AI visibility is becoming an important part of how marketers measure search performance.
If you’re asking whether you’re being mentioned or cited by AI, the answer can be both but they’re different outcomes: a mention is when AI names your brand in an answer, while a citation is when AI uses your website as a source.
For marketers, SEO professionals, and brand strategists, that distinction matters because Ai-generated mentions and AI citations affect brand recognition, search performance, referral traffic, and business results in different ways. A brand appearing in an AI-generated answer and a website being cited as a source are not the same thing, and they shouldn’t be measured as though they are.
A recent Semrush study on AI search in manufacturing illustrates this difference.
Between January and July 2026, AI Overviews grew from appearing across 38% to 57% of tracked manufacturing search volume. Yet AI assistants and Google’s AI Mode accounted for just 0.48% of website sessions in the study.
The most interesting finding, however, was the difference between the brands mentioned and the websites it cited.
Of the top 15 most-mentioned brands and the top 15 most-cited domains, only two appeared on both lists: Vevor and Grainger.
That’s the gap this article examines: the difference between AI mentions and AI citations, how each shapes SEO and brand visibility, what the Semrush data and Brandesis findings show in practice, and how to measure and use AI visibility more effectively.
Are we trying to get AI to recognise our brand, use our content as a source, or send visitors to our website?
Because those are three different outcomes, and we shouldn’t measure them as though they’re the same thing.
What Is the Difference Between AI Mentions and AI Citations?
Before looking at the data, it’s important to understand what these metrics represent.
AI Mentions: When AI Talks About Your Brand
An AI mention occurs when an AI-generated answer names or references a brand.
For example, someone might ask ChatGPT:
Which agencies in Botswana specialise in brand strategy?
Ofcourse Brandses is mentioned.
If ChatGPT includes Brandesis in its answer, that is a brand mention.
The platform has recognised the business as relevant to the question, whether or not it provides a link to the website.
This matters because customers are increasingly using AI to discover companies, compare alternatives and understand their options.
A mention can introduce your brand to someone who has never heard of it.
But it doesn’t necessarily mean the person will visit your website.
They may continue asking questions, search for your brand later or simply remember the name.
That makes mentions particularly relevant to brand recognition, positioning and how a business is associated with specific topics.
However, a mention alone doesn’t prove that AI trusts the brand or that the mention influenced a purchase.
AI Citations: When AI Uses Your Website as a Source
An AI citation occurs when an AI-generated answer references or links to a specific page used to support its response.
For example, someone might ask:
How does brand clarity affect AI search visibility?
An AI platform could answer the question and cite an article from nonofojoel.com without mentioning Nonofo Joel as a recommended brand or expert.
Many platforms generate these citations through the retrieval-augmented generation process, pulling real-time web pages and showing them as clickable footnotes or source references.
The content has contributed to the answer, but the brand hasn’t necessarily become part of the conversation.
This distinction is important.
A website can become a frequently cited source without the company behind it becoming a frequently mentioned brand.
Citations can create a route to website traffic if a user follows the source link.
But a citation still doesn’t guarantee a click.
The reader might already have everything they need from the AI-generated answer.
AI Mentions vs Citations: A Quick Comparison
| Metric | AI mentions | AI citations |
|---|---|---|
| What it measures | Whether AI names your brand | Whether AI references your content or domain |
| Primary value | Brand recognition, mentions build awareness, and topic association | Source visibility and a direct path to the source |
| Requires a website link | No | Typically appears as a citation link or source reference rather than a generic link |
| Can generate traffic | Indirectly | Potentially through referral clicks |
| Guarantees traffic | No | No |
| Guarantees conversions | No | No |
| Main question | Is AI talking about us? | Is AI using our content? |
Neither metric is inherently more valuable.
Their importance depends on what you’re trying to achieve.
What Semrush’s Manufacturing Study Reveals
The Semrush research offers an interesting picture of how AI search is developing in a commercially important industry.
Manufacturing is particularly useful to examine because purchasing decisions can involve substantial research, technical specifications, supplier comparisons and long buying cycles.
Customers may use AI to understand products before approaching a manufacturer or distributor.
Three findings stood out to me.
1. AI Search Visibility Is Growing Faster Than AI Referral Traffic
The study found that AI Overviews appeared across 57% of tracked manufacturing search volume by July 2026, up from 38% in January.
That’s a significant increase over seven months.
At the same time, AI assistants and Google’s AI Mode accounted for just 0.48% of sessions to manufacturing websites.
Direct and organic search still accounted for nearly 80% of sessions.
These numbers tell us two things.
First, AI-generated answers are becoming a more common part of the search experience.
Second, their growing presence doesn’t automatically translate into large volumes of referral traffic.
I think this is where some of the conversation around AI search gets ahead of the evidence.
We’re seeing more AI-generated answers, but that doesn’t mean traditional search traffic has suddenly become irrelevant.
AI visibility is growing, while the traffic it directly generates remains relatively small in this study.
It’s also worth noting that the two figures measure different things. AI Overview coverage measures how frequently that feature appears across tracked search volume, while referral sessions measure visits attributed to particular AI sources. They shouldn’t be treated as directly comparable percentages.
For marketers, the implication is that we need to continue investing in traditional SEO while learning how AI changes discovery.
I’ve written more about this relationship in Why Tracking AI Visibility Matters.
2. The Most-Mentioned Brands Are Rarely the Most-Cited Websites
This was the finding that caught my attention.
Semrush compared the 15 most-mentioned brands with the 15 most-cited domains.
Only two appeared on both lists.
Those were Vevor and Grainger.
AI search results are volatile, with 50% of cited domains changing every month.
Think about what that means.
The brands AI systems frequently name aren’t necessarily the websites those same systems use to support their answers.
A manufacturer may have strong brand recognition, established products and a reputation within its industry.
AI may mention that manufacturer when recommending suppliers.
But when answering a technical question, the platform may prefer to cite a specialist publication, educational resource or independent website.
The reverse can also happen.
A website might publish excellent technical explanations that AI systems frequently cite without the company itself being recommended to customers.
This creates two different visibility opportunities.
One is to become a brand AI associates with a particular market or problem.
The other is to become a source AI uses when explaining that market or problem.
Ideally, businesses should understand how they perform in both. Only 30% of brands stay visible across consecutive AI queries, so repeated checks matter.
3. Being Cited Doesn’t Guarantee Traffic
Another important finding concerned the pages receiving AI referral traffic.
According to the study, cited pages offering a clear next step, such as relevant product pages or original research, earned more AI traffic than generic reference content.
This makes sense when you consider the user’s journey.
Imagine someone asks an AI platform:
What is the difference between stainless steel 304 and 316?
If the AI answer provides a complete explanation and cites an informational article, the user may have no reason to click.
Now imagine someone asks:
Where can I buy 316 stainless steel fittings for a commercial project?
A citation to a relevant supplier’s product page may be more useful because the reader has something practical to do next.
The difference isn’t simply whether the page was cited.
It’s whether visiting the page helps the user complete the next part of their task.
This has implications for content strategy.
We shouldn’t only ask whether our content is citation-worthy enough to be cited.
In some cases, comparison pages can outperform generic reference content when the user is weighing options and needs a clear next step.
We should also ask whether the page provides value beyond the summary an AI platform can generate.
I Checked Brandesis: The Gap Between Mentions and Citations
I wanted to compare the findings with something closer to my own work.
So I looked at the Semrush AI Visibility Toolkit data for Brandesis.
Brandesis is a Botswana-based brand strategy agency working across brand strategy, content, website design and distribution.
The data, dated October 14, 2026, showed an interesting difference between brand mentions and website citations.
Brandesis AI Visibility Overview
| Metric | Result | Reported change |
|---|---|---|
| AI Visibility Score | 14/100 | Low |
| AI Mentions | 3 | +50% |
| AI Citations | 163 | +7.9% |
| Cited Pages | 152 | +5.6% |
| Performing Topics | 140 | — |
| Topic Opportunities | 11 | — |
| Source Opportunities | 55 | — |
Source: Semrush AI Visibility Toolkit, Brandesis visibility overview, October 14, 2026.
At first glance, the numbers might appear contradictory.
Brandesis has only three recorded mentions, yet its website has 163 citations and 152 cited pages.
The AI Visibility Score remains low at 14 out of 100.
But the metrics are measuring different things.
The data suggests that Brandesis content is appearing as a source much more frequently than the brand itself is being named in the tracked AI answers.
That may partly reflect the fact that 85% of brand mentions in AI answers come from third-party pages rather than a brand’s own site.
That distinction is exactly what the manufacturing study highlights.
Our website can contribute information to AI-generated answers without Brandesis becoming a prominent part of those answers.
There is also an important measurement detail here: citations, cited pages and mentions have different counting methods. We shouldn’t interpret 163 citations and three mentions as a direct citation-to-mention conversion rate.
What the figures give us is a reason to investigate further.
Which AI Platforms Are Citing Brandesis?
The Semrush report also breaks down cited pages by platform.
| AI platform | Cited pages |
|---|---|
| ChatGPT | 44 |
| Google AI Overview | 38 |
| Google AI Mode | 27 |
| Gemini | 48 |
These figures show that Brandesis pages are appearing as cited sources across several AI search experiences.
Gemini accounts for the largest number of cited pages in this view, followed by ChatGPT.
Across major AI engines, visibility can vary widely because AI engines often show different mention rate and citation rate patterns from one platform to another.
For example, Gemini has an 83.7% mention rate and 21.4% citation rate, while ChatGPT has an 87% citation rate but only 20.7% mention rate.
Citation overlap can also be low: only 11% overlaps between ChatGPT and Perplexity.
However, we shouldn’t assume that a larger number of cited pages means more traffic.
Each platform has a different interface, citation behaviour and user experience.
Someone reading a ChatGPT answer may behave differently from someone encountering an AI Overview in Google Search.
The next question is whether those citations are producing measurable visits and whether the cited pages give readers a reason to continue.
What Are AI Platforms Actually Associating With Brandesis?
The second part of the Semrush report provides another useful perspective.
It shows the topics where Brandesis appears, along with associated prompts, visibility and estimated AI search volume.
The report identifies 140 performing topics and 164 prompts.
Some of the visible topics include:
| Topic | Visibility | Brand mentions |
|---|---|---|
| Chrome Hearts | 11 | 1 |
| Heart Brand Identity and Marketing | 7 | 1 |
| Designers in Botswana | 6 | 1 |
| SEO and Keyword Optimization | 0 | 0 |
| Value Proposition Design | 0 | 0 |
| Brand Ambassador Programs and Roles | 0 | 0 |
This is where AI visibility becomes a brand strategy question.
Not every topic associated with a website is equally valuable to the business.
For Brandesis, appearing around brand identity, value proposition design, brand strategy and related marketing subjects is more relevant to our positioning than appearing around unrelated topics.
A large estimated AI search volume doesn’t automatically make a topic commercially valuable.
For example, a topic may have substantial search demand but little connection to the services a company wants to sell.
If we chase that visibility simply because the volume looks attractive, we risk building traffic and citations without strengthening the brand’s position.
That’s why I prefer looking at topic relevance before topic volume.
The important question isn’t how many topics we can appear in.
It’s whether we’re becoming more visible in the topics that matter to our customers.
This connects with something I’ve explored in Topical Authority in the Age of AI Search.
A business doesn’t become known for something simply because it publishes one article about it. The association becomes stronger when there is consistent evidence of expertise across relevant content and other credible sources.
What the Brandesis Data Tells Me
Looking at these two reports together, I see three areas worth investigating.
Our Content Is Being Used, but Brand Recognition Needs More Work
The citation figures suggest that Brandesis has content appearing in AI-generated answers.
That’s useful.
But the low mention count and visibility score suggest the brand is not yet being named frequently in the tracked conversations. Tracking mentions alongside citations helps expose this kind of mention-citation gap.
For me, the next step isn’t simply publishing more articles.
It’s understanding which pages are being cited, what questions they answer and whether the topics reinforce the expertise we want Brandesis to be known for.
We Need to Understand Which Citations Have Commercial Value
Not all citations deserve equal attention.
A cited article answering a broad informational question may support visibility without creating a direct path to our services unless the cited page gives the visitor a clear next step.
A cited case study, service page or practical resource may be more valuable when the user’s question suggests they’re evaluating solutions. citations drive traffic only when the citation link leads to a page worth visiting beyond the AI summary.
I would therefore review cited pages against their purpose.
Does the page answer a useful question?
Does it establish relevant expertise?
Is there a logical next step for someone interested in working with us?
Can the reader easily discover related services, case studies or resources?
These are ordinary content and user-experience questions.
AI search hasn’t made them less important.
We Need to Connect Visibility With Website Performance
The Semrush report shows where Brandesis appears across tracked AI answers.
It doesn’t establish how much revenue those appearances generate.
To understand that, I’d need to compare the visibility data with website analytics.
That means looking at AI referral sessions, landing pages, engagement, enquiries and conversions. For citation tracking, I’d combine prompt testing, traffic analysis, and log monitoring.
I’d also monitor branded search demand over time.
However, increases in direct traffic or branded searches cannot automatically be attributed to AI mentions. They can provide supporting context, but there are many other reasons those metrics might change.
The objective is to understand whether improving AI visibility contributes to business performance, not to claim every improvement was caused by AI.
AI Mentions, Citations and Traffic Should Be Measured Together
I think one of the biggest mistakes we can make is treating AI visibility as a single metric.
It isn’t.
A brand can be frequently mentioned but rarely cited.
A website can be frequently cited but rarely mentioned.
And either can happen without meaningful referral traffic.
That’s why I would measure AI search performance across three connected areas.
1. Brand Recognition
Are AI platforms mentioning the business when users ask relevant questions?
Are they associating it with the right services, products and topics?
Are they describing the company accurately?
This is where brand mentions, share of voice and brand clarity become important.
A brand mention is only useful if the association makes sense. Tracking brand mentions in AI also helps measure brand visibility even when there is no citation.
2. Content Visibility
Which pages are being cited?
Brands earn citations when retrieval systems can select a page to support an answer.
What topics do they cover?
Which platforms are citing them?
Are those pages helping establish the expertise the business wants to be known for?
This is where citations, cited pages and source analysis become useful. Content with structured headings is 28–40% more likely to be cited.
The objective isn’t to get every page cited.
It’s to build useful, credible resources around relevant subjects.
3. Business Outcomes
Are AI referrals increasing, and do we need to track citations alongside them when evaluating outcomes?
Which pages receive those visits?
Do visitors engage with the website?
Are they enquiring, signing up or purchasing?
Are we seeing changes in branded search demand that warrant further investigation? You should also track citations by prompt and platform before tying them to business outcomes.
These questions help connect AI visibility with actual marketing performance.
Without them, it’s easy to celebrate a growing citation count while having no idea whether the visibility is helping the business.
Why Your Most-Cited Page Might Not Be Your Most Valuable Page
This is another lesson I take from the manufacturing research.
Marketers often focus on getting informational content cited because educational articles are natural sources for AI-generated explanations.
But the most-cited page isn’t necessarily the page creating the most commercial value.
A detailed article may receive citations across dozens of prompts while generating very little traffic.
Meanwhile, a product comparison or service page might receive fewer citations but attract visitors who are closer to making a decision. Some pages are also more likely to earn both a citation and a brand mention in the same response.
The important distinction is the role each page plays.
Informational content can introduce a subject, demonstrate expertise and support topical relevance.
Commercial content can help users evaluate options and take action.
Both matter.
But they shouldn’t be judged by citation volume alone.
For content teams, this means thinking beyond whether a page can answer a question.
We also need to consider what the reader might need after receiving that answer.
If someone wants more detail, can they find it?
If they’re comparing options, is there a useful comparison?
If they’re ready to enquire, is the next step obvious?
This isn’t about forcing a sales pitch into every article.
It’s about making sure useful content leads somewhere useful.
What Should Marketers Do Differently?
I wouldn’t respond to the Semrush findings by abandoning traditional SEO or suddenly optimising every page for AI citations.
The manufacturing data doesn’t support that.
Direct and organic search still account for the overwhelming majority of sessions in the study.
Instead, I’d make AI visibility part of the existing search and brand measurement process.
Track Mentions and Citations Separately
Don’t combine them into a single success story.
Understand whether the business is being named, whether its content is being used and where those outcomes overlap.
Prioritise Relevant Topics
Focus on the questions customers ask when researching the problems your business solves.
A mention in an irrelevant topic may contribute to a visibility count without improving commercial relevance.
Review the Pages AI Is Citing
Look at their content, structure, usefulness and next steps.
Review pages already getting cited to see what makes them citation-worthy; they may be signalling stronger content authority that you can apply to other pages.
Ask whether they offer something valuable beyond the summary provided by the AI platform.
Strengthen Brand Associations Beyond Your Website
AI systems may encounter your company through reviews, directories, industry publications, partner websites and other third-party sources.
Examples include review platforms and community discussions, which can strengthen brand presence. In practice, third party content often shapes how AI systems understand a category and a brand beyond the company website.
That makes PR, reputation and brand strategy relevant to AI visibility.
I’ve explored this broader relationship in Brand Strategy in the Age of AI Search.
Connect AI Visibility With Analytics
Use the Semrush AI Visibility Toolkit to track AI citations and mentions, topics and competitors.
Then use your website analytics and conversion data to understand what happens when visitors arrive, and compare citation tracking against analytics to see which cited URLs actually attract visits.
Don’t assume that more visibility automatically means more revenue.
AI Visibility Is Not the Same as Business Impact
The manufacturing study reinforces something I’ve been thinking about for a while.
AI visibility is useful, but we need to be clear about what each metric tells us.
Mentions show whether AI is bringing a brand into relevant conversations.
Citations show whether its content is being used as a source.
Referral traffic shows whether some users are moving from AI platforms to the website.
Conversions tell us whether those visits are contributing to measurable business outcomes.
These metrics are related, but they aren’t interchangeable.
The Brandesis data makes that distinction practical for me.
We have substantially more recorded citations than mentions. That gives us evidence of content visibility, but it also raises questions about brand recognition, topic relevance and whether the pages being cited are helping the business.
The answer isn’t necessarily to produce more content.
It may be to improve existing pages, strengthen the topics we want to own, develop better third-party signals or make the next step clearer for readers.
Being mentioned is useful. Being cited is useful. But understanding what each contributes to the business is where AI visibility becomes a strategy.
And with AI-generated answers becoming more common, I think that’s the measurement conversation marketers need to be having.
Frequently Asked Questions
What is the difference between AI mentions and AI citations?
An AI mention occurs when an AI-generated answer names or references a brand. An AI citation occurs when the answer references a source, such as a webpage, to support its information. A brand can be mentioned without its website being cited, and a website can be cited without the brand being named.
Are AI mentions more valuable than citations?
Neither is automatically more valuable. Mentions can support brand recognition and relevant topic associations, while citations can establish content visibility and create opportunities for referral traffic. Their value depends on the business’s objectives and the quality of the visibility.
Do AI citations increase website traffic?
AI citations can generate website visits when users follow source links, but they don’t guarantee clicks. The Semrush manufacturing study found that pages offering a clear next step, such as relevant product information or original research, attracted more AI traffic than generic reference content.
Why does my website have more AI citations than brand mentions?
AI platforms may use your content to answer questions without naming your business in the response. This can happen when a website provides useful information but the brand itself isn’t strongly associated with the topic or recommendation. Citations and mentions also use different counting methods, so the figures should not be compared as a direct ratio.
How can I measure AI mentions and citations?
Tools such as the Semrush AI Visibility Toolkit can help track brand mentions, citations, cited pages and topic visibility across supported AI platforms. You can combine this with manual AI searches and website analytics to evaluate how your brand is represented and whether AI referrals contribute to business outcomes.
Does AI visibility replace traditional SEO?
No. AI visibility adds another dimension to search measurement. Technical SEO, useful content, topical relevance, website experience and traditional search visibility remain important. AI search also brings greater attention to brand mentions, third-party sources and how a business is understood beyond its own website.
