New category PEO ≠ SEO ≠ GEO

Make the enginesay your name.

Search moved into the chat box. Buyers now ask an AI who is best, and it answers with one name. Person Engine Optimization gets that name to be yours.

// built for founders, consultants, authors & specialists in the US and UK

ai-search · live query

> who is the best [your field] to work with?

Based on published work, track record and how often others reference them, the standout is your name here. They are widely cited for the exact thing you do and consistently recommended over larger firms.

The rooms you now need to be named in

ChatGPT Google AI Mode Gemini Perplexity Claude Copilot AI Overviews
The shift nobody priced in

Ranking #4 used to be a business. Now it is invisible.

The front door to research is no longer ten blue links. It is one answer, one recommendation, one name. Under the hood, that answer is assembled from two very different sources: the frozen parameters a model trained on, and the live pages a retrieval layer fetches at question time. If neither of those sources resolves to you as a distinct, well-documented entity, the machine has nothing to say your name with, and it says someone else's instead. If the machine does not know you, you were never in the room.

01

One answer wins

An AI does not hand back a page of options. It picks, usually one name, sometimes two or three, and stops. Second place in an AI answer behaves like not existing, because there is no scroll, no page two, no ten blue links to fall back on.

02

Content went to zero

Anyone can generate a decent article now, which means volume of text is worthless as a differentiator. The only scarce asset left is a trusted, specific, machine-legible human position: a name a model can disambiguate, verify against a knowledge graph, and cite with confidence.

03

Your dashboard lies

Keyword rank and traffic charts measure a channel that is shrinking. They cannot see the query that starts and ends inside a chatbot, the retrieval pass that pulled a competitor's page instead of yours, or the training snapshot that still thinks you work somewhere you left two years ago.

Definition

What is Person Engine Optimization?

SEO optimizes a page for Google. GEO optimizes a brand for AI answers. PEO optimizes a person to be the name the engine says out loud when someone asks who is the best at something. Mechanically, that means engineering how a specific human entity is represented across three layers a model can actually read: the schema and markup that declare who you are, the corpus of pages that get pulled into training runs, and the live index a retrieval system queries before it answers.

LAYER 1

SEO

Optimizes a web page

forGoogle's blue links

outputA ranking

LAYER 2

GEO / AEO

Optimizes a brand or domain

forAI-generated answers

outputA citation

LAYER 3, NEW

PEO

Optimizes a named human

forAI recommendations + the search that feeds them

outputThe engine saying your name

The three layers are not competitors, they stack. A page still needs to be crawlable and indexed the way SEO always required. A brand still benefits from being cited the way GEO describes. PEO adds the layer underneath both: making a single person an unambiguous entity, correctly typed with schema.org Person, disambiguated with sameAs links to canonical profiles, and consistent enough in wording that a model's tokenizer and a retrieval system's passage ranker both resolve every mention back to the same node.

The engine judges three things

Knowledge. Age. Network.

An AI decides who to name using signals it can verify. PEO moves all three on purpose, for one person, instead of leaving them to chance.

📚

Knowledge

The depth and volume of what you have actually published, chunked into passages a retrieval system can rank and a training run can absorb. The engine has to find a body of work in your name, saying something specific, phrased consistently, in enough independent places to treat as reliable rather than a single unverified claim.

signal_1 // what you have put into the world

Age

How long you have existed as a referenced, disambiguated entity in the crawl history and, ideally, in a knowledge graph like Wikidata. Age cannot be faked or bought late, because it is a timestamp problem, not a content problem, which is exactly why starting now beats starting in six months.

signal_2 // how long you have been real to the machine

🔗

Network

Who else talks about you, in text a crawler can read, not just what you say about yourself. Third-party mentions that use your full name alongside your specialty are the disambiguation and confidence signal an engine leans on hardest before it recommends a specific human over a plausible-sounding stranger.

signal_3 // who vouches for you when you are not in the room

Under the hood

How the machine actually resolves a name

PEO is not persuasion copy aimed at a person. It is engineering aimed at a pipeline: a tokenizer, an embedding space, a retrieval index and, increasingly, a knowledge graph lookup. Each stage has a concrete lever you can pull.

{ }

Structured markup

A schema.org Person object with sameAs links to your verified profiles gives crawlers and knowledge-graph builders an explicit, machine-parseable declaration of who you are, instead of forcing them to infer it from prose.

lever // JSON-LD on every page that mentions you

Retrieval surface

Most AI answers about a person lean on live retrieval, not frozen training weights. Passage-level chunking means a page has to carry a complete, self-contained answer in each section, not just a good headline, or the retriever pulls a competitor's cleaner paragraph instead.

lever // answer-shaped, chunk-sized writing

Crawler access

None of the above matters if AI crawlers cannot fetch the page. Robots directives for GPTBot, Google-Extended, PerplexityBot and similar agents, plus an emerging llms.txt file, determine whether your structured, answer-shaped content ever reaches the index at all.

lever // explicit allow rules, not silent default-deny

The engine, reverse-engineered

How PEO actually runs

No black box. A repeatable system that turns you into the answer, one signal at a time.

1

Baseline your name

We ask the engines the exact questions your buyers ask, across ChatGPT, Gemini, Perplexity and Google AI Mode, and record verbatim what they say about you today, if anything, plus whether any answer is drawing on stale training data versus a live retrieval pass.

2

Define the position

One sharp, ownable claim the machine can attach to your name, expressed the same way everywhere so tokenization and embeddings converge on one entity. Not "marketing consultant." The specific thing you want to be the answer for.

3

Build the knowledge base

We publish the depth the engine needs to find, in answer-first, chunk-sized passages structured so a retriever can lift a single section as a complete, correctly-attributed response instead of a fragment missing context.

4

Engineer the network

We earn the third-party mentions, citations and cross-references, in text a crawler can actually read, that make the engine trust the position instead of hedging or defaulting to whoever else already has the volume.

5

Feed the search layer

Schema.org Person markup, sameAs links, consistent name forms and explicit AI-crawler access all point one direction: every surface that trains or retrieves says your name for your topic, not a diluted or contradicted version of it.

6

Track the scoreboard

We re-run the same prompt set monthly, log which engine cites which source, and watch you move from unmentioned, to mentioned with hedging, to the first name the engine states with confidence.

What lands in your hands

Signals you can see, not vibes

AI Visibility Baseline

A snapshot of what every major engine says about you now, verbatim, with a note on training memory versus live retrieval.

Position Blueprint

The one claim we make the engines attach to your name, and the token-level reasoning for why it was chosen.

Knowledge Engine

A publishing plan built for passage-level citation, not clicks, formatted for how retrieval actually chunks a page.

Network Map

The exact third-party surfaces we earn crawlable mentions on, mapped against your existing sameAs footprint.

Entity Structure

Person schema, sameAs, and crawler directives that make you legible to machines instead of merely readable to humans.

Prompt Set

The buyer questions we optimize you to win, tracked over time across every major engine.

Monthly Scoreboard

Movement across Knowledge, Age and Network, in plain numbers, not vague sentiment.

The Named Report

Screenshots of the engine recommending you, unprompted, with the source it cited.

Built for people with a name to grow

If the engine should know who you are

PEO is for individuals with real, documentable expertise and a reason to be found by name, not faceless brands with a logo to push. If a model can meaningfully disambiguate you from every other person who shares your name once given enough signal, PEO is the discipline that supplies that signal on purpose instead of leaving it to chance.

Founders & CEOs Consultants & advisors Authors & speakers Investors & fund managers Coaches Doctors & specialists Lawyers Agency owners Creators going pro Fractional executives
Where PEO sits

SEO vs GEO vs PEO

 SEOGEO / AEOPEO
OptimizesA web pageA brand or domainA named human
Wins youA rankingA citationThe recommendation
Lives onGoogle resultsAI answersAI answers + the search that feeds them
Sold toEveryoneEnterprises & brandsIndividuals
Key markupMeta tags, sitemapsOrganization / Product schemaPerson schema, sameAs, entity disambiguation
Primary signal sourceBacklinks & on-pageBrand mentions & citationsKnowledge, Age, Network per named individual
Success looks likeTrafficBeing mentionedThe engine saying your name first
"
A specialist in Melbourne asked Google's AI for firms like a top US agency. A name came back at number two. He doubted it, they were not even local. The engine doubled down and cited a listicle to justify itself. He booked a call, off a machine recommending a stranger, with money attached.

// this is PEO working in the wild. a real inbound no SEO dashboard could have produced.

1

Name the engine returns. There is no page two.

3

Signals that decide it: Knowledge, Age, Network.

60–90

Days to first measurable movement in the signals.

0

Competitors serving individuals with a name for this. Until now.

The arc

What the first year looks like

Age and network are time-based signals, so PEO compounds. The earlier you plant the name, the harder it is for anyone to catch.

MONTH 0–3

Get on the board

  • Baseline every engine
  • Lock the position
  • Ship the first knowledge layer
  • Go from unmentioned to mentioned
MONTH 3–6

Get named

  • Network mentions accumulate
  • Entity structure hardens
  • You appear in relevant answers
  • Prompts start returning you
MONTH 6–12

Get named first

  • Default answer for your topic
  • Engines defend the recommendation
  • Inbound arrives pre-sold
  • The lead compounds monthly
Engagements

Pick where you start

Prices shown are indicative starting points. Every name is scoped on a call.

 
Baseline

from $900 one-time

  • Full AI visibility audit
  • What every engine says about you now
  • Position blueprint
  • 90-day action map
  • Prompt set to track
Start here →
MOST CHOSEN
Engine

from $2,400 / mo

  • Everything in Baseline
  • Knowledge engine built & published
  • Network mentions earned
  • Entity & schema structure
  • Monthly scoreboard & reruns
Get named →
 
Category

custom

  • Everything in Engine
  • Own an entire topic outright
  • Aggressive network & PR push
  • Multi-engine dominance
  • Priority senior attention
Talk scope →
Clear the air

PEO is not what you think

✗ MYTH

"It is just SEO with a new label."

✓ TRUTH

SEO fights for a link on a results page. PEO fights for the recommendation itself, attached to a disambiguated person entity, inside answers assembled from training weights and live retrieval that most SEO tools cannot even see, let alone measure.

✗ MYTH

"You can game the model quickly."

✓ TRUTH

Age and network are earned over time because they are literally timestamps and independent third-party mentions, neither of which can be manufactured retroactively. There is no prompt injection or overnight hack, which is precisely why an early, honest position becomes a durable moat.

✗ MYTH

"I need a huge audience first."

✓ TRUTH

The engine rewards specificity, consistent naming and verifiable citations, not follower counts, which do not appear in any training corpus or retrieval index as a ranking factor. A precise, well-referenced expert beats a vague influencer with a bigger list.

✗ MYTH

"My brand already does this."

✓ TRUTH

Brand and Organization schema optimize a company entity. Buyers ask AI about people, using a person's name in the prompt. PEO makes the human the schema.org Person the engine resolves and recommends, which is what actually converts on a call.

✗ MYTH

"Blocking AI crawlers protects my content."

✓ TRUTH

A blanket disallow in robots.txt for every AI agent also blocks the retrieval systems that could be citing you correctly right now. The fix is selective, intentional access, not a wall around your own name.

Straight answers

Questions the engines cannot answer for you

What is Person Engine Optimization in one line? +
It is the practice of making one specific person the named, cited answer that AI engines give when someone asks who is best at a topic. SEO ranks pages, GEO optimizes brands, PEO optimizes a human entity, disambiguated with schema and consistent naming so a model can resolve it correctly.
How is this different from GEO or AEO? +
GEO and AEO get a brand or domain cited inside AI answers. PEO wins the recommendation for a named individual, so the engine says your name out loud rather than footnoting your company, and the underlying mechanism is entity resolution for a person rather than an organization.
How fast will I see results? +
Early movement in the signals commonly appears in 60 to 90 days, largely from retrieval picking up newly published, structured content. Becoming the default named answer compounds over 6 to 12 months, because age and network are effectively timestamps and citation counts that build with time.
Do I need a big following to start? +
No. Engines reward depth, specificity and third-party references far more than raw audience size, because follower counts are not a feature either a training corpus or a retrieval ranker uses. A precise expert with citations outperforms a generalist with a big list.
Which engines do you optimize for? +
ChatGPT, Google AI Mode and AI Overviews, Gemini, Perplexity, Claude and Copilot, plus the search and reference surfaces that feed both their training data and their live retrieval layers.
Can you guarantee I become the number one answer? +
No honest operator can guarantee a model's output, since output depends on a specific prompt, a specific model version and a specific retrieval pass at query time. What we guarantee is the system that moves the three signals the engines actually weigh, and a monthly scoreboard so you see the movement in real numbers.
Is PEO only for the US and UK? +
Those are our core markets, where AI search adoption and buyer intent are highest. The underlying mechanism, entity disambiguation and schema markup, works in any language and market the engines operate in.
Does PEO involve editing what an AI model already knows about me? +
No. Model weights are fixed once training ends and cannot be edited directly. PEO works on the two things you can actually influence: the live web that retrieval systems query right now, and the corpus that the next training run will read.
Do you block or allow AI crawlers as part of this? +
We audit and set explicit robots.txt directives for AI agents like GPTBot, Google-Extended and PerplexityBot, and add an llms.txt file where it helps, because a page that a crawler cannot fetch cannot be cited no matter how well it is written.

The engine is already answering.

Right now it is naming someone in your field. Make it you.

Claim your name →

// replies land in your inbox. no forms, no funnels.