How to sell GEO services to clients without overpromising

Why are clients asking about GEO now?

Because their buyers are asking AI engines for advice. A homeowner asks ChatGPT which window firms are good. A finance director asks Gemini which payroll software suits a firm of fifty. Clients have seen this, or tried it, and want to know where they stand.

Many have also been sold confusing ideas. Some think SEO is dead. Others think GEO is a trick that gets a brand into ChatGPT. An agency or consultant who explains it plainly and honestly has a real advantage.

This guide is for agencies and consultants adding GEO to what they offer. It covers explaining it, proving it, reporting on it and pricing it.

How do you explain GEO to a client?

Keep it to one or two sentences. Generative engine optimisation is the work of making sure a business is found, described correctly and recommended when buyers ask AI engines for advice. The what is GEO guide has a fuller version you can share.

Then connect it to what they already know. Much of GEO is good SEO. Google's guide to generative AI features says its AI features are rooted in its core ranking and quality systems. ChatGPT search and Perplexity find pages through web search. So clear pages, a sound site and a good reputation still do most of the work.

What changes is the measurement and some of the focus. There are no ten blue links to rank in. Instead, you track whether the brand is named, how prominently, how it is described and which sources the engines rely on. Accuracy matters more, because engines can repeat wrong facts to many buyers.

What is the best way to open a GEO conversation?

Show, do not tell. Run the client's own buyer questions through the main engines and put the answers in front of them.

A short audit works well as a paid first step or a pitch piece. Take 20 to 30 real questions, such as "best kitchen fitter in Leeds". Run them in ChatGPT, Perplexity, Gemini, Claude and Google. Record who is named, in what order, what is said and which sources are cited.

The results nearly always start the right conversation. A client who sees a competitor named in most answers, or reads a wrong price for their own service, does not need convincing that it matters. The AI visibility audit guide sets out the steps.

Which clients are a good fit for GEO?

Not every client needs it yet. The best fits are businesses where buyers research before they choose, and where a recommendation carries weight.

Home improvement firms, clinics, training providers, professional services, B2B software and travel businesses are good examples. Their buyers ask detailed questions and compare options. Multi-location brands are also a strong fit, because AI engines often get branch details wrong.

A business that sells mostly on impulse, or through a marketplace it does not control, may gain less. It is fine to say so. Turning down work that will not pay off is part of setting honest expectations.

What should you promise, and what should you not?

Promise the work and the measurement. Never promise a mention.

Google's page on whether you need an SEO says no one can guarantee a top ranking in Google. The same is truer still of AI answers. They vary from run to run, differ between engines, and change when models are updated. Any agency promising "guaranteed ChatGPT placement" is promising something it cannot control.

Honest expectations sound like this.

• We will track your visibility across these engines, on these questions, every week.
• We will find and fix wrong facts about your business.
• We will improve the pages and sources engines rely on in your topic.
• We expect to see trends over three to six months, not in the first week.
• Some engines and some questions will move faster than others.

Clients who hear this tend to stay longer. Clients sold a guarantee tend to leave when it fails.

What should a GEO report show?

Trends, causes and actions. A single answer from a single day proves nothing.

A good monthly report covers a handful of things.

1. Visibility over time. Mention rate and share of voice per engine, across repeated runs.
2. Competitors. Who is named most, and whether that is changing.
3. Sources. Which domains and pages the engines cite, including the client's own.
4. Accuracy. Wrong or out of date facts found, fixed and still open.
5. Actions. What was done this month and whether earlier actions worked.
6. Traffic. AI referral visits from GA4, noting that they undercount AI's real effect.

Explain the method once, in plain words, at the front of every report. Remind readers that tracking is a sample. That protects you when one engine has a strange month.

How should you price a GEO retainer?

Build the price from your real costs, then add a fair margin. Avoid inventing a flat price and hoping it covers everything.

There are three parts to the cost.

Tool cost. Most AI visibility tools charge per prompt, per engine, per run, or bundle that into plans. Work out the cost of the client's prompt set, on the engines you agreed, at the frequency you agreed.

Hours. Analysis, content, technical fixes, outreach and reporting. Be realistic. Fixing wrong facts across directories and pages takes longer than people expect.

Margin. Your normal margin on similar work.

Then set tiers that scale with the things that drive cost. Those are the number of prompts, the number of engines, the number of markets or locations, and the amount of content or fix work included. A single-location client with 40 prompts should pay less than a multi-location brand tracked across ten towns.

Should GEO be a separate service or part of SEO?

For most agencies, part of SEO works better. The work overlaps too much to split cleanly. Structured data, site health, content quality and reputation help both.

A common setup is an AI visibility add-on to an existing SEO retainer. It covers tracking, accuracy checks and a share of content aimed at gaps in AI answers. Clients who only want GEO can buy a stand-alone version, but it should still include the SEO basics, or it will not work.

Be careful with positioning. Calling yourself an AI SEO agency is fine if it describes what you do. Suggesting that SEO no longer matters is not honest, and clients who read Google's own guidance will know it. The GEO vs SEO guide sets out what changes and what stays the same.

How do you show results without overclaiming?

Link each action to a measure, record a starting point, and check it later. That is the fairest way to show value.

If you fix a wrong price that three engines were repeating, check whether later answers stop saying it. If you improve a page, check whether it starts being cited. If you get a client listed on a directory the engines trust, watch whether mentions rise on the related prompts. The GEO research paper that named the field found that visibility in generative engines can be changed by what is published. Your reports should show whether it was, for this client.

Where nothing moved, say so and say what you will try next. Clients trust agencies that report honest misses far more than those whose reports are always good news.

How Axiom GEO helps

Axiom GEO is built for agencies and SEO consultants, with multi-client workspaces isolated at the database level and staff and client logins. Client reports are white label monthly PDFs with a weekly digest, manual recommendations and automated claim checks on the numbers before a report is sent. Accuracy findings can be hidden from client users until your staff turn them on. Work management carries tasks with evidence and client sign-off, and records a baseline so it can later measure whether a fix worked. Every action has a published credit price, so you can cost a client's prompt set before you quote.