The Complete Guide to AEO

The 7 Signals of Answer Engine Optimisation

AI doesn't recommend brands it doesn't understand. To earn AI visibility, your brand needs to send clear, consistent, and structured signals across 7 dimensions.

What is AEO?

Answer Engine Optimisation (AEO), also known as generative engine optimisation, is the practice of making your brand discoverable, citable, and recommendable by AI systems, including ChatGPT, Perplexity, Google AI Overview, and other AI-powered answer engines.

We’re a good fit if:
We’re a good fit if:

When someone asks an AI assistant "What's the best agency for B2B marketing in Singapore?" The answer it gives is not random. AI draws from structured, authoritative, and consistent information across the web to decide who to recommend.

AEO is the discipline of engineering that information so AI recommends you.

But most AEO approaches get it backwards. They start with tools. Pick a topic, generate prompts, track rankings. That means you're optimising for prompts you guessed your audience uses.

At Underscore, we start with the brand. Then we map the audience. Then we engineer 7 distinct signals that collectively determine whether AI recommends you or your competitor.

The signals

The 7 Signals Framework

AI doesn't recommend brands it doesn't understand. To earn AI visibility, your brand needs to send clear, consistent, and structured signals across 7 dimensions, organised in three tiers.

Tier 1 — Strategic Foundation. Brand Signal sits at the top. It defines who you are, who you serve, and what makes you different. Without it, every other signal lacks direction.

Tier 2 — Owned Presence. These three signals cover everything on your own site: how it feels (Experience), what it says (Content), and how AI agents can access it (Agent). You have full control here.

Tier 3 — External Presence. These three signals cover everything beyond your site: your search visibility (Search), your social footprint (Social), and what third parties say about you (Authority). You influence these but don't fully control them.

deep dive

Explore Each Signal

Each signal page covers the core problem, key components, common mistakes, and actionable audit checklists.

‍🎯

Brand Signal

What defines you as a brand and who your audiences are.

Brand Signal is the foundational layer. It defines who you are, what you stand for, who you serve, and what makes you different, articulated in ways AI can parse and repeat.

This is the umbrella for everything else. AI needs explicit, factual differentiation, not a beautiful logo or clever tagline. It needs clear statements about what makes you different that it can extract, understand, and repeat.

Brand Signal also determines your prompt strategy. Most AEO tools auto-generate tracking prompts. But how do you know those prompts are what your customers actually ask AI? The honest answer: you don't. At Underscore, we use a 4-source methodology to build prompt lists with up to 80% accuracy, compared to the industry standard of essentially guessing.

✍️

Content Signal

What you publish on your own site.

Content Signal is the body of content on your owned properties, structured so AI can extract, understand, and cite it. This is where most AEO advice starts. We believe it's the third signal, not the first.

Content Signal is not just "write good content." It's about structuring content so AI systems can extract key facts, attribute statements to your brand, and cite your pages as sources.

This signal also introduces a critical distinction most practitioners miss: the difference between being cited (AI links to your content as a source) and being mentioned (AI names your brand as a recommendation). These require different strategies, and understanding the difference changes how you approach content.

Read deep dive

Agent Signal

The technical infrastructure that allows AI agents to crawl, parse, and understand your site.

Agent Signal is the technical layer, your site's ability to be crawled, parsed, and understood by AI agents. This covers structured data, schema markup, crawl accessibility, and emerging AI-agent protocols.

This is the most technical of the 7 signals and the one fewest marketers think about. As AI systems shift from pre-trained knowledge to real-time web retrieval (Perplexity, ChatGPT Browse, Google AI Overview), the technical accessibility of your site becomes a direct ranking factor.

A site with perfect Content Signal but poor Agent Signal is like a library with great books but locked doors.

Read deep dive
👤

Experience Signal

How users perceive and interact with your brand.

Experience Signal is about brand perception and giving users the opportunity to understand what you sell. This signal has never changed and will always be relevant, AI or no AI.

This is the signal most AEO practitioners ignore because it's not "AI-specific." But when AI sends someone to your site via a citation link, what do they land on? A dead-end blog post or a page that moves them toward a decision?

Experience Signal ensures the handoff from AI recommendation to your owned experience is seamless. If steps 1-3 of the AI discovery journey are optimised but the landing experience is broken, you've wasted the AI visibility.

Read deep dive
🔍

Search Signal

Your existing search practices and keyword-focused content efforts.

Search Signal bridges traditional SEO with AEO. Your existing keyword data and search rankings are not irrelevant in an AI world. They're one of the inputs that build accurate prompt lists.

SEO isn't dead. Pages that rank well on Google are more likely to be sourced by AI systems. Your existing SEO infrastructure is an asset, not a liability.

The value of Search Signal is in translation. Taking high-performing keywords and reframing them as the natural language questions someone would ask an AI assistant. SEO keyword data is the third most reliable source for building AEO prompt lists, after CRM data and user research.

Read deep dive
🏛️

Authority Signal

Your presence on third-party sites run by other organisations.

Authority Signal is your earned media footprint. PR coverage, guest articles, directory listings, podcast appearances, and any instance where a third-party site vouches for your brand.

This is the signal that makes the difference between being cited and being mentioned. AI systems cross-reference multiple sources. If your claims only exist on your own website, AI treats them with lower confidence. If independent third parties say the same things, AI's confidence in recommending you increases dramatically.

For newer brands, Authority Signal is often the weakest, and it's the reason why citation comes before mention in the content strategy.

Read deep dive
💬

Social Signal

Your presence on LinkedIn, Substack, Reddit, Instagram, YouTube, and other social platforms.

Social Signal is your brand's footprint across social platforms. AI systems increasingly source information from social content. Reddit threads, LinkedIn posts, YouTube transcripts, and Substack newsletters all feed into AI training data and real-time retrieval.

This is a signal most AEO practitioners completely ignore. But Reddit is explicitly licensed as training data for multiple AI models. LinkedIn posts are crawled and indexed. YouTube transcripts are searchable by AI. Social platforms are becoming primary sources, not secondary ones.

Social Signal also doubles as a research input. What questions do people ask in Reddit threads about your industry? What pain points surface in LinkedIn comments? These map directly to how your audience prompts AI.

Read deep dive
Methodology

How We Apply the 7 Signals

A 5-phase methodology that starts with brand and ends with measurable AI visibility.

1

Brand Signal Foundation

Define the Brand Signal — a strategic articulation of who you are, who you serve, and what makes you defensibly different.

2

Audience-First Prompt Mapping

We audit all 7 signals, identifying which are strong, weak, or absent. This produces a prioritised action plan.

3

Signal Assessment

We audit all 7 signals, identifying which are strong, weak, or absent. This produces a prioritised action plan.

4

Signal Engineering

We execute across the signals that move the needle fastest. Typically Content Signal and Agent Signal first, then Social Signal and Authority Signal.

5

Measurement

AEO is an awareness and consideration play. ROI shows up as increased bot crawl activity, improved citation metrics, first-time visits from AI referral sources, and overall brand metric uplift.

Regional Considerations

AEO works differently region by region. The same prompt in English vs Mandarin may return different recommendations. AI models have uneven knowledge depth across markets. For brands operating across multiple markets, each region needs its own Signal assessment.

Frequently Asked Questions

  • What is AEO (Answer Engine Optimisation)?

    AEO stands for Answer Engine Optimisation. It is the practice of optimising your brand's online presence so AI-powered search tools like ChatGPT, Perplexity, and Google AI Overview can discover, cite, and recommend you. Unlike traditional SEO, which focuses on ranking in search engine results pages, AEO focuses on earning visibility within AI-generated answers. Some practitioners also refer to this as generative engine optimisation (GEO).

  • What is the difference between AEO and SEO?

    SEO optimises for search engine rankings — getting your pages to appear in Google results. AEO optimises for AI recommendations — getting your brand cited and mentioned when someone asks an AI assistant for advice. They're complementary, not competing. Strong SEO feeds your Search Signal, which is one of the 7 Signals of AEO.

  • Do I need to do all 7 Signals at once?

    No. We typically start with Brand Signal (foundation), then Content Signal and Agent Signal (most controllable), then layer in Social Signal and Authority Signal over time. The priority depends on your current strengths and gaps.

  • How long does it take to see results from AEO?

    AI models update their knowledge at varying intervals. Some results (like Perplexity citations from real-time retrieval) can appear within weeks. Others (like ChatGPT mentions from training data) take longer. The typical engagement sees measurable citation improvements within 60-90 days.

  • How do you measure AEO ROI?

    Conversion is not a direct-attribution metric for AEO. The real measurement includes first-time visit capture ("How did you hear about us?"), UTM tracking from AI citation links, overall brand lift in page views and bot crawl activity, and pipeline attribution from leads that mention AI as a discovery channel.

  • Is traditional SEO dead?

    No. SEO is one of the 7 Signals (Search Signal). Pages that rank well on Google are more likely to be sourced by AI systems. Your existing SEO infrastructure is an asset that feeds into your AEO strategy.

Ready to make your website work harder for you?

Book a short call to explore which service fits your situation.