The most reliable approach is to crawl the site with specialized software and supply the raw diagnostic exports to the AI for analysis.

Which AI model handles large crawl data files best?

Google Gemini 1.5 Pro and Claude 3.5 Sonnet handle large datasets best due to their massive context windows (1M+ tokens and 200k tokens respectively). They can ingest multi-megabyte CSV exports of crawl logs, redirect maps, and search queries without truncating or losing context.

Does an AI audit replace specialized tools like Screaming Frog or Ahrefs?

No. AI complements those tools rather than replacing them. Crawler tools collect raw technical data and backlink metrics; AI models interpret that complex data, detect subtle patterns, draft remediation code, and format client-friendly audit reports.

How do you prevent AI from hallucinating keyword metrics?

Enforce explicit constraints in your prompts stating: “Use only the provided CSV data as the source of truth. Do not estimate search volume, keyword difficulty, or backlink counts.” If certain metrics are not in the uploaded file, instruct the AI to state “Data not supplied” rather than estimating.

How do you adapt these prompts into automated workflows?

You can integrate these prompt structures into custom Python scripts, Make.com scenarios, or n8n automation pipelines using OpenAI or Anthropic APIs. Simply pipe raw crawl CSV outputs into the prompt template via API calls to generate automated audit drafts at scale.

Complete SEO Audit Quality Assurance Checklist

• [ ] Crawl data and server response codes are sourced from empirical diagnostic tools.
• [ ] Robots.txt directives and XML sitemap URLs have been validated manually.
• [ ] Canonical tags and meta robots indexing directives do not present conflicting signals.
• [ ] Title tags and meta descriptions have been reviewed for character length and brand tone.
• [ ] Heading hierarchy follows a single logical H1 down through descriptive H2 and H3 subheadings.
• [ ] Core Web Vitals remediation steps have been verified against mobile PageSpeed tests.
• [ ] Internal linking suggestions use natural, entity-relevant anchor text.
• [ ] All generated JSON-LD structured data passes the Schema.org and Google Rich Results validators.
• [ ] Content improvements focus on information gain and verifiable primary evidence.
• [ ] The final action roadmap is prioritized into an executable 30-60-90 day schedule based on impact versus effort.