---
title: "Does AI visibility matter if SEO is already working?"
description: "Effective SEO does not show whether AI recommends your business. Learn what Google and Bing reports measure and how to monitor your brand's presence in AI answers."
published_at: "2026-11-02"
modified_at: "2026-11-02"
url: "https://dotgrow.net/en/knowledgebase/ai-visibility-and-seo/"
markdown_url: "https://dotgrow.net/en/knowledgebase/ai-visibility-and-seo/index.md"
lang: "en"
---

A business whose SEO brings in traffic and enquiries has every reason to ask why it needs another marketing channel. Google rankings show whether people can find the business. They do not show whether an AI assistant mentions it when a customer is gathering options and comparing suppliers – and the new Search Console and Bing Webmaster Tools reports reveal only part of that presence.

Business owners and marketing directors now hear about “AI search optimisation” from agencies, their own teams or their competitors’ activities. The decision about whether to invest in it should be based on data.

This article examines the question using Google documentation, research from 2025–2026 and our own experience. Some studies cover Poland, others the US market.

## AI visibility and effective SEO – at a glance

AI visibility matters, but its importance varies from one business to another. Good SEO helps when it is based on customers’ intent and questions, rather than popular keywords alone. It does not tell you whether your brand appears in answers from AI models such as ChatGPT, Claude, Gemini or Perplexity – where some customers now make their choices.

At the very least, every business with effective SEO should include AI in its organic marketing plan and regularly check how assistants answer a selection of questions from potential customers.

The main takeaways for businesses:

1.  **Good SEO helps with AI visibility when it answers customers’ questions, not just targets keywords.** SEO that targets keywords which attract traffic but not customers does not carry over well into assistants’ answers.
    
2.  **AI in Google and AI assistants are two different places.** SEO principles apply directly to AI Overviews and AI Mode, while ChatGPT and Perplexity operate outside Google Search.
    
3.  **A high Google ranking does not mean AI assistants will cite your website.** Most links cited by ChatGPT, Gemini and Copilot come from outside Google’s top ten results.
    
4.  **Customers ask AI when gathering information and comparing offers, but make the decision themselves.** The assistant helps narrow the options, while the customer has the final say.
    
5.  **SEO reports reveal only part of your AI presence.** Google and Bing reports measure page impressions or citations, while Google Analytics records recognised visits from AI assistants. They do not show whether or how an assistant described or recommended your brand.
    
6.  **The minimum is to monitor a selection of customer questions.** A fixed set of questions, with some regularly replaced, is enough to get started.
    
7.  **The scope of further work depends on the buying process.** The more often customers use AI to compare suppliers, the more reason there is to go beyond monitoring.
    

No activity guarantees a mention, citation or recommendation. AI answers vary, and appearing in them does not automatically lead to enquiries.

We explain how AI visibility works and how to build it in our [guide to AI search optimisation](/en/knowledgebase/ai-search-optimisation/).

## Do good Google rankings mean you appear in AI answers?

To an extent. Google’s AI search features, AI Overviews and AI Mode, follow the same principles as SEO, so good rankings do help. The relationship is much weaker for standalone assistants such as ChatGPT, Claude, Gemini and Perplexity, because only a small proportion of the pages they cite come from Google’s top ten results.

### AI in Google and AI assistants

[Google explicitly states](https://developers.google.com/search/docs/appearance/ai-features) that SEO best practices still apply to AI Overviews and AI Mode, with no additional optimisation required. Rankings do not translate directly into citations, though. An [Ahrefs analysis from March 2026](https://ahrefs.com/blog/ai-overview-citations-top-10/), covering 863,000 search results pages, found that 37.9% of pages cited in AI Overviews appeared among the first ten elements on the results page, including ads and Google features such as “People also ask”. Even within Google, a high ranking therefore increases the chance of a citation without guaranteeing one.

The relationship is weaker still in AI assistants. According to an [Ahrefs study from August 2025](https://ahrefs.com/blog/ai-search-overlap/), an average of around 12% of links cited by the assistants studied (ChatGPT, Gemini, Copilot and Perplexity) appeared in Google’s top ten for the same query. For Perplexity, the figure was around 28.6% (in a predominantly English-speaking market). This does not mean Google rankings have stopped mattering. It means a high ranking does not tell you whether an AI assistant will mention or recommend a business. That presence needs to be checked separately.

|  | AI in Google (AI Overviews, AI Mode) | AI assistants (ChatGPT, Gemini, Perplexity) |
| --- | --- | --- |
| Where the customer asks | In Google search results | In a separate app or on the assistant’s website |
| Relationship with Google rankings | Strong: SEO principles apply, but a ranking does not guarantee a citation | Weak: only a small proportion of cited pages rank in Google’s top ten |
| What SEO reports show | Search Console reports link impressions in AI Overviews and AI Mode, but not the content of the answers | GA4 shows recognised visits after a click. Bing Webmaster Tools reports citations in Copilot and Bing’s AI answers, but not brand recommendations |

### Keyword-focused SEO and SEO built around customer questions

In our experience, large brands that were already strong before AI assistants developed tend to appear both in Google and in AI answers, because language models were trained partly on content about them. For younger businesses, SEO alone is often not enough – especially when it attracts visitors through keywords that do not bring in customers.

SEO helps with AI visibility when it is based on analysing intent and segmenting potential customers.

### Mention, citation, recommendation

A brand can appear in an AI answer in three different ways, which we explain in more detail in the guide:

-   **Mention:** the assistant names the brand.
-   **Citation:** the assistant links to the business’s website as a source.
-   **Recommendation:** the assistant suggests the brand as a choice for the person asking.

An [Ahrefs analysis from 2025](https://ahrefs.com/blog/ai-overview-brand-correlation/), covering 75,000 brands, found that online brand mentions correlated with visibility in AI Overviews about three times more strongly (0.664) than the number of backlinks (0.218; the closer to 1, the stronger the relationship). This is a correlation, not proof of causation. It does suggest, though, that AI visibility deserves a look beyond what a brand publishes on its own website to what others write about it.

“

> SEO shows that people can find us. AI visibility shows whether we are present when a customer already knows they are looking for something and is comparing options. Those are two very different questions. Most businesses today address only the first.

![Daniel Andraszewski](/_astro/avatar_daniel.CpC5s8jg_ReWpF.png)

Daniel Andraszewski

Founder and CEO of dotGrow, a technology partner for business growth.

## At what stage of a purchase do customers ask AI?

Customers mainly ask AI when they are gathering information, comparing offers and looking for opinions – before they even visit a business’s website. According to an [On Board Think Kong study from April 2026](https://itwiz.pl/juz-61-polakow-korzysta-z-chatbotow-ai-podczas-zakupow/), 61% of Polish consumers had used AI when planning purchases. They usually keep the decision in their own hands, but the shortlist they choose from may already have taken shape in a conversation with an assistant.

In the same study, the most frequently reported use of AI was comparing products and services (50.8%), while 42.6% reported looking for recommendations when making their final choice. Respondents could select several uses. These are consumers’ own accounts, mainly from the B2C market, rather than transaction data.

According to the [Gemius report “E-commerce in Poland 2026”](https://gemius.com/pl/news/raport-e-commerce-w-polsce-2026-jest-juz-dostepny/), AI tools had 16 million real users in Poland in June 2026. At the same time, 71% of respondents do not want to let AI make a purchasing decision on its own. For a business, this means AI rarely chooses on the customer’s behalf, but may influence which suppliers the customer chooses between.

A similar pattern appears in B2B, although these figures come from the US market. In a [Semrush study from March and April 2026](https://www.semrush.com/blog/how-ai-shapes-b2b-buying/) of 519 US professionals who use AI at work, 66% regularly research suppliers with it, and 92% say AI influenced their shortlist. Almost all verify the recommendations before deciding. The study only included people who already use AI, so it does not show how widespread this behaviour is across the whole market – it shows how AI works where it is already being used.

A B2B buying journey might look like this for an operations director at a manufacturing business looking for a warehouse management system supplier (illustrative example):

![Customer journey from asking AI, through a supplier shortlist and comparison, to visiting a website and making contact](/_astro/ai-visibility-customer-journey.CbGlcmrM_1c6fo2.png)

An illustrative customer journey: asking AI → supplier shortlist → comparison → visiting suppliers’ websites → enquiry. The shortlist takes shape before the website visit.

1.  They ask an AI assistant what to look for when choosing a system.
2.  They ask for a few businesses that implement these systems in Poland.
3.  They compare three or four names from the answer.
4.  They visit two or three suppliers’ websites and check what they offer.
5.  They send an enquiry to one of them.

Traffic reports show, at most, step 4. Google and Bing reports may also show earlier page impressions or citations, but not the process of creating the shortlist itself.

The shortlist was created earlier, in a conversation with the assistant.

## What do Search Console, Bing Webmaster Tools and Google Analytics show – and what do they miss?

Search Console and Bing Webmaster Tools already reveal part of a website’s AI presence: link impressions or citations. Google Analytics shows recognised visits to the website. They do not show whether a brand was mentioned without a link, how it was described or which other businesses appeared alongside it. If a customer reads an answer without clicking a link, traffic reports record nothing. That is why SEO data alone cannot tell you how visible a brand is in AI.

### Google Search Console: AI impressions

Since June 2026, Google has offered a [generative AI performance report](https://support.google.com/webmasters/answer/16984139?hl=pl). It has been available globally since 31 August 2026, including for Polish websites, although it may not appear if a site has too few impressions.

The report shows how often links to a website appeared in AI Overviews and AI Mode, broken down by factors including URL, country, device and date. A separate report covers generative AI in Discover. It does not show the content of answers, users’ questions or whether the assistant mentioned or recommended a brand. There is also no click metric, so impressions cannot be linked directly to visits.

### Bing Webmaster Tools: AI citations

Since February 2026, [Bing Webmaster Tools has offered the AI Performance report](https://blogs.bing.com/webmaster/2026/2/Introducing-AI-Performance-in-Bing-Webmaster-Tools-Public-Preview/). It covers Microsoft Copilot, AI answers in Bing and selected partner integrations.

The report shows the number of times a website was cited, the most frequently cited URLs, changes over time and _grounding queries_, the phrases AI uses when searching for sources. These are not users’ exact questions. The number of citations does not tell you whether AI recommended a business or how many visits its appearance in an answer generated.

Google and Bing results should not be compared directly: Google reports link impressions, while Bing counts citations.

### Answers without a click

Some encounters with a brand end with the answer itself. In a [Pew Research Center study](https://www.pewresearch.org/short-reads/2025/07/22/google-users-are-less-likely-to-click-on-links-when-an-ai-summary-appears-in-the-results/) (browsing data from 900 US adults, March 2025), users clicked a traditional result in 8% of visits with an AI summary in Google and 15% of visits without one, while a link within the summary itself was clicked in around 1% of visits. The findings concern Google Search in the US, but illustrate the mechanism well: the more complete the answer, the less often users click and the less evidence remains in analytics.

### GA4 only captures some traffic from AI assistants

GA4’s default channel group includes an [“AI Assistant” channel](https://support.google.com/analytics/answer/9756891?hl=pl), which includes recognised visits from services such as ChatGPT, Gemini, Claude and Copilot. It only counts visits made by following a link from an assistant, though. It does not include AI Overviews or AI Mode, and GA4 may classify visits from assistants that do not pass a referrer as Direct traffic.

In practice, the report therefore shows a lower bound for AI traffic, not its full scale. We are not currently aware of any tools that reliably capture the rest – which is why asking customers remains the most reliable signal.

### The best signal: asking new customers

The most reliable information comes from asking new customers how they heard about the business. You can add the question to a contact form and ask it during the first conversation with a new customer, for example “How did you hear about us?” with options such as “Google Search”, “an AI assistant (e.g. ChatGPT)”, “a recommendation” and “other”. It is a simple approach that also captures contacts which began with an answer without a click.

### From dotGrow’s experience

-   **Observation:** when businesses first check AI answers, they are usually most surprised by the complete absence of brand mentions, rather than a lack of recommendations. A brand may be reasonably visible in Google, including in AI Overviews and AI Mode, yet not appear at all in most AI assistants.
-   **Interpretation:** ChatGPT, Claude, Gemini and Perplexity are not directly connected to Google Search, so keyword-focused SEO does not automatically carry over into them, although some businesses assume it does.
-   **Main takeaway:** we check a brand’s presence in AI assistants separately, regardless of its Google results.

## How much attention should your business give AI visibility?

No business with effective SEO should overlook AI as a channel, but the scope of the work depends on how its customers choose suppliers. Monitoring a selection of customer questions is the minimum, while SEO built around questions along the customer’s decision journey should be standard practice. Separate work makes sense once customers are comparing businesses and solutions in AI and the brand is missing from the answers.

From our perspective, the question is how to include AI in SEO, rather than whether to work on it. We do not know of a business with effective SEO that we would advise to do nothing beyond checking AI answers, without including AI in its organic marketing plan.

### Three levels of involvement

| Level | When | What to do | What not to expect |
| --- | --- | --- | --- |
| **1\. Minimum: monitoring** | Any business doing SEO | A fixed set of customer questions, with some regularly rotated. Include AI in the organic marketing plan. | That monitoring alone will increase the brand’s AI presence |
| **2\. SEO with AI elements** | Customers gather information and compare offers before making first contact | Map the ideal customer’s decision journey, list their questions and problems, and build a content strategy that answers them | Guaranteed mentions or quick results |
| **3\. Separate AI search optimisation work** | Customers compare suppliers and look for recommendations, while the brand is missing from answers across several rounds of checks | A separate plan for the brand’s presence in AI assistants | That a mention will automatically generate enquiries or sales |

### Questions that help you choose a level

1.  Do customers compare several suppliers before making first contact?
    
2.  Is your offering complex or expensive enough that customers spend a long time gathering information before choosing?
    
3.  Do new customers name an AI assistant as their source of information about the business?
    
4.  Is your brand missing from answers in which competitors appear across several rounds of checks?
    

The more “yes” answers, the more sense a higher level makes. A brand’s absence from answers may be a missed opportunity, but it is not proof of lost revenue – studies show that AI supports the choice, but do not describe an individual business’s buying process.

“

> The way people look for information is changing. An AI assistant is always close at hand, on your phone, and that is simpler than searching Google. That is why it is worth including this channel when planning organic marketing.

![Jake Smolarek](/_astro/avatar_jake.iQsiQxLX_29tWqh.png)

Jake Smolarek

Strategic partner at dotGrow, specialising in AI search positioning and traditional SEO.

## How can you check and monitor your brand’s AI presence without a large investment?

A manual test is enough to start, without tools or a large budget. Gather 10–20 questions that customers ask before choosing a supplier. In **incognito mode**, ask each question in two or three AI assistants and in Google, then record whether the brand appears, how it is described and which other businesses are mentioned alongside it. Repeat this at set intervals, for example weekly.

### 1\. Customer questions instead of the business name

Start with questions from the decision journey: about the problem, comparing solutions and choosing a supplier. Asking about the business by name tells you little, because the customer often does not yet know it at this stage. Sales conversations and requests for quotations are good sources of questions.

### 2\. Two or three assistants and AI in Google

Two or three tools from ChatGPT, Claude, Gemini and Perplexity are enough. Check AI Overviews and AI Mode separately, because they operate within Google Search, while AI assistants operate outside it.

### 3\. Repeat the same question several times

Ask each question several times in separate sessions, because AI answers vary considerably. A single answer therefore tells you little – what matters is the brand’s share of appearances across many answers. If the brand appears in, say, two out of three cases, that is an early signal, not proof of lasting visibility.

### 4\. Record the brand, description, sources and competitors

Record the results in a simple spreadsheet. For each answer, note whether the brand appeared, how it was described, which sources the assistant cited and which businesses it named alongside it. After several rounds, you will see a pattern that a single check cannot reveal.

### 5\. A fixed set with some questions rotated

Keep most questions unchanged so that you can compare results over time. Replace part of the set at regular intervals, for example monthly, because customers ask about different things.

A manual check is a starting point, not a full audit. It is worth supplementing it with data from Google and Bing reports, remembering that they do not show complete answers or brand recommendations. We explain how to measure visibility more broadly in our [guide to AI search optimisation](/en/knowledgebase/ai-search-optimisation/).

## Which conclusions should you avoid?

The most common mistakes in assessing AI visibility come from applying SEO logic to AI assistants and judging them on a single answer. A high Google ranking, an appearance in AI Overviews or a single ChatGPT mention does not yet tell you how the brand performs across answers, nor does it justify moving budget from SEO to AI.

-   **“We are in Google’s top three, so AI knows us.”** A Google ranking helps with AI Overviews and AI Mode, but has a weak relationship with assistants’ answers, which – as shown above – mainly cite pages outside Google’s top ten.
    
-   **“We appear in AI Overviews, so ChatGPT will recommend us too.”** AI Overviews are part of Google Search, while ChatGPT, Gemini and Perplexity are separate platforms. Appearing in one place does not determine whether you appear in another.
    
-   **“We were missing from one answer, so we are not visible.”** AI answers change even for the same question, so a single absence – just like a single mention – proves nothing. What matters is the brand’s share of appearances across a series of answers.
    
-   **“An AI mention means new customers.”** A mention, a citation, a recommendation and a customer enquiry are different outcomes. A brand mentioned in an answer may not make the customer’s shortlist, and its absence does not prove lost revenue – it is a signal to examine the situation more closely.
    
-   **“We need to move budget from SEO to AI.”** According to Google, SEO best practices also apply to AI Overviews and AI Mode, so SEO built around customer intent remains the foundation. AI is added to the organic marketing plan rather than replacing it.
    

These mistakes share a simple pattern: a result from one channel or one answer is treated as the whole picture.

We explain which content AI is more likely to cite in our [article on content cited by AI](/en/knowledgebase/ai-citation-ready-content/).

## Frequently asked questions

### Does SEO still make sense when customers ask AI?

Yes. SEO built around customer intent still brings in traffic. According to Google, the same SEO practices are also the basis for visibility in AI Overviews and AI Mode.

### Will AI replace Google Search?

We do not have data that supports this. Customers use both channels, and Google itself is developing AI features in search results.

### Does appearing in AI answers bring customers and traffic?

Not automatically. A mention may help a brand reach a customer’s shortlist, but some encounters end without a click, and customers make their own choice of supplier.

### Is GEO a separate service or SEO under another name?

GEO (Generative Engine Optimisation) grows out of SEO and uses the same foundations. It also covers a brand’s presence in AI assistants outside Google. You may also come across the terms AI SEO or AEO. We explain this further in the [guide](/en/knowledgebase/ai-search-optimisation/).

### Can anyone guarantee that AI will recommend my business?

No. Nobody controls the answers AI models produce. You can increase the chance of your brand appearing by answering customers’ questions and building brand authority.

## Summary: keep SEO and add AI to the plan

SEO remains the foundation of visibility when it is built around customer intent and questions, rather than keywords alone. A brand’s presence in AI assistants still needs to be checked separately, because Google and Bing reports reveal only part of its AI presence, while the scope of further work depends on how customers choose suppliers.

The question, then, is not “SEO or AI”. It is whether we know what customers hear about us before they visit the website.

Start with the questions customers ask when deciding whether to work with you, and check the answers regularly. Several rounds will show how often the brand appears in answers and whether AI deserves more attention in your organic marketing plan.

## About the authors

![Daniel Andraszewski](/_astro/avatar_daniel.CpC5s8jg_ZWg6BN.png)

Daniel Andraszewski

Founder and CEO of dotGrow, a technology partner for business growth.

![Jake Smolarek](/_astro/avatar_jake.iQsiQxLX_Z26t9y8.png)

Jake Smolarek

Strategic partner at dotGrow, specialising in AI search positioning and traditional SEO.

## Latest articles

-   [![Does AI visibility matter if SEO is already working?](/_astro/ai-visibility-and-seo-hero.6SPa2t4m_Z198gmP.png) 2 Nov 2026 ### Does AI visibility matter if SEO is already working?](/en/knowledgebase/ai-visibility-and-seo/)
-   [![Which content is most likely to be cited by AI models?](/_astro/ai-citation-ready-content-hero.CMdOC9Mt_Z2v6NST.png) 12 Sept 2026 ### Which content is most likely to be cited by AI models?](/en/knowledgebase/ai-citation-ready-content/)
-   [![AI Search Optimisation: a practical guide for businesses in 2026](/_astro/ai-search-optimisation-hero-v2.C7Ak7T5c_UuDSC.png) 13 Aug 2026 ### AI Search Optimisation: a practical guide for businesses in 2026](/en/knowledgebase/ai-search-optimisation/)

## Popular topics

[ai](/en/knowledgebase/tag/ai/)[ai search optimisation](/en/knowledgebase/tag/ai-search-optimisation/)[content marketing](/en/knowledgebase/tag/content-marketing/)[seo](/en/knowledgebase/tag/seo/)