Skip to content

Article 11 min read

Answer Engine Optimization: AEO Was Born Before AI, in the Answer Google Already Gave Without a Click

by Daniel Sócrates published

Type “answer engine optimization” into Google and notice the order of the screen. We ran the search on September 14, 2026, on Google in Brazil, where we are based. Three ads came first. Two of them sold GEO and AEO services. 1

Just below, before any organic link, came the AI Overview. It answered in English, on a results page set to Portuguese, with a sentence taken from HubSpot. AEO, it said, is “the practice of shaping your online content so that AI systems cite your brand when answering user questions.” 1

The answer machine explained, on its own, how to optimize for the answer machine. It sounds like a 2026 topic. The story starts a decade earlier, when Google began answering without asking for a click.

What answer engine optimization is

Think of the information desk at a bus terminal. You ask when the next bus to Philadelphia leaves. The attendant does not hand you a stack of brochures from every bus line. He reads the right line of the schedule out loud and tells you which company it came from.

A search engine that answers this way works like that desk. This is called a direct answer.

Answer engine optimization, or AEO, is the work of getting your page ready to be the line the attendant reads. The name shows up in a record from 2018. 2 The desk is older than the name.

Before AI, Google already answered without a click

2012, the knowledge panel. On May 16, 2012, Amit Singhal announced Google’s Knowledge Graph on the company blog. The title summed up the shift: “things, not strings.” 3 At launch, the catalog held more than 500 million objects. It also held more than 3.5 billion facts about them and the relationships between them. 3

The knowledge panel is the display case for that catalog. Google’s help describes it as an information box. It appears when you search for an entity that is in the Knowledge Graph. 4 The panel is generated automatically, with information from various sources across the web. 4 What to do when it shows the wrong data is on the page about the knowledge panel.

2014, the featured snippet. In January 2014, Google launched featured snippets. The date comes from a 2018 post by Danny Sullivan. He recalls the fear of the time: “there were some concerns that they might cause publishers to lose traffic.” 5

Google’s documentation describes special boxes where the format of a regular search result is reversed. The descriptive snippet comes first. 6 A snippet is the short summary that appears under the title of each result.

Related questions. A featured snippet can also appear inside a group of questions. Google calls that group “People Also Ask.” 6 Each question you open there hands you a slice of some site, already cut out.

Voice. Google’s help says featured snippets are “helpful when you use a phone or talk to your device.” 7 Sullivan described how the spoken version worked. Google cited the source page in the spoken answer itself. And it sent a link to the page in the Google Home app. 5

On a smart speaker with no screen, the list of ten results disappears. In 2018, Search Engine Watch put the problem this way: “the search engine needs to be able to provide a single answer, or none at all.” 2

Behind the box sits a design Google patented. US 9,940,367, with an August 2014 priority date, covers searches identified as questions. 8 The system takes the top-ranked pages for that question. It cuts out candidate passages and scores each one. The winner appears apart from the results, “as in an ‘answer box.’” 8

One detail in the text matters to anyone who writes. Under one of the criteria described, only text “subordinate to a particular heading” goes into a candidate passage. 8 The patent describes a mechanism. It does not prove that Google uses exactly this today.

Who coined the term AEO

The person who claims the term is Jason Barnard, founder of Kalicube. The company’s page on the term says it was “coined by Jason Barnard in 2017.” 9 Another page from the same company says “Back in 2018, Jason Barnard publicly coined Answer Engine Optimization (AEO).” 9

His personal site points to a milestone on September 15, 2017. That day, Trustpilot reportedly gave out a printed white paper at BrightonSEO, written with him, on voice and AEO. 10 As proof, the page points to an event recap published by Footprint Digital. That recap does not mention Barnard, the white paper, or the acronym. 10

His talk at BrightonSEO in April 2018 is published. The slides were uploaded on April 28, 2018, under the title “A Universal Strategy for Answer Engine Optimisation (beyond position 0).” 11

The oldest independent record I found is from February 7, 2018. Rebecca Sentance, of Search Engine Watch, reported on a webinar by Chee Lo, of Trustpilot, and Jason Barnard. 2 She writes that the strategy “has come to be known as AEO.” In other words, the name was already circulating. 2

The same article lays out AEO’s terrain in three pieces: featured snippets, the Knowledge Graph, and voice search. 2 None of them is generative AI.

So the record stands like this. Barnard’s claim has a 2018 slide deck and coverage from the time linking the name to him. The dates change from one of his pages to another. And I found no independent document that proves who used the term first.

What changed with AI Overviews and AI Mode

On May 14, 2024, Elizabeth Reid, Google’s VP of Search, announced AI Overviews for everyone in the United States. The post’s title was an invitation: “Let Google do the searching for you.” 12

On August 15, 2024, Google announced AI Overviews in six more countries, Brazil among them, with local language support. 13 AI Mode started rolling out to everyone in the U.S. on May 20, 2025. On September 8, 2025, it expanded to five more languages, Brazilian Portuguese among them. 14

The difference from the featured snippet lies in who does the writing. The featured snippet shows “a little piece of a website,” as it stands. 7 In an AI answer, the system retrieves several pages from the index. It reviews their information and writes the answer, with links to the sources. 15

For the full mechanism, including query fan-out, see the page about how to show up in AI answers.

The fight over the click came back with the same script. In 2018, Google wrote that “featured snippets do indeed drive traffic.” 5 In 2024, the AI Overviews announcement said something similar. It said links inside the answer got more clicks than the same page as a traditional listing. 12

An outside measurement came from the Pew Research Center. Athena Chapekis and Anna Lieb analyzed the browsing of 900 American adults in March 2025. 16 Users who saw an AI summary clicked a traditional result in 8% of visits. Users who did not see one clicked in 15% of visits. 16 A click on a link inside the summary happened in 1% of visits to pages with a summary. 16

Two caveats come with these numbers. It is a U.S. measurement, from a single month, with a summary in 18% of the searches in the study. 16 And Google and Pew measured different things. Google compared the same link in two formats. Pew compared pages that had a summary with pages that did not.

What still holds

The snippet is still the door. For Google’s AI features, a page must be “indexed and eligible to be shown in Google Search with a snippet.” 15 It is the same slice of text that feeds the featured snippet.

Even the control is the same. The documentation says to use the nosnippet rule to opt out of featured snippets. 6 Today, that rule also keeps the content from being a direct input for AI Overviews and AI Mode. 17 The max-snippet rule applies to web search, the Assistant, AI Overviews, and AI Mode. 17 It is one switch, from the 2014 box to the generated answer.

The answer that stands on its own. The featured snippet was always a small slice of a site, read away from its page. 7 In the 2014 patent, one of the criteria cuts the passage from under a heading. 8 A paragraph that answers on its own, right under a heading that asks the question, serves the old desk and the new one.

Google says nobody needs to chop a page into tiny pieces for AI. 15 The ask here is smaller. Each important answer sits in a paragraph that makes sense without its neighbor.

The clear entity. The knowledge panel appears for entities that are in Google’s catalog. 4 In my book SEO de Entidade, published in Portuguese, I describe two possible states. The system finds your company’s name and understands one specific thing, with attributes and relationships. Or it understands a string of letters that looks like other strings. 18

In a generated answer, the credit tends to go to whoever the machine knows. That bridge is my reading, and it is in the book. 18

Structured data, in the right measure. Here the market needs a correction. Google’s guide says “Structured data isn’t required for generative AI search, and there’s no special schema.org markup you need to add.” 15 It is still useful for rich results in Search. 15

The only markup Google built with voice in mind, speakable, is still in beta. It works only “for users in the U.S. that have Google Home devices set to English.” 19

AEO, GEO, and LLMO: where this page stops

Google itself defines the acronym in its July 2026 guide. “‘AEO’ stands for ‘answer engine optimization.’” The term describes work focused on visibility in AI search experiences. 15

Notice the shift. In 2018, AEO’s terrain was featured snippets, the Knowledge Graph, and voice. 2 In 2026, the owner of the search engine describes the same name through AI. 15 The guide closes by saying that, for Google Search, optimizing for generative AI is still SEO. 15 This is Google talking about itself.

The fight between the acronyms, with our measurement of the bare acronym “aeo” in Brazil, is on the page about LLMO, GEO, and AEO. The academic origin of GEO has its own page. So does the step-by-step for featured snippets.

The question to ask from today on

When someone offers you AEO, ask which desk the proposal starts at. Featured snippets, related questions, the knowledge panel, voice, or AI answers. Each one has its own rules, and almost all of them were born before generative AI.

Nobody guarantees the spot. Google writes that you cannot mark your own page as a featured snippet. Its systems do the choosing. 6

The first step takes an afternoon. Write down the five questions customers ask before they buy. Search for each one on Google and write the date next to it. Note what the screen shows before the links, and whose passage it is.

Then open your page on each question. Look for the paragraph that answers on its own, under a heading that asks the question. The rest of the work in this section is in SEO for AI.

Frequently asked questions

Is AEO different from SEO?

For Google Search, Google itself says no. The July 2026 guide says optimizing for generative AI search is still SEO. 15 A limit comes attached. This is the owner of the search engine talking about its own engine, and the sentence does not cover other systems.

Who coined the term answer engine optimization?

Jason Barnard claims it, with the dates 2017 and 2018 on his own company’s pages. 9 The oldest independent record I found is from February 2018. It links the term to a webinar Barnard gave with Trustpilot. 2 I found no proof of who used the term first.

No. Google’s documentation answers in two words: “You can’t.” Its systems decide whether a page makes a good featured snippet and, if it does, elevate it. 6

Do I need structured data to appear in Google’s AI answers?

No. The guide says structured data is not required for generative AI search. There is no special schema.org markup to add either. 15 It is still useful for rich results.

Google’s documentation still describes featured snippets, in a version updated in December 2025. 6 In our search for “answer engine optimization” on September 14, 2026, on Google in Brazil, the AI Overview came right after the ads. No featured snippet appeared. 1 What the screen shows changes from one search to the next, and from one day to the next.

Notes and sources

  1. SEO Genome’s own measurement, September 14, 2026, in two readings, in Brazil, where we are based. Semrush, Brazil database: “answer engine optimization” with 170 searches a month and keyword difficulty 57; the top ten results are about answer engine optimization; the report does not state the capture date. Google in Brazil, live, with Portuguese as the language, Brazil as the country, and the parameter that asks for non-personalized results, in a browser signed in to an account, from a single location: three ads at the top, two of them selling GEO and AEO services; then the AI Overview, labeled “Visão geral criada por IA” in the Portuguese interface, written in English and opening with HubSpot’s definition, “the practice of shaping your online content so that AI systems cite your brand when answering user questions”; then nine web results, all about answer engine optimization, with no featured snippet on the page. Internal record of the firm.
  2. Rebecca Sentance, “The rise of Answer Engine Optimization: Why voice search matters,” Search Engine Watch, February 7, 2018, https://searchenginewatch.com/2018/02/07/the-rise-of-answer-engine-optimization-why-voice-search-matters/. Accessed September 14, 2026. Reports on a webinar with “Chee Lo, Head of SEO at Trustpilot, and Jason Barnard, SEO consultant at Kalicube.pro.” Passages: “This strategy has come to be known as AEO, or answer engine optimization”; “Innovations like featured snippets and Knowledge Graph have contributed to Google’s aim of providing the answer to a search query without requiring a user to click through to another website”; and, on devices without screens, “the search engine needs to be able to provide a single answer, or none at all.”
  3. Amit Singhal, “Introducing the Knowledge Graph: things, not strings,” official Google blog, May 16, 2012, https://blog.google/products/search/introducing-knowledge-graph-things-not/. Accessed September 14, 2026. Passage: “It currently contains more than 500 million objects, as well as more than 3.5 billion facts about and relationships between these different objects.”
  4. Knowledge Panel Help, “About knowledge panels,” https://support.google.com/knowledgepanel/answer/9163198. Accessed September 14, 2026. Passages: “Knowledge panels are information boxes that appear on Google when you search for entities (people, places, organizations, things) that are in the Knowledge Graph” and “Knowledge panels are automatically generated, and information that appears in a knowledge panel comes from various sources across the web.”
  5. Danny Sullivan, “A reintroduction to Google’s featured snippets,” The Keyword (Google’s blog), January 30, 2018, https://blog.google/products/search/reintroduction-googles-featured-snippets/. Accessed September 14, 2026. Passages: “When we introduced featured snippets in January 2014, there were some concerns that they might cause publishers to lose traffic”; “It quickly became clear that featured snippets do indeed drive traffic”; “It’s especially helpful for those on mobile or searching by voice”; and “When it comes to spoken featured snippets, we cite the source page in the spoken result and provide a link to the page within the Google Home app”.
  6. Google Search Central, “Featured snippets and your website,” https://developers.google.com/search/docs/appearance/featured-snippets, page last updated December 10, 2025. Accessed September 14, 2026. Passages: “Featured snippets are special boxes where the format of a regular search result is reversed, showing the descriptive snippet first”; “They can also appear within a related questions group (also known as ‘People Also Ask’)”; the nosnippet rule to block both regular snippets and featured snippets; and “How can I mark my page as a featured snippet? You can’t. Google systems determine whether a page would make a good featured snippet for a user’s search request, and if so, elevates it.”
  7. Google Search Help, “How Google’s featured snippets work,” https://support.google.com/websearch/answer/9351707. Accessed September 14, 2026. Passages: “This box shows a little piece of a website that helps answer your question” and “They’re helpful when you use a phone or talk to your device.”
  8. Google, patent US 9,940,367 B1, “Scoring candidate answer passages,” priority date August 13, 2014, filed August 12, 2015, granted April 10, 2018, https://patents.google.com/patent/US9940367B1/en. Accessed September 14, 2026. Passages: “a question query that seeks an answer response”; the passage generator “only uses the top N ranked resources”; the passage shown “separate and distinct from the search results, e.g., as in an ‘answer box’”; and “only text that is subordinate to a particular heading may be included in a candidate answer passage”. A patent describes a mechanism and does not prove current use.
  9. Kalicube, “Answer Engine Optimization,” https://kalicube.com/entity/answer-engine-optimization/, with “coined by Jason Barnard in 2017” and, on the same page, “Created by Jason Barnard, 2018”; and Kalicube, “How Jason Barnard Defined Answer Engine Optimization Before Google AI Mode,” June 2, 2025, https://kalicube.com/learning-spaces/faq-list/generative-ai/googles-ai-mode-is-changing-how-search-works-and-what-visibility-really-means/, with “Back in 2018, Jason Barnard publicly coined Answer Engine Optimization (AEO)”. Accessed September 14, 2026. These are pages from the company of the person who claims the term.
  10. Jason Barnard, “The Birth of Answer Engine Optimization: Trustpilot’s BrightonSEO 2017 stand, then the 2018 public record,” page modified September 1, 2026, https://jasonbarnard.com/digital-marketing/conferences/offline/the-birth-of-answer-engine-optimization-jason-barnard-trustpilot-and-brighton-seos-defining-moment-in-digital-search/. Accessed September 14, 2026. Passage: “On 15 September 2017, Trustpilot exhibited at BrightonSEO and gave out a printed white paper written with Jason Barnard”. As proof, the page points to Footprint Digital’s recap of BrightonSEO in September 2017, dated September 18, 2017, https://www.footprintdigital.co.uk/library/brighton-seo-september-2017/, accessed September 14, 2026. That recap does not mention Jason Barnard, the white paper, or the acronym; Trustpilot appears only in a list of companies giving talks and exhibiting at the event.
  11. Jason Barnard, slides “A Universal Strategy for Answer Engine Optimisation (beyond position 0),” SlideShare, uploaded April 28, 2018, https://www.slideshare.net/slideshow/a-universal-strategy-for-answer-engine-optimisation-beyond-position-0/95315822. Accessed September 14, 2026.
  12. Elizabeth Reid, “Generative AI in Search: Let Google do the searching for you,” official Google blog, May 14, 2024, https://blog.google/products-and-platforms/products/search/generative-ai-google-search-may-2024/. Accessed September 14, 2026. Passages: “today, AI Overviews will begin rolling out to everyone in the U.S., with more countries coming soon” and “the links included in AI Overviews get more clicks than if the page had appeared as a traditional web listing for that query.”
  13. Hema Budaraju, “New ways to connect to the web with AI Overviews,” official Google blog, August 15, 2024, https://blog.google/products/search/new-ways-to-connect-to-the-web-with-ai-overviews/. Accessed September 14, 2026. Expansion of AI Overviews to the United Kingdom, India, Japan, Indonesia, Mexico, and Brazil, with local language support in each country. The original separates those two passages with a dash, so this page uses indirect speech.
  14. Hema Budaraju, “AI Mode in Google Search expands to more languages,” official Google blog, September 8, 2025, https://blog.google/products/search/ai-mode-expands-more-languages/. Accessed September 14, 2026. Passage: “we’re bringing AI Mode, our most powerful AI search experience, to five new languages for users around the globe: Hindi, Indonesian, Japanese, Korean, and Brazilian Portuguese.” The U.S. rollout comes from Elizabeth Reid, “AI Mode in Google Search: Updates from Google I/O 2025” (headline on the page: “AI in Search: Going beyond information to intelligence”), official Google blog, May 20, 2025, https://blog.google/products-and-platforms/products/search/google-search-ai-mode-update/, accessed September 14, 2026: “AI Mode will start to roll out for everyone in the U.S. today.”
  15. Google Search Central, “Optimizing your website for generative AI features on Google Search,” https://developers.google.com/search/docs/fundamentals/ai-optimization-guide, page last updated July 10, 2026. Accessed September 14, 2026. Drawn from it: retrieval-augmented generation, “also known as grounding,” which retrieves pages from the Search index and reviews their information; the condition of a page “indexed and eligible to be shown in Google Search with a snippet”; “There’s no requirement to break your content into tiny pieces”; “Structured data isn’t required for generative AI search, and there’s no special schema.org markup you need to add,” with the caveat that it helps with rich results; and the passage on the acronyms: “‘AEO’ stands for ‘answer engine optimization,’” a term used “to describe work specifically focused on improving visibility in AI search experiences,” and “From Google Search’s perspective, optimizing for generative AI search is optimizing for the search experience, and thus still SEO.”
  16. Athena Chapekis and Anna Lieb, “Google users are less likely to click on links when an AI summary appears in the results,” Pew Research Center, July 22, 2025, https://www.pewresearch.org/short-reads/2025/07/22/google-users-are-less-likely-to-click-on-links-when-an-ai-summary-appears-in-the-results/. Accessed September 14, 2026. Base: 900 U.S. adults, browsing data from March 2025. Passages: “Users who encountered an AI summary clicked on a traditional search result link in 8% of all visits. Those who did not encounter an AI summary clicked on a search result nearly twice as often (15% of visits)”; a click on a link in the summary in “just 1% of all visits to pages with such a summary”; and “Some 18% of all the Google searches in our study generated an AI summary”.
  17. Google Search Central, “Robots meta tag, data-nosnippet, and X-Robots-Tag specifications,” https://developers.google.com/search/docs/crawling-indexing/robots-meta-tag, page last updated March 24, 2026. Accessed September 14, 2026. The nosnippet rule “will also prevent the content from being used as a direct input for AI Overviews and AI Mode”; max-snippet applies to “Google web search, Google Images, Discover, Assistant, AI Overviews, AI Mode” and “will also limit how much of the content may be used as a direct input” for those two features.
  18. Daniel Sócrates, SEO de Entidade (published in Portuguese), 2026, chapter 3 (the two states a system ends up in when it finds a company’s name: one specific thing or a string of letters) and chapter 9 (the gate of attribution). The bridge between a clear entity and credit in the generated answer is the author’s reading.
  19. Google Search Central, “Speakable (BETA) Schema Markup,” https://developers.google.com/search/docs/appearance/structured-data/speakable, page last updated September 8, 2026. Accessed September 14, 2026. Passage: “The speakable property works for users in the U.S. that have Google Home devices set to English, and publishers that publish content in English.”

Want to see SEO Genome more often on Google? Add it as a preferred source.

Daniel Sócrates

Founder and CEO of SEO Genome

Author of AI Search Optimization. Writes about what the company applies in client work.

danielsocrates.com.br

Let us look at what the machine understands about your company

A diagnostic conversation, with no slide deck. You bring your website address. We bring what Google and the AI assistants already know about it today.

Talk to us on WhatsApp