Question.Marketing

The dated AEO record

SEO Isn't Dead. The Name Is.

Agents don't search, they choose. Simon Young on why the name SEO no longer describes the job, the alternatives (AEO, GEO, AAO, AX) and the cited, recommended, chosen ladder.

Published Running time 34:16

What does this video cover?

  • 01The name of the job has always followed the machine: search engines gave us SEO, answer engines gave us AEO, and the next machine is the agent.
  • 02Simon reviews the names in circulation (AEO, GEO, LLMO, AAO and AX) and what each gets right and where it falls short.
  • 03The ladder: cited (the source behind the answer), recommended (the name the AI suggests) and chosen (the agent books you).
  • 04Five questions an agent asks about you: can I get in, read it, trust it, compare it and act on it?
  • 05The honest counterweight: Google says optimising for its AI features is still SEO, and AI referrals are still a small share of traffic.
  • 06Simon's suggestion: stop naming the machine and start naming the outcome.

Full transcript

Short answer: SEO still matters as the foundation, but the name describes a machine that is no longer making the decision. Search engines show you a list. Answer engines tell you an answer. AI agents decide and act for you. The job has moved from being found to being chosen, and the name should follow.

Watch the full video on YouTube.

The name always follows the machine

Search engines gave us search engine optimisation. Answer engines gave us answer engine optimisation. The next machine is the agent: you say "find me a reliable roofer and book me a quote for Tuesday" and there is no results page, no ten blue links, and one business gets the job.

The names already in circulation

NameWhat it gets rightWhere it falls short
AEO, answer engine optimisationFits AI search today"Answer" undersells an agent that acts
GEO, generative engine optimisationThe mechanics, with academic roots in 2023Names the tech, not the outcome
LLMO, AIO, AI SEOPlain descriptionStopgaps tied to today's tech
AAO, assistive agent optimisationJason Barnard, 2025, built for agentsClosest yet, still names the machine
AX, agent experienceDeveloper world: can an agent use you?About usability, not being chosen

Jason Barnard coined answer engine optimisation in 2017 and popularised it in 2018. His Kalicube framing, that search engines present choices, assistive engines recommend and agents decide, underpins the ladder below.

The ladder: cited, recommended, chosen

  • Cited: you are the source behind the answer. Needs exact answers, original knowledge nobody else has, and crawlers allowed in.
  • Recommended: you are the name the AI suggests. Needs other people saying it about you: reviews and mentions that agree.
  • Chosen: the agent books you, buys from you or signs you up. Needs real prices, published availability and booking without a phone call.

Being cited is not the same as being chosen. Pew Research found Google users clicked a result on 8% of visits when an AI summary appeared, against 15% without one, so ranking higher no longer guarantees the click.

Five questions an agent asks about you

  1. Can I get in? Since July 2025 Cloudflare blocks known AI crawlers by default for new domains, so check your firewall settings.
  2. Can I read it? One page that clearly says who you are and what you do.
  3. Can I trust it? Reviews and genuine mentions elsewhere that agree with you. A review that says "picked up on the A1 and home within the hour" tells a machine far more than "cheers, Gary".
  4. Can I compare it? Real prices and terms, not "contact us".
  5. Can I act on it? Book, buy or enquire without a human on the phone.

The honest counterweight

Google's own guide says optimising for its generative AI features "is optimizing for the search experience, and thus still SEO". AI referrals are still a small share of web traffic. Findable still matters: a guide we wrote in September for our client Stitch AI became the first source card in Google's AI Overview for its question. Findable is the floor, not the finish line.

So what should we call it?

Stop naming the machine and start naming the outcome. Recommendation optimisation? Agent optimisation? Or simply selection: are you selected? I'll keep using AEO for now, because the job underneath is what's changing.

Try this today

Ask ChatGPT, Claude, Perplexity or Copilot to book someone in your trade for next Tuesday. If it can't, or it picks someone else, you know which step of the ladder to work on.

Tell me I'm wrong: https://www.linkedin.com/in/aeo/

Full transcript

Full transcript coming shortly. It will be added here once the captions have been checked. In the meantime, watch the full video on YouTube.

This entry is part of a dated archive, what we said, when we said it. For what works in AI search today, start with AEO at Question.Marketing. For why the dates matter, see the receipts.

Where next?

The library holds every video we have published. The AEO pages explain the method behind them, what it costs and what the evidence actually supports.