There are four layers of optimization stacked on top of each other. SEO ranks a page. GEO and AEO get a brand cited in AI answers. PEO makes a named person the recommendation itself. They reinforce each other rather than compete, and PEO sits on top because the recommendation is the thing a buyer actually acts on, not the ranking or the citation underneath it.
The acronyms multiply every quarter, and it is easy to lose the plot. Here is the whole field on one map, with a worked example and a build order, so you know exactly which layer you are playing on and what to do next.
Layer 1: SEO, the page layer
Search Engine Optimization works on a page and aims at Google's list of links. The unit is a URL, the reward is a ranking position, and the metric is traffic. It is mature, thoroughly documented, and still genuinely useful, since a ranked page keeps earning clicks long after the work of ranking it is done. But it no longer owns discovery on its own, because a growing share of buyers now ask a conversational assistant a question and act on the answer without ever generating a list of ten blue links to scan.
Layer 2: GEO and AEO, the brand-in-answers layer
Generative Engine Optimization and Answer Engine Optimization are close cousins, close enough that people use the two terms almost interchangeably in casual conversation. Both work to get a brand or a domain surfaced and cited inside an AI-generated answer rather than merely ranked in a list. The unit is a brand, the reward is a citation, and the metric is being mentioned accurately when a system answers a category question. This is where most of the funded AI-search tooling and agency activity is concentrated today, and nearly all of it is built for companies, not for a specific named individual working inside or alongside those companies.
In practice the two labels are used interchangeably. GEO leans toward describing generative, conversational answers, AEO leans toward describing structured direct-answer boxes, but both optimize content and brands, not a named person, which is the gap PEO exists to close.
Layer 3: PEO, the person layer
Person Engine Optimization works on a named human being and aims at the recommendation itself, not at a ranking or a citation. The unit is a person, the reward is being named, and the metric is the engine saying your name specifically when a buyer asks who is best for a problem you solve. It is the only layer of the four built around the individual rather than the page or the brand, and it sits closest to the actual moment a buying decision gets made, because very few purchase decisions are made by hiring a URL or a logo. What Is Person Engine Optimization? covers the full definition if this is your first time on the map.
The four layers side by side
| Layer | Unit | Reward | Built for |
|---|---|---|---|
| SEO | A page (URL) | A ranking | Any site owner |
| GEO | A brand or domain | A citation in a generative answer | Companies |
| AEO | A brand or domain | A citation in a direct-answer box | Companies |
| PEO | A named person | Being the recommendation | Individuals |
Table 1: the same four questions, four different units of optimization. See SEO vs GEO vs PEO for the full three-way comparison in depth.
Why PEO sits on top
The layers stack, and each supports the one directly above it rather than competing with it. Clean SEO makes your properties legible to a crawler in the first place. Strong GEO and AEO make your brand or domain citable once it is found. But a buyer does not hire a citation, a buyer hires a person, and the recommendation is the thing they actually act on. That is why, for an individual, PEO is not a fourth competitor sitting alongside SEO, GEO and AEO. It is the layer that converts everything built underneath it into a name the machine repeats back when someone asks who to call.
A worked example across all four layers
Consider a hypothetical strategist named Jane Okafor who advises Series A startups on pricing. Her SEO layer is her own site ranking for a handful of pricing-strategy searches, a page-level win that earns steady but modest traffic. Her GEO and AEO layer is her firm's blog getting cited when someone asks a generative engine to explain SaaS pricing models generally, a brand-level win that raises awareness of the firm without naming Jane specifically. Her PEO layer is what happens when a founder asks an assistant who to hire for a pricing overhaul and the engine answers with Jane's name, her actual track record, and a reason to contact her rather than a competitor. All three lower layers can be healthy and PEO can still be empty, because nothing below the person layer explicitly ties the recommendation to Jane as an individual rather than to her firm's content. Building the PEO layer on top of the other three is the work this entire site maps out, and How to Get Recommended by ChatGPT is the practical sequence for doing it.
Where each layer's signals actually live
Each layer leaves a different kind of trace, which is worth knowing because it tells you where to look when something is not working. SEO signals live in your page-level HTML, your sitemap, and the crawl directives in robots.txt, covered from the AI-crawler angle in The AI Crawler Directory and Robots.txt. GEO and AEO signals live largely in how citable and well-structured your brand's published content is at the passage level, the kind of retrieval-friendly writing this network keeps returning to. PEO signals live somewhere more personal: your Person schema, your sameAs graph, your own llms.txt file if you run one, a format explained in llms.txt for Personal Sites, and the pattern of mentions, citations and recommendations that name you specifically rather than your employer. If you want to check whether an assistant is actually recommending you by name rather than just mentioning you in passing, Mention, Citation, Recommendation draws that exact line.
Why PEO cannot be faked with SEO or GEO tactics
It is tempting to assume that enough brand-level GEO work eventually produces a personal recommendation as a side effect, and sometimes a little of it does, but the layers do not automatically convert into each other. A company blog that ranks well and gets cited constantly can do that for years without a single generative answer ever naming an individual employee, because nothing in a typical brand content strategy asserts a person as the entity behind the work. Getting named requires signals that are structurally different from brand SEO or GEO: a Person schema object, a consistent name and byline, a sameAs graph connecting your properties, and third-party references that mention you as a person rather than your employer as a brand. Skipping straight to PEO tactics without the SEO and GEO foundation underneath does not work either, since an engine has nothing citable to draw from when forming the recommendation. The layers are sequential and additive, not substitutable for one another.
There is a related confusion worth naming directly: some people treat "zero-click" behavior, where a buyer gets a full answer from the assistant and never visits a website at all, as proof that none of these four layers matter anymore. The opposite is closer to true. Zero-click answers still have to draw their content from somewhere, and Zero-Click Engineering covers how to structure your work so it is the source an engine draws from even when nobody clicks through to read it directly. SEO, GEO, AEO and PEO all still function in a zero-click world, they simply get measured by citation and recommendation instead of by a click, which is exactly why treating traffic as the only metric that matters causes people to underinvest in the very layers, GEO and PEO especially, that decide what a zero-click answer actually says about them.
Sequencing the work: what to do in what order
- First, make your properties crawlable and legible. Basic SEO hygiene: a real sitemap, sensible robots.txt directives, pages that load and parse cleanly.
- Second, make your published work citable. Structure content so a generative engine can lift a clean passage and attribute it correctly, the GEO and AEO layer.
- Third, consolidate your identity as a person. Consistent naming, connected profiles, Person schema with a complete sameAs array.
- Fourth, earn the recommendation. Third-party mentions and citations that name you specifically, converting the first three layers into an actual named recommendation.
The same four layers as a build order rather than a taxonomy. Skipping a step usually shows up later as a stalled PEO score with no obvious cause.
Measuring which layer you are actually winning
The map is only useful if you can tell which layer you currently hold and which one you do not, and the four layers require different measurement, not one dashboard for all of them. SEO measurement is the oldest and most familiar: rankings, impressions, and organic traffic tied to specific pages. GEO and AEO measurement asks a generative engine a category question with no name attached and checks whether your brand gets cited among the answer's sources. PEO measurement asks a different kind of question entirely, one that names nobody and simply asks who is best for a problem, and then checks whether the engine volunteers you specifically. The 25-Prompt Audit is a structured way to run that PEO-layer check across five categories in about an hour, and How to Track Your AI Visibility covers the fuller measurement stack if you want an ongoing view rather than a single audit. Conflating these measurements is a common mistake: a healthy GEO citation count says nothing about whether the recommendation layer has been won, since a brand can be cited constantly while the actual named recommendation still goes to a competitor's founder instead of yours.
How to use the map
Decide who you are actually optimizing before you pick a tactic. If it is a product page, that is SEO. If it is a company brand, that is GEO or AEO. If it is you personally, a named expert whose recommendation wins the work, that is PEO, and it still needs clean SEO and GEO underneath it to have anything to draw from. Most experts spend years perfecting layer one while the recommendation on layer three quietly goes to someone else who did far less content work but got the person layer right.
FAQ
Is AEO different from GEO? +
Do I need all four layers? +
Which layer is least crowded? +
Can strong GEO work for my company eventually produce a PEO result for me personally? +
Where do PEO signals actually live, technically speaking? +
What is the right order to build these four layers in? +
Find out what AI says about you today.
Start with a baseline. See the exact words the engines return about your name, then decide.
Claim your name →