The conversation most account managers are avoiding

At some point in the last year, a client probably asked you what you thought about AI search. Maybe they saw something about Google’s AI Overviews. Maybe a competitor told them they were doing “GEO”. Maybe they just noticed their traffic numbers shifting in a way that didn’t line up with their rankings.

Most account managers have one of two responses. They either brush it off (“it’s too early, nothing to worry about yet”) or they try to explain it and end up losing the client in a tangle of technical language about large language models and citation training.

Neither lands well.

This article gives you a clear, honest way to explain GEO to a client who only knows traditional SEO. Something you can use in a call, in a proposal or in a follow-up email without needing a PhD in machine learning to back it up.

What GEO actually is, in plain language

GEO stands for Generative Engine Optimisation. It means making sure your content appears in the answers that AI tools generate, not just in the list of ranked results below them.

That’s it. That’s the concept.

Where traditional SEO focuses on ranking your pages in Google’s blue link results, GEO focuses on being included when tools like Google’s AI Overviews, ChatGPT Search or Perplexity generate an answer to a question your potential customers are asking.

These are two different outcomes. A page can rank well in SEO terms and still be completely absent from AI-generated answers. A brand can appear prominently in AI answers while having a fairly modest organic ranking position. Both matter. They require different types of work.

The short version you can say to a client: “SEO gets you into the list of search results. GEO gets you into the answer that appears before anyone looks at that list.”

How search has changed since AI Overviews arrived

Google introduced AI Overviews in 2024. They appear at the top of results for a significant and growing portion of queries, particularly those involving research, comparison, advice and complex decisions. The AI generates a synthesised answer and pulls from sources it considers authoritative and well-structured.

ChatGPT has a search function. Perplexity has built its entire product around AI-generated answers with cited sources. Microsoft’s Bing has had AI-generated answers built in for over two years.

More importantly, the behaviour is catching on. People who discover they can ask a detailed question and get a useful synthesised answer in seconds are forming a habit. For certain query types, they’re skipping the list of blue links entirely and reading the generated response instead.

That creates a gap. If your client’s brand or content isn’t being cited or referenced in those responses, they’re invisible to a portion of their potential audience during the exact moment that person is deciding what to do or who to trust.

Worth flagging to clients: AI Overviews have reduced click-through rates on the organic results below them for many query types. Good SEO rankings are still valuable, but the traffic distribution has shifted for research-heavy searches.

The analogy that works every time

Clients tend to understand GEO immediately when you use this comparison.

SEO is like getting a good table in a restaurant guide. You’re listed. Your rating is visible. When someone browses the guide, they might find you. The better your SEO, the higher up the list you appear, and the more likely someone is to click.

GEO is like being the restaurant the waiter recommends when someone walks in and asks where to eat nearby. The waiter doesn’t show them a list. They just give an answer: “The Italian place two streets over is excellent.” That recommendation is worth far more than a listing three pages into the guide.

You can have both. You can have one without the other. They are related in some ways, but they are genuinely different outcomes that require different work.

Most clients immediately see why they’d want to be the recommended restaurant rather than just the well-listed one. That’s usually where the conversation opens up.

Why existing SEO work doesn’t fully cover it

This is the question clients ask next: “We already pay for SEO. Doesn’t that cover it?”

Partly. But not entirely, and the gap matters.

Good SEO work, including quality content, a clean technical foundation and a solid backlink profile, does carry some weight with AI models. AI tools use signals similar to search engines when assessing whether a source is trustworthy. So a well-maintained site with strong SEO foundations is a better starting point for GEO than a neglected one.

But AI models don’t rank pages in the same way a search algorithm does. They pull from content that clearly answers specific questions. They favour sources that define their subject matter precisely, use structured language and are cited or referenced elsewhere on the web. They respond to entity clarity, which means how well the web as a whole understands who your client is and what they do.

Here’s the table that usually helps clients see the distinction:

AreaTraditional SEOGEO
Primary goalRank in Google’s organic resultsAppear in AI-generated answers
Optimises forKeywords, backlinks, technical signalsClear answers, entity authority, structured content
Measured byRanking position, organic traffic, click-throughsAI visibility, branded search growth, citation tracking
Content approachKeyword-targeted pages and blog postsDirect answers to real questions, well-attributed sources
Brand signalsDomain authority, anchor text, page authorityEntity consistency, knowledge panel data, third-party citations
Timeline3 to 12 months typically3 to 9 months, varies by sector and existing authority
Works without good SEON/AHarder. Strong SEO foundations help GEO work faster.

The honest message for clients is that existing SEO investment is not wasted. It carries over. But the specific optimisation work for AI visibility is a separate job on top of what they’re already doing.

What GEO work actually involves

Keep this section concrete. Clients don’t need a technical deep dive. They need to understand what you’ll be doing and why it makes sense.

There are four main areas.

1. Structured content that answers real questions directly

AI models are much more likely to pull from content that asks a question and answers it clearly than from content written to rank for a keyword. This means reworking or creating pages and articles that directly address the questions your client’s audience is typing into AI tools. Short, clear answers. No padding around them.

2. Entity clarity

AI models build a picture of who a brand is from multiple sources: the website, the Google Business profile, Wikipedia if applicable, LinkedIn, Crunchbase, third-party mentions and structured data. If those sources tell a consistent story, the model has a clear “entity” to reference. If they’re inconsistent or thin, the brand is harder to cite with confidence. GEO work includes auditing and improving that entity footprint.

3. Source authority and third-party mentions

AI tools tend to cite sources that are referenced elsewhere. A brand that is mentioned in industry publications, partner sites, directories and reputable editorial sources has a stronger claim to appearing in AI answers than a brand that only talks about itself on its own website. This overlaps with traditional link building but with a slightly different lens on what counts.

4. Schema markup and structured data

Schema markup helps both search engines and AI models understand what a page is about at a structural level. FAQ schema, article schema, organisation schema and product schema all contribute to making content easier to parse and cite. Most sites with active SEO will have some schema in place already. GEO-focused work often extends this further.

A script for the client conversation

Here is a version of the explanation you can use almost verbatim in a call. Adapt the industry references to fit each client.

Client-ready script

You know how we’ve been working on your Google rankings? That work is still valuable and we’re going to keep doing it. But there’s something else happening in search that I want to flag.

When someone searches for something like “best accountant for small businesses in Manchester” or “which project management tool is right for a team of ten”, they’re increasingly getting an AI-generated answer at the top of the page. Not a list of websites. An actual synthesised response, sometimes with a few source links underneath it.

That’s Google’s AI Overview. ChatGPT and Perplexity do the same thing. And a growing number of people read that answer and don’t scroll down to the blue links at all.

So the question for us is: when someone asks a question your business should be answering, is your brand being mentioned in that AI response? Right now, for most businesses, the answer is no. Not because their SEO is bad, but because nobody’s done the specific work to make that happen yet.

That’s what GEO is. It’s making sure that when those AI answers are generated, your brand is in them. It works alongside your SEO, not instead of it. And the agencies that start doing this now are going to have a meaningful head start over the ones that wait.

If the client asks “so is this replacing SEO?”, the answer is straightforward: no. AI models use many of the same trust signals that search algorithms use. Strong SEO foundations make GEO work faster and more effectively. They are complementary, not competing.

Which clients should hear about this first

Not every client has the same urgency here. The ones most exposed to the shift in AI search behaviour tend to share a few characteristics.

Their customers research before buying. Long consideration cycles, comparison searches, “best X for Y” queries.
They operate in professional services, B2B, financial, legal, healthcare or education sectors.
Their target audience skews toward tech-comfortable decision makers who are likely to use AI tools.
They’ve noticed a drop in organic traffic or click-through rates despite stable or improving rankings.
A competitor has started talking about GEO or AI search in their marketing.
They have high-consideration eCommerce products where buyers read reviews and ask detailed questions before converting.

For clients who sell commodities, run local service businesses, or operate in sectors where people search with high purchase intent and low research need, the urgency is lower. GEO is still relevant but it’s not the first conversation to have.

Start with the clients where AI search is most likely already affecting their discovery. That’s where the case for action is clearest and the results are most measurable.

Common client objections and how to handle them

Three objections come up in almost every GEO conversation. Here’s how to respond to each without getting defensive or overcomplicating it.

We already do SEO. Isn’t that the same thing?
SEO gets you into the ranked results. GEO gets you into the AI-generated answer that appears above them. The work overlaps in places, but the specific optimisation for AI visibility is a different job. Think of it as extending what you’ve already built rather than starting over. Your existing SEO investment actually puts you ahead of brands that haven’t done that foundation work.
We can’t measure it. How do we know if it’s working?
GEO measurement is more manual than traditional rank tracking right now, but it’s not invisible. You can track how often your brand appears in AI Overviews for your target queries, watch branded search volume as a proxy for AI-driven awareness, and monitor direct traffic trends. The tools for tracking AI visibility are improving quickly. Setting honest expectations upfront, rather than promising a dashboard that looks like position tracking, is the right approach.
Our customers don’t use AI to search. They’re not that technical.
They don’t need to use a standalone AI tool. Google AI Overviews appear in standard Google searches and the majority of people using Google have already seen them, even if they don’t know the name for it. If your clients’ customers use Google, they are already encountering AI-generated answers. The question is whether your brand is in those answers or not.