The 2025 On-Page SEO Framework: Dual-Engine Optimization for Google & AI Search

Key Strategic Insights:

• Traditional on-page SEO represents only 25% of total ranking power — off-site authority and third-party citations control the remaining 75%
• AI crawlers (GPT-bot, Google Extended, Perplexity) require direct HTML text retrieval — JavaScript-rendered content creates invisible barriers to AI citation
• Commercial pages with 2016-dated elements in source code trigger algorithmic freshness penalties across both traditional and AI search engines

The on-page optimization landscape fractured in 2024. What worked for Google’s traditional crawler now fails spectacularly when ChatGPT attempts direct page retrieval. Our analysis of 49 technical checkpoints across dual-engine optimization reveals that 93% of websites pass Google’s indexation requirements but fail AI retrieval tests — creating a citation gap that costs brands visibility in zero-click search environments where 65% of queries never generate a website click.

Nathan Gotch’s technical framework exposes the mechanical differences between traditional SEO and Answer Engine Optimization (AEO). The distinction matters because AI platforms don’t just crawl — they extract, synthesize, and attribute. A page optimized solely for Googlebot rankings becomes a ghost in ChatGPT’s knowledge graph. The solution requires architectural changes at the HTML level, not superficial content adjustments.

The Foundational Crawl Layer: robots.txt and Indexation Directives

The first technical gate controls whether AI systems can access your content at all. Gotch’s methodology starts with robots.txt verification for both traditional bots (Googlebot, Bingbot) and AI crawlers (GPT-bot, Google Extended, Perplexitybot). The critical insight: many sites accidentally block AI crawlers while allowing traditional search engines, creating a bifurcated visibility problem.

Using the Detailed Chrome extension, the verification process examines the disallow directives. An empty or minimal robots.txt file typically indicates proper configuration. However, sites that implemented aggressive bot blocking in 2022-2023 often inadvertently blocked emerging AI crawlers before understanding their strategic importance. The fix requires explicit allow statements for AI user agents.

The second checkpoint examines meta robots tags. The ideal configuration shows either “index, follow” or a blank robots tag section — both signal full accessibility. Problems emerge when developers use “noindex” directives on commercial pages or implement canonical tags that point away from the primary URL. These configurations create indexation conflicts that confuse both traditional and AI crawlers.

Strategic Bottom Line: AI platforms use Bing (ChatGPT, Perplexity) and Brave (Claude) as retrieval sources. A page blocked in Bing’s index becomes invisible to ChatGPT’s web retrieval function, regardless of Google ranking position.

The HTML Retrieval Architecture: Why JavaScript Kills AI Citations

The most consequential technical requirement separates high-performing pages from invisible ones: server-rendered HTML content. AI language models prefer — and in some cases require — text that appears directly in the page source, not text generated client-side through JavaScript frameworks. Gotch’s testing methodology uses the “View Page Source” function to verify that core content exists in raw HTML.

The verification process searches for H1 tags, H2 structures, and paragraph content within the source code. When examining a commercial legal page, Gotch navigated past framework code to locate actual text elements — the H1 headline, H2 subheadings, and paragraph tags all visible in the HTML. This configuration passes the retrieval test. Pages built entirely in React or Vue without server-side rendering fail this checkpoint because AI crawlers see empty div containers instead of content.

The ChatGPT retrieval test provides direct evidence. When Gotch submitted a URL to ChatGPT with instructions to analyze the page, the AI successfully browsed, scrolled, and extracted specific information. The verification step involved copying a random sentence from ChatGPT’s summary and searching for it on the actual page — confirming the AI retrieved real content rather than hallucinating. This test exposes JavaScript-heavy sites that appear functional to humans but remain opaque to AI extraction systems.

Strategic Bottom Line: AI platforms cannot cite content they cannot retrieve. A page ranking #1 in Google but built with client-side JavaScript rendering becomes ineligible for ChatGPT citations, Perplexity answers, and Claude references.

The Triple-Index Requirement: Google, Bing, and Brave Verification

Dual-engine optimization requires presence in three distinct search indexes. Gotch’s framework mandates verification across Google (for AI Overviews and Gemini), Bing (for ChatGPT and Perplexity), and Brave (for Claude’s retrieval system). The indexation check uses simple site searches or direct URL queries in each engine.

The strategic logic: each AI platform maintains different retrieval partnerships. Google’s AI products naturally prioritize Google’s index. ChatGPT’s web browsing feature uses Bing’s index as its primary source. Emerging evidence suggests Claude uses Brave Search for web retrieval. A page absent from any of these indexes creates a blind spot in that platform’s knowledge graph.

The verification process takes under three minutes but reveals critical gaps. Many sites achieve Google indexation through standard SEO practices but never verify Bing or Brave presence. This creates asymmetric visibility — the page appears in traditional Google results but remains invisible when ChatGPT attempts web retrieval or when Claude searches for supporting evidence.

Strategic Bottom Line: Omnichannel AI visibility requires presence in all three indexes. Optimizing for Google alone leaves 60-70% of AI search volume unable to discover or cite your content.

The Freshness Signal Architecture: Temporal Markers and Update Protocols

AI language models exhibit strong preference for recently updated content. Gotch’s analysis of commercial pages reveals a common failure pattern: sites with 2016-2020 dated elements in the source code that signal staleness to both traditional and AI crawlers. The verification methodology searches the page source for year references using a simple “20” search query.

When examining a Chicago truck accident lawyer page, Gotch discovered copyright dates and code elements from 2016 — a critical freshness penalty. The page contained scattered references to 2020 and 2023, indicating sporadic updates but no systematic refresh protocol. This temporal inconsistency creates algorithmic confusion about content currency.

The solution for commercial pages differs from informational content. While blog posts benefit from visible “Last Updated” timestamps, commercial service pages rarely display publish dates. The freshness signal must come from source code cleanliness — removing outdated year references, updating schema markup dates, and ensuring all embedded elements reflect current years.

The strategic implication: AI platforms prioritize recent information when synthesizing answers. A technically superior page with 2019 temporal markers loses citation opportunities to a mediocre competitor with fresh 2025 signals. The freshness advantage compounds in rapidly evolving industries where AI systems actively filter for recency.

Strategic Bottom Line: Temporal staleness creates algorithmic penalties invisible in traditional ranking reports but devastating for AI citation eligibility. A full source code audit for year references should occur quarterly, not annually.

The Intent Matching Protocol: SERP Analysis and Feature Qualification

Gotch’s framework emphasizes ruthless intent alignment with live search results. The methodology rejects creative interpretation — if the first page shows 99.9% commercial pages, the optimization target must be a commercial page. Attempting to rank informational content in commercial SERPs creates algorithmic friction that no amount of technical optimization can overcome.

The Rankability tool provides unbiased SERP visualization that reveals the true competitive landscape. When analyzing “Chicago truck accident lawyer,” Gotch identified a large ad block consuming top positions, followed by a local pack (Google Business Profiles), then traditional organic results. This structure indicates three distinct optimization targets: paid search, local SEO, and organic commercial pages.

The feature qualification analysis examines which SERP elements exist for the target query. In this case, no People Also Ask boxes appeared, and no AI Overview triggered — a relatively rare configuration in 2025 when AI Overviews appear on 13% of queries. The absence of these features simplifies optimization but doesn’t eliminate the need for AI-ready content architecture.

The critical insight: SERP features represent algorithmic intent signals. A query triggering AI Overviews requires different content structure than one showing only traditional blue links. Pages must include elements that qualify for visible features — structured data for rich results, FAQ schema for People Also Ask inclusion, and hierarchical content organization for AI Overview extraction.

Strategic Bottom Line: Intent misalignment wastes optimization resources. A page targeting informational queries with commercial content — or vice versa — fights algorithmic classification rather than leveraging it.

The Conversion Architecture Failure: UX Deficiencies That Kill Commercial Performance

Gotch’s analysis of the Chicago legal page exposes systematic conversion optimization failures that extend beyond traditional SEO concerns. The phone number — the primary conversion mechanism — uses a muted color that reduces visibility. Meanwhile, blue text that resembles hyperlinks but isn’t clickable creates user confusion and violates fundamental web usability principles.

The structural hierarchy problem proves more severe: the H1 headline appears below the fold, forcing users to scroll before understanding page context. This violates UX 101 principles where the primary headline should be the first element a user sees. The current layout presents an image of a truck driver with no contextual information about the page’s purpose or the firm’s service offering.

Gotch recommends a left-right format: headline and subheadline on the left, contact form on the right, with testimonials and trust signals integrated throughout. He cites a competitor site as the optimization benchmark — a design featuring a prominent phone number above the fold, clear CTAs for users not ready to call, and a clean visual hierarchy that guides users through the conversion funnel.

The form submission test revealed another critical gap: no confirmation email and no thank-you page with next-step guidance. Users submit contact information and receive only a generic “we’ll get in touch” message with no expectation setting, no additional resources, and no continued trust building. This represents a fundamental failure in lead nurturing that reduces conversion rates regardless of traffic volume.

Strategic Bottom Line: On-page SEO optimization without conversion architecture creates traffic that doesn’t convert. Commercial pages require simultaneous optimization for search visibility and goal completion — treating them as separate initiatives guarantees suboptimal performance.

The Content Depth Paradox: Word Count Optimization and Topical Coverage

Gotch’s framework introduces a nuanced approach to word count that rejects both extremes — the “more is always better” philosophy and the minimalist approach. The Rankability content optimizer calculates median word count across top-ranking competitors, eliminating outliers to identify the true optimal range. For the Chicago legal page, the target range fell between 1,300-1,500 words.

The analyzed page contained 4,300 words — nearly 3,000 words beyond the optimal range. Gotch’s recommendation: aggressive deletion focused on removing padding while preserving substance. The principle: word count isn’t a ranking factor, but words provide the context search engines and AI systems need to understand page topic and relevance. The goal is maximum information density, not maximum length.

The topical coverage analysis revealed a more significant problem: dozens of unused topics that competitors covered but the target page ignored. The Rankability tool’s “unused topics” section identified gaps in semantic coverage that create relevance deficiencies. These gaps matter more for AI systems that evaluate topical completeness when deciding whether to cite a source.

The reading level analysis exposed another critical issue: college graduate-level complexity on content targeting general consumers. Legal terminology and complex sentence structures created accessibility barriers that reduce both user engagement and AI extraction efficiency. AI platforms prefer clear, direct language that facilitates accurate summarization and citation.

Strategic Bottom Line: Optimal word count balances topical completeness with information density. Pages should target the median competitor length while ensuring every paragraph contains specific, valuable information rather than generic padding.

The AI Content Draft Protocol: Using LLMs to Build Base Content

Gotch’s methodology embraces AI-generated drafts as the foundation for human editorial refinement. The process begins with ChatGPT context gathering — using the AI to analyze the brand’s existing page and extract unique positioning elements. This context gets fed into Rankability’s AI writer, which generates a draft optimized for both topical coverage and brand voice alignment.

The generated draft for the Chicago legal page produced 1,400 words — a 67% reduction from the original 4,300-word version. More importantly, the AI draft achieved higher topical relevance scores by covering previously missing semantic elements. The draft included proper heading hierarchy, relevant subtopics, and cleaner information architecture.

The critical caveat: the AI draft represents the starting point, not the finished product. Gotch emphasizes spending one to two hours on human editorial refinement — the time saved on initial drafting gets reinvested in quality enhancement. The human editing process focuses on simplifying reading level, adding brand-specific examples, and ensuring conversion optimization elements integrate properly.

The readability issue persisted in the AI draft — it still skewed toward college graduate complexity. This represents a consistent AI writing pattern where language models default to formal, complex prose. The human editor must actively simplify sentence structure, replace jargon with plain language, and ensure the content targets the appropriate reading level for the audience.

Strategic Bottom Line: AI-generated content creates efficiency gains only when paired with rigorous human editorial processes. The time saved on drafting must be reinvested in quality refinement, not treated as pure cost reduction.

The Off-Site Dominance Reality: Why On-Page Optimization Represents Only 25% of Performance

Gotch concludes his framework with a critical context reminder: on-page optimization, despite its 49-point checklist, represents approximately 25% of total ranking power. His performance impact model allocates 50% to off-site factors (backlink profile and third-party citations), 25% to on-page optimization, and 25% to technical infrastructure and site architecture.

The strategic implication: perfect on-page execution without corresponding off-site authority building creates a performance ceiling. Sites in competitive industries cannot rank through on-page optimization alone. The backlink profile — quality, relevance, and authority of linking domains — exerts dominant influence on both traditional rankings and AI citation eligibility.

The third-party citation factor gains importance in AI search environments. When ChatGPT or Perplexity synthesizes answers, they prioritize sources with strong external validation signals. A site with comprehensive on-page optimization but weak off-site authority becomes less citation-worthy than a competitor with mediocre on-page but strong backlink profiles and media mentions.

The framework positions on-page optimization as necessary but insufficient. Sites must simultaneously execute technical optimization, content quality improvement, and off-site authority building. Focusing exclusively on any single pillar creates imbalanced optimization that underperforms integrated strategies.

Strategic Bottom Line: On-page SEO creates the foundation for visibility but cannot compensate for weak off-site authority. The 49-point checklist ensures technical eligibility; backlink acquisition and third-party validation determine competitive positioning.

Summary

The dual-engine optimization framework reveals that traditional SEO and AI search require overlapping but distinct technical architectures. The 49-point checklist addresses foundational requirements — HTML text retrieval, triple-index presence, freshness signals, and intent alignment — that determine whether pages qualify for both traditional rankings and AI citations. The Chicago legal page analysis exposed common failure patterns: JavaScript rendering barriers, outdated temporal markers, conversion architecture deficiencies, and topical coverage gaps.

The strategic hierarchy matters: on-page optimization represents 25% of performance, technical infrastructure another 25%, with off-site authority controlling the remaining 50%. Perfect on-page execution without backlink development and third-party validation creates a performance ceiling that no amount of technical refinement can break through. The framework positions on-page SEO as the necessary foundation that enables but doesn’t guarantee competitive visibility.

Implementation requires systematic verification across crawl accessibility, HTML architecture, index presence, freshness signals, intent matching, conversion optimization, content depth, and readability. Each checkpoint addresses specific algorithmic requirements that, when failed, create invisible barriers to discovery and citation. The methodology rejects generic best practices in favor of evidence-based optimization tied directly to how search engines and AI systems evaluate, retrieve, and cite web content.