The way people discover and evaluate businesses has already changed.
For years, organic search meant earning a prominent position in Google and persuading someone to click through to your website. That model still matters, but it now sits alongside another route to discovery: buyers asking ChatGPT, Gemini, Perplexity and Google’s AI-powered search experiences to research a problem, compare suppliers and recommend what to do next.
This is not a future trend.
At Growth London, we are already seeing the commercial effect. Organic traffic and inbound enquiries from traditional search have fallen as AI-mediated discovery has accelerated. We cannot attribute every lost visit to AI, but the direction of travel is clear: a growing share of the research that once happened on websites is now happening inside an AI-generated answer.
That changes the question marketing teams need to ask.
It is no longer enough to know:
“Do we rank on Google?”
You also need to understand:
“When an AI system answers the questions our buyers ask, does it mention us, cite us and describe us accurately?”
That is the practical purpose of Answer Engine Optimisation (AEO).
And for HubSpot users, AEO has moved quickly from an emerging marketing idea to something you can now measure and operate directly inside HubSpot.
Answer Engine Optimisation is the practice of improving how your organisation, expertise, products and content appear in AI-generated answers.
You may also hear terms such as Generative Engine Optimisation (GEO), AI search optimisation, AI visibility, LLM optimisation or AI SEO. The terminology is still settling.
Rather than becoming too concerned about the acronym, the useful question is whether the AI systems your buyers use can find sufficiently strong evidence to understand your organisation and include it in relevant answers.
No.
SEO remains one of the foundations of AEO.
Google is particularly explicit about this. Its guidance for AI Overviews and AI Mode says that the same fundamental SEO practices continue to matter and that there are no special technical requirements or special AI schema required to appear in its generative search experiences.
Good technical SEO still matters. Crawlability still matters. Useful content still matters. Internal linking still matters. Original expertise still matters. Authority still matters.
The difference is that marketers now have another layer of visibility to consider.
A page can perform well because someone clicks it from Google. It can also perform a different job by supplying evidence, facts or explanations that an answer engine uses when responding to a potential customer.
A modern content strategy needs to account for both.
There is no single mechanism used by every answer engine.
Different systems have different indexes, retrieval mechanisms, models and citation behaviours.
Google’s AI Overviews and AI Mode are closely connected to Google’s existing Search infrastructure. These experiences can use query fan-out, where a complex question results in multiple related searches across different subtopics.
ChatGPT can also retrieve current information from the web. OpenAI advises publishers that allowing OAI-SearchBot to access public content helps that content become discoverable and eligible to appear with citations in ChatGPT search.
Other systems work differently again.
Instead, create content that is accessible, indexable, factually clear, genuinely useful, well structured, supported by evidence, connected logically to the rest of your expertise and, where appropriate, corroborated by trustworthy external sources.
These are useful characteristics regardless of which retrieval system sits behind the answer.
A buyer may ask:
“What is the best CRM for a 50-person B2B SaaS business?”
Behind that question, an answer engine may need to investigate a much wider set of questions: which CRMs are designed for B2B, what works for a company of that size, implementation requirements, pricing, integrations, customer feedback, limitations and trade-offs.
Your content therefore needs enough depth to answer more than one isolated keyword.
This does not mean producing hundreds of tiny artificial content fragments. Clear, self-contained sections are useful because they improve understanding, not because an arbitrary word count supposedly unlocks AI visibility.
When we originally wrote this guide, much of the AEO process had to happen outside HubSpot.
That has changed.
HubSpot now includes native AEO functionality that can help teams measure AI visibility, track commercially relevant prompts, analyse citations, compare competitor share of voice and identify practical actions to improve visibility.
For HubSpot customers, this changes the operating model from:
Guess → publish → manually ask an AI assistant → guess again
to:
Measure → investigate → prioritise → improve → measure again
That is a much more useful way to treat AEO.
HubSpot can also use business context such as your ICPs and products to support prompt tracking and recommendations.
For teams already investing in HubSpot AI and agents, this is an important development: AEO is becoming part of the same operational environment as your content, CRM and AI deployment work.
Before changing your website, understand how your brand currently appears.
Set up HubSpot AEO with accurate information about your brand, common brand-name variations, your main competitors, your products or services and your Ideal Customer Profiles.
Then establish an initial set of prompts.
Do not judge performance from a single test. Generative answers vary. A useful baseline should help you answer:
This gives you something much more useful than a generic AEO score: a map of where the real gaps are.
AEO measurement is only useful if you track useful questions.
Avoid building a prompt list entirely around your own brand name.
A prospective customer who already asks “Is Growth London a good HubSpot partner?” is very different from somebody asking “Who can implement HubSpot for a UK professional services company?”
A healthy prompt set should include different stages of the buying journey:
That makes your prompt library much more useful than a list of vanity questions designed simply to make your brand appear.
Before assuming that another blog article is the answer, look at what answer engines are actually referencing for the topics you care about.
You may discover that they favour documentation, independent reviews, comparison articles, research, YouTube, Reddit, industry publications or detailed product pages.
That changes the content decision.
If third-party research dominates the citations, another generic article on your own website might not solve the visibility gap. You may need original research or external coverage instead.
If competitors’ product pages are repeatedly cited, your own equivalent pages may lack sufficient clarity or depth.
If Reddit appears frequently, the opportunity may be genuine participation in relevant discussions rather than another SEO landing page.
AI optimisation does not rescue a site that search engines cannot properly crawl and understand.
Check the basics first:
If visibility in ChatGPT search matters to you, also check that you are not inadvertently blocking OAI-SearchBot.
“Answer-first” content remains one of the most useful habits from the original version of this guide.
If a section asks “What is HubSpot Smart CRM?”, do not make the reader work through five introductory paragraphs before you answer it.
Give a clear answer first. Then add detail, examples, caveats, evidence and implementation guidance.
Think in terms of complete sections, not arbitrary AI chunks.
A useful section should normally make sense when someone lands directly on it.
Traditional keyword thinking can encourage very narrow pages.
AEO should push you towards stronger topical coverage.
If you are writing a guide to HubSpot onboarding, do not stop after defining onboarding. A useful resource might also cover data migration, integrations, user permissions, sales pipelines, marketing setup, reporting, training, common mistakes, internal ownership, timescales and what happens after onboarding.
Ask:
What additional questions would someone need answered before they could make a confident decision?
Then answer those questions well.
Structured data remains useful.
Depending on the page, you might use appropriate schema for your organisation, articles, products, services, people or breadcrumbs.
But we would now change one important recommendation from the original version of this guide.
FAQ schema is not a magic AEO lever.
Structured data should accurately describe the content users can already see on the page and should primarily be considered part of good search and content hygiene.
By all means create useful FAQ content. Just do it because the questions help your reader and complete the subject — not because adding FAQ markup supposedly guarantees an AI citation.
For practical HubSpot implementation guidance, see our guide: How to Add Schema to Your HubSpot Website (and Why It Matters for AEO).
AI assistants are extremely useful during evaluation.
People ask questions such as:
If these questions matter commercially, create content that genuinely helps somebody evaluate the choice.
A strong comparison should discuss who each option suits, important differences, trade-offs, implementation implications, cost considerations where appropriate, limitations and situations where you would recommend the alternative.
If ten websites already contain variations of the same generic advice, an eleventh rewritten version does not add much.
Original material gives both humans and retrieval systems a reason to use your source.
Examples include:
You cannot control AEO entirely from your CMS.
Answer engines regularly cite third-party sources.
That means digital PR, community participation, partnerships and customer advocacy increasingly sit alongside traditional SEO within the AEO strategy.
The important word is genuine.
Do not turn “Reddit gets cited” into a brief for your marketing team to manufacture promotional Reddit comments.
Instead, contribute expertise where you genuinely have something useful to add, earn customer reviews, pitch useful data and expertise to journalists, contribute to industry discussions, build relationships with relevant publications and publish material worth referencing.
Content maintenance still matters.
But avoid the simplistic rule that every article needs changing every few months because AI supposedly prefers a particular age of content.
Freshness depends on the subject.
A guide to current HubSpot pricing needs frequent review. A tutorial about a rapidly evolving AI product needs frequent review. A long-standing conceptual explainer may remain accurate for years.
Instead of refreshing content to change the date, review important content for factual accuracy, product changes, outdated screenshots, superseded statistics, broken links, new evidence, missing questions, changed terminology and changes in what your buyers actually ask.
The biggest mistake would be treating AEO as a one-off website project.
A much better operating model is:
Measure → diagnose → prioritise → act → re-measure
HubSpot’s AEO recommendations help formalise much of this process.
In some cases, those recommendations can also feed into HubSpot’s content creation tools. If you are using Content Agent or other credit-consuming AI features, it is worth understanding how that usage fits into your overall account-level credit pool. Our guide Sitting on HubSpot Credits? Here’s What to Do With Them is a useful starting point.
One useful tactic from the original guide is still worth keeping, but with more nuanced positioning: create a clear, factual page about your organisation.
This is not a magic file for LLMs. It is simply a well-structured, first-party source of truth that explains who you are, what you do, who you serve and how your services fit together.
A useful page might include:
For a practical example of how to approach this in HubSpot, see Build Your ‘AI Info Page’ in HubSpot — a blueprint for AI Engines.
There is another development worth highlighting for HubSpot teams.
HubSpot’s connector can expose AEO information to supported AI assistants, including ChatGPT and Claude.
That means you can work with your HubSpot AEO data conversationally.
For example, you might ask:
For teams already using ChatGPT alongside HubSpot, this matters.
AEO stops being another dashboard somebody needs to remember to inspect. Instead, the information can become part of your normal content planning, reporting and optimisation workflow.
AEO needs a different scorecard from traditional SEO.
Organic traffic and rankings still matter, but they no longer tell the whole story.
Useful metrics now include:
Keep AEO visibility metrics and actual business results connected.
It is entirely possible to increase the number of AI mentions without creating any meaningful commercial impact.
The objective is not to win an AEO dashboard. The objective is to become more discoverable and credible when the right buyers are researching a problem you can genuinely solve.
Write useful content for people. Clear structure helps retrieval, but machine-oriented prose normally makes the content worse.
There is no magic 100-word or 150-word chunk size. Explain the subject properly.
Structured data remains useful, but there is no special AEO schema that guarantees inclusion in AI-generated answers.
Treat new technical conventions on their evidence. Do not implement them simply because they are fashionable.
If third-party communities matter in your market, participate in them genuinely.
Generative answers vary. Track a consistent prompt set over time instead.
The technical and editorial foundations of good search optimisation remain highly relevant to AI-driven discovery.
Answer Engine Optimisation is not a future consideration. It is already part of how people discover, compare and evaluate businesses.
The practical challenge is not to chase every new AEO tactic. It is to make your expertise easier to find, understand and trust wherever buyers are asking questions.
That means creating things worth citing, explaining your expertise clearly, making important content easy to discover, answering the questions your buyers really ask, building credible evidence about your organisation on and beyond your own website, and measuring what answer engines actually do rather than assuming you know.
For HubSpot users, this is becoming much easier to operate.
HubSpot’s native AEO tooling now gives marketing teams a way to track prompts, monitor brand visibility, analyse citations, compare competitors and turn those observations into specific recommendations inside the same platform they already use to manage content, CRM and marketing.
At Growth London, we help HubSpot teams turn that into a practical programme: establishing the AEO baseline, deciding which questions matter, identifying visibility and citation gaps, improving the underlying content and technical setup, and measuring what changes.
If you want to understand how your brand currently appears in AI search — and what you can realistically do about it inside HubSpot — get in touch.