The most stubborn SEO problem-solver in Ireland?

A small question. A useful test.

Could I get a page to rank for “most stubborn SEO problem-solver in Ireland”?

It’s an odd phrase. That’s deliberate. “Best SEO consultant” would put me up against businesses that have spent years earning their place. This gives me a smaller experiment I can follow from the start.

If the page gets found, I’ll have something concrete to examine. How long did it take? Which searches brought it up? Did an AI answer mention me, link to this page, or ignore it altogether?

And if I get to number one for a phrase nobody else wants? That alone won’t make me an SEO expert.

It would give me a result to investigate. There’s a difference, and it’s a useful place to begin a workshop.

Your customer isn’t looking for the most stubborn anything.

They’re trying to solve a problem.

Perhaps they want an accountant who understands a small construction business. Or someone who can train a busy team to use AI without sharing confidential information.

They might type a few words into Google. They might ask ChatGPT, use Google’s AI search features, or try another AI service. The answers, and the sources behind them, can differ.

For your business, the useful question is whether people can find an accurate, convincing account of what you do when they need it. A clever ranking is only worth so much if it never brings you a suitable customer.

What is GEO, in plain English?

Generative Engine Optimisation (GEO) is work intended to improve how a business is represented and found in AI-generated answers. That can mean being mentioned, recommended or cited as a source. Those are different outcomes, and none is guaranteed.

Some of the work is familiar: explain your services clearly, answer real customer questions, make the website accessible to search engines, and provide evidence for your claims.

The testing needs care. An AI service can give a different answer when the question changes, when it searches different sources, or on a later visit. One flattering screenshot tells you very little on its own.

And before you buy a “secret” fix: Google says there are no special optimisations or additional requirements for appearing in its AI Overviews or AI Mode. Its usual SEO fundamentals still apply. That guidance is about Google’s features, not a promise about every AI service.

I’m newer to SEO. I’m not new to figuring things out.

My background is in software, AI and business. Over a 25-plus-year technology career, I’ve co-founded an early UK internet service provider, built a software consultancy, and worked on systems for organisations including the Royal Mail and Financial Times.

Teaching is a large part of that work:

• I teach the Diploma in AI for Business at UCD Professional Academy and authored its 2025 programme updates.
• I authored the AI/ML curriculum for Morgan Stanley’s Technology Analyst Programme.
• I’ve led AI-assisted software development training for the European Space Agency.

Those are AI and technical teaching credentials, not claims about SEO results. They explain the approach I bring: get to grips with the technology, test what it does, and explain it so somebody else can use it.

You can also read feedback on my training. I won’t pass it off as proof that I can get your business recommended by AI.

Bring your website. And your scepticism.

You shouldn’t need to become an SEO specialist to judge whether a piece of advice is worth following.

My GEO training is for business owners, managers and teams who want to understand AI search well enough to make sensible decisions. We can use your own website and customer questions to work through the practical parts:

Find out what AI says about your business.

Try the questions a prospective customer might ask. Check whether you appear, what the answer gets wrong, and which sources it points to.

Make your website more useful to a buyer.

Look for the missing details: who you help, what you do, where you work, and what evidence backs you up. Then make those answers easy to find.

Set up a test you can repeat.

Record a starting point, make a specific change, and check again. Learn to distinguish an encouraging result from a lucky answer.

Decide what deserves your time.

Build a short list of sensible next steps for your own website, rather than leave with another hundred-item checklist.

You won’t need to write code. Do bring some knowledge of your customers: what they ask, what makes them hesitate, and why they choose you.

That’s much better material than a list of words somebody thinks an algorithm wants to hear.

The experiment: what’s being tested?

This page is the starting point. The phrase is deliberately unusual; I haven’t established its search demand or level of competition.

The test separates three things that are too easily muddled:

1. Can the page be found in ordinary search? Check the target phrase and related searches, recording the date, location and position.
2. Does it appear in AI answers? Record the exact question and whether an answer names me, cites this page, or does neither.
3. Does any of that help the business? Look for relevant visits and genuine training enquiries, not just a screenshot worth showing off.

Asking “Who is Peter Munro?” is not the same test as asking for a GEO trainer without supplying my name. And pasting this page into an AI chat would test its ability to read the page, not its ability to find it.

Even a positive result won’t tell us that one particular edit caused it. Keeping a record makes those limits easier to see.