The AEGIS A·E·G·I·S Framework — Redesigning Content Structure Around the 5 Stages AI Uses to Read Your Site

Why "How Many Pages" Is the Wrong Question — It's About Structure

Many brands write blog posts diligently, polish their product pages, and fill out an FAQ section. Yet they still go unmentioned in AI search. The reason is usually not the quantity of content, but its structure. AI answer engines don't browse a site the way a human does. They read the entire sitemap.xml at once and structurally judge how the URLs relate to one another, what kind of organization runs the site, and how far they can be trusted as a source of information.

The problem is that most startup sites leave this structure to chance. Blog posts accumulate, but there's no author bio page. A product page exists, but there's no guide explaining how to actually use it. To an AI, a site like this reads as "content-rich but ambiguously trustworthy." The AEGIS Framework was built to solve exactly this problem — classifying every URL on a site into five stages, Acquisition · Engagement · Generation · Intelligence · Sales, and diagnosing which stage is empty.

What Is the AEGIS Framework? — Its Relationship to the AEGIS Index (Score)

First, a distinction that's easy to conflate. The AEGIS Index is a composite score from 0–100 that measures a brand's AI visibility (SEO×0.25 + AEO×0.25 + GEO×0.50 — this formula and its 38 metrics were covered in depth in our earlier post, "AEGIS Index Methodology"). The AEGIS Framework (A·E·G·I·S), by contrast, is not a score at all — it's a structural model that classifies a site's URLs by role. The score tells you "where you stand right now." The Framework tells you "which part is empty, and that's why the score won't move." Both operate within the same diagnostic tool (AEGIS Insight), but one is the result and the other is the root-cause diagnosis.

When AEGIS Insight scans a site's structure, it collects up to 300 URLs from sitemap.xml and automatically classifies them into these 5 stages. The order isn't arbitrary — it mirrors the actual sequence in which a consumer comes to trust a brand.

The 5 Stages in Detail — Definitions and SEO/AEO/GEO Checklists

Stage	Definition	Representative Pages
A — Acquisition	Securing AI visibility · authority pages that prove who this organization is	About, Contact, Author, Team
E — Engagement	Brand entry · the pages consumers meet first	Home, Product, Pricing, Services
G — Generation	AI citation generation · evidentiary content AI cites in its answers	Blog, Research, Insights, Reports
I — Intelligence	Practical resources · content offering usage instructions and decision criteria	Guides, Tools, FAQ, Docs
S — Sales	Conversion creative · pages that drive actual action	Comparison pages, Landing pages, Categories

A (Acquisition) — Nothing Starts Without Authority

AI answer engines don't cite content without evidence. About, Author, and Contact pages are the evidence base for "who wrote this" and "does this organization actually exist." This is also where E-E-A-T (Experience, Expertise, Authoritativeness, Trustworthiness) signals concentrate.

Checklist: Author profiles marked up with Person JSON-LD schema · team, history, and contact info on the About page · Organization schema structuring business entity information · real credentials and experience in author bios

E (Engagement) — First Impressions Are Structural Signals Too

Home, product, and pricing pages get the most human traffic, but they're also the primary evidence AI uses to define "exactly what this service does." If this is ambiguous, AI will avoid mentioning the brand in its answers altogether.

Checklist: inverted-pyramid structure placing the core answer in the first paragraph of the homepage · SoftwareApplication/Product schema · PriceSpecification schema on pricing pages · core category clearly stated in title and meta description

G (Generation) — Where AI Actually Cites You

Blog posts, research, and reports are the content AI actually draws on when generating an answer. The reason the AEGIS Index weights GEO twice as heavily is that content at this stage ultimately determines whether you get cited at all.

Checklist: Article/BlogPosting schema · a direct-answer paragraph (a definitional sentence under 50 characters) placed at the top of the article · FAQPage schema structuring Q&A pairs · verifiable statistics and data as supporting evidence

I (Intelligence) — Answering "How"

Guides, tools, and FAQs answer the practical questions consumers ask right before a purchase decision — "how do I use this," "how is this different." AI prioritizes content at this stage for HowTo-style and comparison-style answers.

Checklist: HowTo schema structuring step-by-step procedures (the highest-weighted schema type) · comparison tables written in actual HTML table markup · FAQ titles phrased exactly as real search queries

S (Sales) — When Structure Breaks, Conversion Breaks Too

Comparison pages, categories, and landing pages sit right before final conversion. No matter how solid the first four stages are, if this stage is empty, AI has no basis to suggest a "next action."

Checklist: concrete differentiators stated in competitive comparison content · clear value propositions repeated around CTAs · BreadcrumbList schema on category pages to signal position within the site

Self-Diagnosis — Check Your Site's A·E·G·I·S Distribution With AEGIS Insight

You don't need to understand this framework only in theory. Enter a URL into AEGIS Insight and it will parse your sitemap.xml directly and automatically classify your URLs into the 5 stages above. This scan is a token-free pre-analysis step, so it returns real measured values within seconds. On the results screen, clicking each category shows you exactly which URLs fall into that stage.

What matters here isn't the URL count — it's which stage is empty. Two failure combinations show up constantly.

A Common Failure Pattern — Sites That Stack Only G While A and I Sit Empty

The most common pattern on early-stage startup sites is diligently stacking blog content (G) while neglecting About/Author pages (A) or guides/FAQ (I). To AI, this reads as "content-rich, but I don't know who wrote it and there's no explanation of how to actually use it." As a result, even the G-stage content gets deprioritized in citation — because the weighted AEO/GEO metrics tied to author profile and E-E-A-T signals drop.

The opposite pattern is just as common: solid A and E (a traditional corporate-brochure-style site) with no G at all. A site like this reads as "trustworthy, but with no evidentiary content for AI to actually cite," and scores poorly on the AEGIS Index's GEO metrics. Ultimately, if even one of the five stages is empty, it drags down the effectiveness of the other four — which is exactly why the right question isn't "how many pages do we have," but "is our structure complete."

A 90-Day Execution Roadmap

Period	Priority Stage	Actions
Month 1	A · E	Build or strengthen About/Author pages, add Organization/Person schema, rewrite home and product pages in inverted-pyramid structure
Month 2	G	Publish blog content with direct-answer paragraphs, apply FAQPage/Article schema, add verifiable data as supporting evidence
Month 3	I · S	Build HowTo guides and comparison content, complete structured data on category/landing pages, re-diagnose to confirm 5-stage balance

Closing

Site structure in the AI search era operates on a different logic than navigation designed for humans. The AEGIS Framework simplifies that logic into five stages, shifting the question from "what more should we write" to "which stage is empty, and dragging everything else down with it." Check your site's A·E·G·I·S distribution for free at AEGIS Insight.