From unknown business to recommended? How GEO gets you there

Published

Contents

Action points

  • Pick one specific question your customers really ask.
  • Gather your proof from client examples, lessons and your own data.
  • Create one strong page that answers the whole decision.
  • Ask for a review after every job and measure the same questions every month.
Reading type0 min

Prefer to outsource?

Outsource GEO →

Big brands always seem to win in AI. Yet as a small business you can win that race, with specific questions and your own proof.

The empty restaurant problem of small businesses

Take a look at this.

You are facing an authority problem: how can smaller businesses build authority in GEO without first being a well-known name? Because if you are not a well-known name yet, how on earth do you get the robot to show you? And if you look at what GEO actually shows, you very often see big, well-known brands.

So how do you get shown? If you get shown as a smaller business, it helps you build your brand. At the same time, it is harder to build that brand without being shown everywhere.

So you are stuck with the empty restaurant problem: nobody walks in until someone is already sitting there.

Owner of a new café in Brabant in her empty restaurant, with a queue at a busy chain across the street

Big brands start out ahead

Big brands start this race with a head start. They already have more recognition and more searches for their brand name. If you look at Google Trends, you can compare yours with theirs.

They have more content, more backlinks, more mentions, more PR. They have bigger marketing budgets and spend more on advertising. If you spend a lot more on ads, more people simply see your brand and that helps it grow, assuming you have a good product or a good service.

So it is really hard for a small business to match that.

This limited recognition and authority makes it feel impossible for a small brand to win. Often it is not. It is an obstacle. One you can actually overcome.

There are strategies for this.

And when we get into those strategies, the strategy is definitely not to copy what a big brand does. You will lose that race outright. Especially if you are a very small start-up with little to no budget.

You can still get results. You simply cannot use the same strategy as these big brands. You need to take a different approach. And this different approach works very well for smaller brands.

From 20 mentions, things take off

We see this in AI too, for brands of all sizes.

How many earned media mentions does it take before AI recommends a brand?
Up to 20 mentions a month, you are taxiing down the runway. After that, the chance of a recommendation climbs fast, up to cruising altitude.
Chance that AI recommends a brand, by number of earned media mentions per month
PhaseMentions per monthChance of a recommendation
Low visibility0–4low
Emerging presence5–20starts to rise
Tipping point21–36rises steeply
Strong recommendation range37–58high
Dominant presence59+highest, levels off
Source: MEDO MEDIA / zoekmachinespecialist.com, calibrated to the Dutch market. Sample size: 840. Data from 26 business-to-consumer (B2C) sectors.

It starts to get interesting once you get more than 20 mentions every month. Above, you can see that recommendations then start to shoot up. The more websites mention yours – and the right type of website is where this actually gets a bit pickier – the bigger the chance that a language model presents you.

Not if you get 1, 2 or 3 websites.

At some point you need quantity.

Not hundreds either. This varies by industry - in the end, you can see where we are going with this.

Specific questions change the maths

What we also see is that specific questions change the maths.

When you get more specific with the questions people type in and focus on those specific questions, you have a much better chance of doing well.

You might say: sorry Dominic, those specific questions do not have the search volume.

I say that is fine.

When you work in a very specific, intent-driven way, they may not have the volume. They do convert very well and lead to more revenue.

More revenue also means more buying power for marketing, to expand and take on bigger players.

Together, SEO and GEO are a very profitable game that lets you scale up pretty well.

What we have found is that confidence rises as questions get more specific. The more specific the question someone types in, the more likely it is that smaller brands do more than act confident about their ability to appear. They actually do appear.

I use this for a lot of my clients.

Piano tuner from Brabant at work at an open piano in her own workshop

It works very well. We can still win and go after the bigger searches that the bigger brands focus on.

It just takes longer. Look, as a small business, can you really afford to invest in marketing for a year or more and only then see results? No.

That is unrealistic.

You need a strategy that focuses on very specific questions. You will have to go all in on that.

Google AI Mode names Moore DRV Breda among accountants for SMEs, with its own Breda page as the first source
Source: Google AI Mode, 11 November 2026.

Take Moore DRV in Breda. Ask Google's AI Mode for an accountant for SMEs in Breda. Moore DRV appears among the big full-service firms, in second place. Its own page about Breda is the first source next to the answer. That page answers exactly this question.

And it is not as simple as: we will just go after very specific questions.

You can expand from there and target the branches that grow out of them. There are actually quite a few other strategies too. We will dive into those shortly.

You do need to work methodically and deliberately.

If you just go too broad like the bigger brands and hope to show up everywhere, you will find you burn a lot of money. It then takes a long time before you see enough results to actually generate an ROI.

Visibility is not the same as revenue

People talk about visibility and say: "Look at me, I show up for some of these searches."

Showing up and generating an ROI – actually bringing in revenue and sales – are two different things.

The strategy I am breaking down here is not just about showing up. Yes, you do show up. What it is really about is turning that visibility into revenue.

If the revenue part is missing, it does not matter.

I do not care if someone says: "Look at this report from Profound and look at our visibility score, it has gone through the roof."

Well, if it does not bring in more revenue, who cares? At the end of the day you are spending money and not getting more revenue back for it.

To me, that really is a failed marketing campaign.

I will show you how to get the visibility that drives revenue for a small business. That needs to be your goal: revenue, in a profitable way.

And you are better off overspecialising than overpublishing.

Overspecialise instead of overpublishing

This is the nuance a lot of SMEs miss.

They try to publish too much.

You do not need to scale up your content and publish 50 articles a week or use loads of AI.

There is something called SynthID that Google has made open source. Google uses it. Claude uses it. OpenAI uses it.

Most language models use it or variations of it. Microsoft has not announced what it uses. Grok has not specifically announced what it uses either, yet the big, popular models do use SynthID.

It is Google's way of putting a watermark on images, videos and text.

Icon of a text page with a hidden watermark under a magnifying glass

Anthropic, the company behind Claude, has released a statement. It is on TechCrunch somewhere. If you search for 'watermarked content Claude' or 'Anthropic TechCrunch', it should come up.

In it, they state that changing a few paragraphs does not remove the watermark. They still know it was generated by AI.

You need to rewrite pretty much the whole piece.

AI content can still have value.

If you look at the traffic statistics that have been published (Lily Ray has also published statistics on this, as have many other SEOs and marketers), you will see that most AI content brings in very short-lived traffic.

After that it really dies off, from both a search traffic and a GEO perspective.

They tend to cite little of what they create themselves. When AI cites AI, there is a bigger chance of hallucinations.

They want to keep their systems from hallucinating by learning from the AI content they create themselves and then generating new AI content from that.

That just opens up a whole can of worms.

How to find the questions that matter

As you just saw, specific questions can really change the maths for smaller brands.

So how do you find the right questions?

As I said, they are the questions that matter for both customers and revenue. In the end, that starts with one specific customer.

As I always say: traditional SEO practices still work (keywords and that sort of thing).

Do start at the beginning: think about the person you are trying to reach.

Who is asking the question? What situation are they in? What are they trying to decide? What would make an answer useful? And in the end: what would they ask next?

If you can answer all these questions, you can find a question worth claiming.

In that respect, you need to take the details into account.

Which customer questions do bigger competitors answer poorly?
Every hole in the cheese is a question big competitors answer poorly. The bigger the hole, the more room for you.
How well bigger competitors answer customer questions, as a percentage of answers
Customer questionVery wellSomewhat wellNeither well nor poorlySomewhat poorlyVery poorlyUnsurePoorly answered
Prices, total costs and trade-offs8%21%16%31%19%5%50%
Specialised uses and product limitations9%19%22%29%17%4%46%
Implementation steps, timelines and requirements11%24%18%28%15%4%43%
Local or regional considerations9%21%22%28%12%8%40%
Common customer mistakes and how to avoid them12%23%24%26%9%6%35%
Specific sector needs14%32%19%21%8%6%29%
Source: MEDO MEDIA / zoekmachinespecialist.com — November 2025, calibrated to the Dutch market. Data from B2B survey respondents.

It is no secret that big brands cover the broad topics very well. As you can see from the weaker spots in this chart, there is plenty of opportunity for smaller brands to win.

These are questions that usually take more hands-on experience to answer. Something a specialist like you has and a big brand really cannot imitate.

Every weak spot you see here is an opening your business could claim.

And when it comes to these details, we need to remember specificity.

Take this ladder.

Climb down the ladder of specificity
With every rung down, the question gets more specific. The bottom question names an audience, a constraint and a decision.
Broad “How does GEO work?”
More specific “How should a small business improve its visibility in AI searches?”
More useful “Which GEO activities should a local service business with no research team and a limited content budget prioritise?”
audience constraint decision

Anyone can ask the first question: how does GEO work? Simple, easy. You will find plenty of results for it.

You could consider getting a bit more specific with an engaged audience. For example: 'How does a small business improve its AI search visibility?'

In the end, though, you want to reach this last rung, where it is most useful. This question names an audience (in this case a local service business), a constraint (no research team or a limited budget) and a decision about what should get priority.

The more specific it gets, the fewer competitors can answer it well.

This is your time to shine.

The good news is that AI also favours these specific questions, especially when you can give a very specific answer. This is something a brand of any size can publish.

You just need to make sure you back it up with outside validation and your own data. By doing this, you give AI a real reason to prefer your pages over a bigger name.

Three tests for every question

Before you just invest in a question, though, you need to make sure it passes these three tests. Very simple.

First: do real prospects ask this question? You should know this. Ask your sales team. I will go a bit deeper shortly into how you can find out what prospects ask.

Second: you need to know whether answering it steers the right customer towards a decision. This is all about moving the customer through the pipeline, to get them to make a purchase or take a next step.

Third: do you have the genuine experience to back it up? Having that expertise and hands-on experience is what makes you the specialist and sets your content apart from a bigger brand.

Does the question pass all three? Then put it on your shortlist.

Does one fail? Then maybe pause it for now and see what you can do, because that perfect question is out there somewhere.

The questions are already within reach

Finding these questions is really simple.

You do not need a huge research budget to find them. They are essentially already within reach.

They are in discovery calls, in sales objections, support tickets and even in comments on social media.

Icon of an open hand with question bubbles and a support ticket

Just ask your sales team: what are you tired of explaining? Is it a question about how to use something? Is it a question about how the buying process works?

They are a treasure trove of information for those specific questions you want to answer.

Then look at which questions also show up in searches. The questions that pop up everywhere are your clear winners.

Once you have chosen your question, the next challenge is proving that you are credible.

Turn your experience into proof

That is exactly the question: how do I prove that I am credible?

You might be thinking right now: I am a small brand. I am not some global mega-company with thousands of pieces of content on the internet. How do I show AI that I am actually credible?

It is a lot easier than you think.

What you want to do here is turn your experience into proof that you publish. What do I mean by that?

Look, anyone can give generic answers. Anyone can write a blog about 'choose the approach that best fits your business needs'. That is everyone's advice.

What if you publish the answer to a question that says: "For organisations with multiple locations: why not decide first whether each location needs separate reports?"

That choice then shapes your setup, your permissions and how you compare results.

Suddenly the reader has an answer. It is not some generic piece. It is based on your own business and expertise.

The reality is that most teams are sitting on a mountain of proof.

They have information that could make great content, including great answers to these questions.

Think of customer questions, objections or recurring feedback you get every day. Practical lessons your team has learned while delivering projects, or client examples.

With permission, of course – I am going to say that a few more times as a reminder. If you use client examples, ask for permission.

Those examples make great proof of your knowledge.

You can interview the experts who work for you and built the products. You can host webinars with people who live in the world of franchise content marketing every day and discuss how a franchise stands out.

You can make use of all these sources.

So the problem is not a lack of expertise. It is how you turn the things you already know into content.

And that is easy to fix.

I recommend showing the work behind every claim.

Show what you observed. Under what circumstances? Why do you think this happens? What could a client in a similar situation learn from this?

And also be open about what your claim cannot prove. Just because something worked in one situation does not mean it always works.

It is absolutely fine to admit that in your content.

You already have the information in-house to write this content, because you do this work every day.

Upholsterer from Brabant leafing through a folder of photos of his own projects

Experience beats fame

Hands-on experience outweighs brand recognition.

Remember how I said earlier that brand recognition is hard and takes a long time? First-hand experience wins every single time.

In study after study, we have found that hands-on experience, original data, detailed case studies and independent coverage by industry experts outperform overall brand recognition in AI and GEO visibility.

Building recognition takes years. Documented proof can start moving the needle for your visibility this month.

The best part is that original research can start small.

I am not saying you need to run huge, well-funded studies. If you have the budget: go for it. Great.

No budget for that? You can still start right away.

Instead, analyse a well-defined set of your own projects. Document patterns you keep seeing. Explain an unusual decision that you or your clients made in depth.

Moore DRV projects page with client projects from the IT consultancy team
Source: moore-drv.nl, 11 November 2026.

Moore DRV already does this. On its projects page, the IT consultancy team shows which business owners it helped, such as Vyva Fabrics and Verfwebwinkel. For each project you can read what the team helped with. That is proof from its own work. No big brand can copy that.

Dig into the things you see every day, because that is exactly what original research and original data are.

And that delivers results faster than an arbitrary yardstick like brand recognition.

Make no mistake: brand recognition is important. You just cannot wait until your brand is as big as your biggest competitor.

You need to start now.

Answer the whole decision

As I said: you have the proof. It is easy to get hold of.

What are you going to do with it now?

You need to make sure you know your questions and that the proof you have is in a format that customers and search systems can actually use.

Make sure you answer the whole decision.

Answering a simple question with yes, no or a short snippet is simple. Important as that is, you also want to make sure you explain why it matters.

When does it change? What are the trade-offs? Is there a real example?

Google AI Mode on the monthly price of an accountant in Breda, with sources that state a price
Source: Google AI Mode, 11 November 2026.
The Moore DRV Pro page ends at a quote request, without a price indication
Source: moore-drv.nl, 11 November 2026.

Ask AI Mode what an accountant in Breda with a dedicated point of contact and online bookkeeping costs per month. Moore DRV is not in the answer. The sources are sites that state a monthly price. Moore DRV offers exactly this service: Pro, with a dedicated adviser and a dashboard. On that page, the route ends at a quote. Add a price indication, or explain what determines the price. Then AI has something to mention.

There is a common thread that leads to the next question, because you are trying to guide prospects here.

For example, if buyers compare lead times, consider publishing your real process. Be honest and transparent about it, think about what affects the timing and perhaps share an example of a project that ran late.

This kind of honesty really builds trust.

Kitchen maker from Brabant pointing at a planning board in her showroom

Own it, explain why something took a long time and what improvements are being made. That makes you stand out, because you answer the whole decision instead of just that one question.

These kinds of comparisons also help you gain AI visibility.

Formats that a generic writer or big brand cannot imitate win the most. Honest comparisons and your own data make you stand out.

Consider building your priority pages around these formats, or FAQ and how-to pages, which you know AI likes to see.

Make sure your pages are reachable

Still, even the best-optimised pages achieve nothing if they are hard to reach.

I cannot stress enough how important it is that your priority pages are ready to be visited.

Fortunately, there is no secret technical trick to it.

Google says that AI features use the same search eligibility criteria as everything else that gets indexed.

So check that your pages are crawlable and linked from related pages, that they clearly answer their question, are written in real page text and, where possible, are attributed to a real expert.

Icon of a web page with an open door that a search robot walks through

We are not trying to game the system here.

You do not need a tiny page for every search variation.

If you create location pages and only change the place name each time, people and systems will see through it. It is simply not worth the effort.

One strong page beats 10 empty pages

Instead, create one strong page per real decision and make sure it has enough depth to help someone choose.

One page that helps someone choose beats 10 pages that say nothing.

Owner of a record shop in Brabant holding up one record as a recommendation

Those 10 empty pages can actually hurt your SEO as well.

You are not just optimising for GEO. You are still optimising to get Google traffic (whether that is Gemini, traditional search or search with AI Overviews).

If you have a lot of junk pages, that hurts your rankings in Google.

This is why many people who score well in GEO are now complaining that their SEO rankings are collapsing. They create loads of junk pages because the algorithms of ChatGPT and other players are not as advanced as Google's.

And then they find their SEO is going downhill.

They do get visibility, yet they see no revenue growth, simply because they are losing revenue through another channel.

You have robbed Peter to pay Paul.

You need to optimise for both.

Build your own area of expertise

In the end, that one strong page is a great starting point for building a whole area of expertise.

So you are not just putting out junk. You are building on a strong question that prospects want an answer to.

How do you grow from that one narrow answer into real topical authority? That takes three steps: choosing a territory, building a cluster and staying consistent.

Choose a territory

Choose a territory you can claim. For example, focus on one customer type, an industry, a recurring problem or a type of decision.

Think of inventory and bookkeeping for a restaurant group with three locations. That is narrow enough to cover in depth, yet big enough to drive real buying decisions.

This is the niche you can own.

Build a cluster per topic

For each priority topic, you then build a cluster.

Think of an overview, a decision guide, extensive FAQs, a case study and an expert's view.

Make sure you link them to each other, so readers move from understanding the problem to choosing a solution.

Icon of five connected pages that together form a path towards a flag

You are building a path that leads them to the right solution. Not a random pile of unconnected pages.

Stay consistent everywhere

Finally: consistency.

The machine-validated authority scorecard
Each leg is one authority signal. The longer the leg, the higher the score. With one short leg, the stool wobbles.
Score per authority signal
SignalScore
Expertise77%
Validation56%
Evidence40%
Consistency (weakest)38%
Source: MEDO MEDIA / zoekmachinespecialist.com — October 2025, calibrated to the Dutch market. Data from a survey of 150 marketers.

In this chart you can see four authority signals. Consistency is the weakest. It does matter. And it is almost entirely within your own control.

It needs to be loud and clear on your site.

What I mean by this is that on every single page you say in the same way who you serve, what you solve and who your experts are.

This consistency is something you can claim very quickly.

It is not about needless repetition. Consistency is the key for GEO.

Of course, that expertise is of limited use if it only stays on your website. You need to make sure it gets out there.

I recommend choosing one or two external channels your buyers actually use and showing up there on a schedule.

Again: one or two, not 10.

Consistency beats volume.

That is a nice bridge to the next topic.

Getting visible on a small PR budget

I know what you are thinking.

You are saying: I do not have a huge PR budget. How can I become visible outside my own website? I just do not have the budget for it.

How do you earn validation outside your own website without that budget?

Do you have a huge PR budget? Then keep reading for a moment. I will come back to you.

For those without that budget, there are lots of ways to become visible quickly.

To be clear: when I say you do not have a huge budget, I do not mean you are paying the Dutch business magazine Quote to mention you.

I mean you are not paying a PR agency to pitch journalists.

I am not talking about paying to take part (pay-to-play) here. I do not believe you should pay for a placement.

You can hire an agency if you do not have the time yourself. Your product, content and services do need to stand on their own two feet and convince someone to write about them.

Now, having a great product does not automatically mean they will write about you.

That is why people pay PR agencies: they know exactly who to email and how to pitch it.

It is not as simple as: "I have a great product, everyone is going to write about it."

Sorry, I just want to make sure this is clear.

So how do you fight your way to visibility without a PR agency?

The answer is (and this should not surprise you, because I have been talking about it this whole article): your own experts and expertise.

Go to the places your buyers already go to learn. Sometimes those are industry publications, professional associations, podcasts or webinars.

See if you can write an article or team up with someone whose work overlaps with yours.

Send your expert into the field on your behalf. Let them own their expertise.

Interior designer from Brabant as a guest in a small podcast studio

Get in touch with local or specialist media and say: "Hey, we have put together this great finding that we would love to share."

Collect your client stories (with permission) and share them.

Reach out to communities of professionals in your field. Your buyers already go there.

Sometimes these are websites, sometimes communities in places like Reddit.

Be careful with Reddit at times, because marketing is a bit odd there. You really need to present yourself as someone who gives valuable advice, not just as a mouthpiece for your brand.

These communities are great places to start.

It is fine if you are not yet mentioned alongside the biggest names in digital PR. Mentions on modest but relevant websites still make a difference.

We tested whether mentions on sites with high domain authority make ChatGPT recommend you more often.

The answer: a little.

Do mentions on sites with higher domain authority make ChatGPT recommend you more often?
Each mast is a group of sites. The higher the flag, the greater the chance ChatGPT recommends you after a mention on such a site.
What to do? Go for more mentions. Look beyond the domain authority of the site that mentions you.
Average chance of a ChatGPT recommendation, by domain authority of the site with the mention
Domain authorityAverage chance
0–200.84
21–400.79
41–60 (highest)0.91
61–800.81
81–1000.83
Source: MEDO MEDIA / zoekmachinespecialist.com June 2025, calibrated to the Dutch market. We studied 10 sites (less than a year old), tracked their AI visibility and placed 160 articles on sites with different domain authorities.

Just look at that middle range, a domain authority of 41 to 60. That is the highest bar in the chart.

And look at the lowest domain authority too: the 0 to 20 range still gets great visibility mentions.

The idea is to go after more mentions and look beyond domain authority. Go after the channels your buyers trust.

Pitch them a strong contribution to knowledge, a useful finding or an informed view on a current question. Pitch them a practical framework that can help readers make a decision.

Offer these channels an expert who can explain a complicated issue.

The idea is: actually make the pitch.

You do not need a PR agency to do this. If you do have one, that is great.

Even so, you can approach that whole list of industry organisations yourself and pitch them a story based on your expertise.

Teams are currently shifting from one-off mentions to delivering proof.

It is important to create proof that others actually want to cite, because that keeps working long after a one-off placement fades.

If you put out a news story that a major publication like Quote picks up, that is fantastic. Yet if people do not keep referring to that proof, the mention may not pay for itself in the long run.

So shift your mindset towards proof that others want to cite.

Look at where your peers focus their efforts and get ahead of them. Make sure you are there before they are.

Reviews are proof that keeps growing

Also remember that reviews are proof.

It is proof you can collect after every job.

Companies like HighLevel have processes that send review requests directly and automatically after every single job. That way your proof database keeps growing by itself.

You can use that proof for your pitch and say: "If you look at the responses, you can see that we solve this problem."

One thing I want to stress: you will probably have fewer reviews than a big, established organisation.

That is fine.

Google profiles side by side: Moore DRV Breda with 1 review, 216 Accountants with 7 reviews
Source: Google Business Profiles, 11 November 2026.

Moore DRV Breda has 1 review on Google, with 5 stars. 216 Accountants in Breda has 7. In its answer to 'beste accountant in Breda' (best accountant in Breda), AI Mode specifically mentioned the positive client reviews of 216 Accountants. Ask for a review after every set of annual accounts. That way you quickly close the gap.

We see language models do the following.

Take, for example, a company that sells phones and competes with Apple.

Apple might have 10,000 reviews. This company asks every customer for a review and gets dozens more each month as a result.

Apple, on the other hand, gets (let's say) only a few more in proportion over that same period.

What this tells the language model (such as ChatGPT) – assuming both sides get high-quality reviews – is that the newcomer has fewer reviews, yet they come in at a much faster pace.

The language model then thinks: "Wait, is the new trend that people prefer this product?"

This really helps a lot.

Now, there is a nuance to why I mentioned HighLevel here. It is not just about asking for a review.

What you can do is rotate the platform you ask for the review on (such as Google, Yelp or Better Business Bureau). You rotate these based on the sites where you fall short versus the sites where you are already strong.

Icon of three review cards with stars on a circling arrow

This gives you a head start, because you grow at a faster pace across all the platforms the AI looks at, not just on one platform.

Give your proof two home ports

The goal of all this work is not to replace what is on your website.

The idea is that you gather all this proof to answer that big question you chose to claim.

That is your campaign for this month.

You are going to give this proof two home ports: on your own page and off your page (external).

Owner of a boat hire business on the jetty of a marina in Brabant

External coverage helps people find you. Your own site gives them the chance to go deeper.

You can turn a customer observation into an article. You can turn those reviews into a tested recommendation. Turn that into an expert interview, a podcast or a magazine article and lead everything back to that in-depth resource on your own site.

You earn your mentions through your real work, the work you do every day. You can turn that straight into proof.

Search results about the Singelloop Breda: the Moore DRV site and Explore Breda with the Moore DRV Business Run
Source: Google Search, 11 November 2026.

Moore DRV has sponsored the Singelloop in Breda for years. Search for it and you will see both home ports. Its own site tells the story behind the sponsorship. On Explore Breda, the run for businesses is called the Moore DRV Business Run. Another site mentions the name. Its own site provides the depth.

Where can you compete as a smaller brand? Where can you create an advantage the big brands cannot replicate?

You cannot beat them on scale. You can beat them on depth and relevance.

Small brands can win on that every single time.

Check where the big brands fall short. Where are details missing, or advice for a specific audience segment? What outdated assumptions do they use?

Which of those gaps can you fill today?

Figure out what you know more about, then say less.

Map out the topics you can own, especially the niche use cases or implementation details big brands do not cover, such as local or regional considerations.

Go to the places big brands write about and cover the same topic, only sharper and more to the point.

If a big brand has made a complete guide, you might write about a very specific regional case.

Suddenly you are answering a question readers actually have.

You expand this one question at a time and only add a page when it really adds value, backed up by proof.

Measure what it delivers

You are looking for a big question. I know you might be thinking right now: "How do I know all this works and helps revenue?"

It is easier than you think.

It starts with tracking the same question over time. Build a fixed set of questions, linked to your priorities.

Record where you appear, how you are described and which pages are referenced.

Baseline measurement with 5 questions about Moore DRV Breda in Google AI Mode

This is what a baseline measurement looks like. I asked AI Mode five questions about an accountant in Breda, from broad to very specific. Moore DRV appears in four of the five answers. For the question about family businesses, the sources come from other firms. For the price question, Moore DRV drops out. Those are the questions to claim this month.

This is really an observational metric, a signal to keep an eye on, rather than a simple ranking number.

It is super exciting when your business gets a mention. Most teams celebrate that and then lose the thread.

How far can smaller brands trace the value of GEO?
Share of smaller brands that can measure at least this stage. The deeper into the funnel, the closer to revenue.
Percentage of smaller brands that can measure at least this stage
StageCan measure at least this stage
1. Brand appears in AI answers47%
2. Source citations or referenced pages29%
3. Website visits via AI referrals19%
4. Qualified leads or intro calls9%
5. Sales opportunities or pipeline6%
6. Revenue2%
Do not measure AI visibility at all55%
Source: MEDO MEDIA / zoekmachinespecialist.com, calibrated to the Dutch market. Data from a survey of B2B respondents.

As this funnel shows, about half of smaller brands can see whether they appear in AI answers. Only 2% can link that to revenue.

See it as a chain where you add one link at a time: referral visits, qualified leads and finally the pipeline.

When it comes to measuring your authority, think in layers.

Visibility and impressions are the top of the funnel.

The next layer is searches for your brand name and direct visits.

And finally the real business results: enquiries, the pipeline and revenue.

Moore DRV contact form asking how someone found the firm, with no option for AI
Source: moore-drv.nl, 11 November 2026.

Moore DRV's contact form already asks how someone found the firm. There are eleven options, from trade fair to search engine. An AI assistant is still missing. Add that option. Then you will see every month how many enquiries come from AI.

Fewer visits can still mean more revenue. That is what you are looking for.

It is important to judge pages on quality. Five qualified prospects are worth more than thousands of irrelevant visits.

Visitors referred by AI convert faster and are often worth seven times more per visitor. This is because they are pre-qualified: the AI has already done the comparison and they trust you before they click.

Now it is up to you to offer that seamless experience with a clear CTA.

Below you can read how to put all this into practice with my 30-day plan.

Your 30-day plan

You can do this work. Within a month.

It does not have to take a year before you see results.

Start with the valuable work no one else is doing yet. Look for activities your team rates as impactful and manageable, such as improving existing specialist pages or publishing client examples.

Infographic of the 30-day plan in four weeks: angle, proof, publishing and circulation

This week (week one), you choose your angle: one valuable customer segment or use case. Gather the questions prospects ask and find out where competitors are incomplete.

By the end of the week, you know exactly which question is worth making your own.

Next week (week two), you work on the proof. Interview experts, review client examples and note down lessons and trade-offs. Ask permission for client data or anonymise it.

By the end of the week, you have the proof behind your answer.

In week three, you publish. Create or improve the page on your site that answers the question. Give the answer first and then back it up. Add a clear next step, check your indexing and your internal links.

In week four, you put that expertise into circulation. Share the insight through your channels, pitch it to one or two publications or partners.

Set up a baseline. Where do we stand today?

And then write down the next question.

Start again next month.

Authority starts with one question

Authority is built by being useful where it counts.

As a smaller brand, you have a path to authority, starting with one great answer to an overlooked question.

You add the 'E' of E-E-A-T (experience) and show hands-on experience to back up your statistics.

Make it easy to understand, so AI can easily pull the answer out of it.

Then you expand to similar questions, without rambling on endlessly.

Measuring is crucial to be sure you generate the right kind of revenue.

Finally, you contribute by sharing your expertise on your website, social media and other channels. Consistent messaging across all these platforms helps them cite you more often.

It simply starts with one question. From there, you grow outwards.

You do not have to do 50 a month. Just do a few and make sure you do them well.

Frequently asked questions about authority in GEO for small businesses (FAQ)

Can a small business build authority in GEO?

Yes. Do not copy the big brands. Focus on very specific questions, back up your answer with your own experience and make sure other websites mention you.

How many mentions do you need to be recommended by AI?

From about 20 mentions a month, it starts to get interesting. 1, 2 or 3 websites will not get you there. The right number varies by industry.

Why do specific questions work better for small businesses?

The more specific the question, the fewer competitors have a good answer to it. Those questions have less search volume, yet they convert very well.

Do you need to publish lots of articles to become visible in AI?

No. You are better off overspecialising than overpublishing. A few strong answers with your own proof do more than 50 articles a week.

Do you need a PR agency to get mentions?

No. You can approach industry publications, professional associations, podcasts and professional communities yourself. Come with a strong finding or an expert from your own team.

How do you know if GEO earns revenue?

Track a fixed set of questions over time. Then measure in layers: visibility first, then branded searches and direct visits, then enquiries and revenue.

Conclusion

As a small business, you do not win the race in AI by copying big brands. You win on depth and relevance. Choose one specific question, back up your answer with your own experience and make sure others mention you. Then measure what it delivers and take on the next question. That is how you build authority step by step, even without a well-known name.

The next step

How we tackle this together

Discover your biggest growth opportunities with a free opportunity analysis (worth €416) — a tailor-made report, free of charge and with no obligation.

1Analysis
2Action plan
3Proposal
Schedule a call
Dominic Pietersz

About the author

Dominic Pietersz

Your dedicated point of contact

Dominic helps SMEs grow through Google and other channels. With his international experience, he turns challenges into concrete growth.

Follow Dominic on LinkedIn

Inspiration · The big DIY store had twenty kinds of paint on the shelf. The little paint shop two streets away was the only one that knew which paint sticks to my old oak window frames.