⚡ TL;DR: Brand mentions now outweigh backlinks for AI search visibility. Ahrefs' study of 75,000 brands found web mentions have a 3x stronger correlation with AI Overview citations than backlinks (0.664 vs 0.218). Co-citations, where your brand appears alongside competitors in roundups and reviews, are the new link-building. A GA4 AI chatbot lets you chat with Google Analytics to track how co-citation strategies shift your referral traffic, and an AI Google Analytics assistant like Meaning makes measuring this impact effortless.

Co-citations are instances where your brand is mentioned alongside other brands in the same piece of content, such as "best of" roundups, comparison reviews, or community discussions. Brand mentions are any references to your brand name across the web, whether or not they include a hyperlink. In the era of AI search, these signals now carry far more weight than traditional backlinks when it comes to getting recommended by AI engines. With natural language analytics, you can track exactly how co-citation efforts translate into real visits, conversions, and revenue inside Google Analytics 4.

Backlinks have been the currency of search visibility for two decades, but AI search engines operate on a fundamentally different model. Google's PageRank algorithm treated every link as a vote of confidence, and SEO strategies revolved around earning (or manufacturing) those votes.

AI search engines work differently. When ChatGPT, Google's AI Overviews, or Perplexity generate a response, they don't count links pointing to your site. They synthesise information from across the web, and the brands they recommend are the ones they've encountered most frequently, most consistently, and in the most authoritative contexts.

This isn't speculation. Ahrefs analysed 75,000 brands and measured which factors correlate with appearing in AI Overview responses. The results were unambiguous:

  • Branded web mentions: 0.664 correlation (strongest factor)
  • Branded anchor text: 0.527 correlation
  • Branded search volume: 0.392 correlation
  • Backlinks: 0.218 correlation
  • Domain Rating: 0.130 correlation

The implication is clear: your brand's presence across the web matters far more than links to your website. As Ahrefs put it, brands earning the most web mentions receive up to 10x more mentions in AI Overviews compared to the next closest quartile.

If you're new to the broader discipline behind this shift, our introduction to Generative Engine Optimisation covers the fundamentals.

What are co-citations and why do they matter?

Co-citations are the AI-era equivalent of backlinks: third-party signals that tell search engines your brand belongs in a given category. The concept is borrowed from academic research. In scholarly work, two papers are "co-cited" when a third paper references both. The more frequently two papers appear together in reference lists, the stronger their perceived relationship.

In AI search, co-citation works similarly. When your brand is consistently mentioned alongside established competitors in "best of" roundups, comparison articles, review platforms, and community discussions, AI models learn that your brand belongs in that category.

Think of it this way: if a hundred articles about "best project management tools" mention Asana, Monday.com, and your product together, AI engines develop a strong association. When someone asks ChatGPT for project management recommendations, your brand has a seat at the table, not because you have the most backlinks, but because you've been discussed in the right contexts.

This is fundamentally different from traditional link building:

Traditional SEOAI Search (GEO)
Links from site A to site BMentions of brand B on site A
Anchor text optimisationContextual co-occurrence with peers
Domain authority flows through linksBrand authority builds through citations
One link = one voteRepeated mentions across contexts = pattern recognition

We explored how structuring your content influences AI citations in our content structure guide. Co-citations are the off-site complement to that on-site work.

The data: where AI engines pull brand signals from

AI engines draw brand signals from a diverse range of platforms, with some carrying significantly more weight than others. Understanding which sources matter most helps you prioritise your co-citation efforts.

YouTube: the citation powerhouse

YouTube has emerged as a dominant source for AI citations. According to BrightEdge, AI search engines cite YouTube 200 times more frequently than any other platform. SE Ranking found YouTube makes up 4.43% of all AI Overview citations, and Search Engine Land reports that 29.5% of Google AI Overviews cite YouTube, making it the top domain overall.

Ahrefs' expanded study (December 2025) confirmed this: YouTube mentions show the strongest correlation with AI visibility at approximately 0.737, even surpassing general web mentions.

For brands, this means a presence on YouTube isn't optional. Product reviews, tutorials, and comparison videos that mention your brand create citation signals that AI engines weight heavily.

Reddit: authentic discussion as signal

Reddit has become a surprisingly influential source for AI-generated answers. Data from Profound shows that AI systems pull from Reddit for both positive (5% of citations) and negative (6.1% of citations) brand sentiment. Google AI Overviews cite Reddit in approximately 2.2% of responses, while Perplexity draws from it in 6.6%.

What makes Reddit particularly powerful is its perceived authenticity. AI models aren't looking for marketing copy. They're looking for genuine evaluation. A Reddit thread comparing your product to competitors, with real users discussing pros and cons, carries more weight than a polished landing page.

Review platforms: G2, Trustpilot, and Capterra

Review sites play an outsized role in AI recommendations, particularly for B2B software. G2's own analysis of 30,000 citations found that a 10% increase in reviews correlates with a 2% increase in AI citations. Between one-third and three-quarters of all review-site citations in AI answers come from G2 specifically, far surpassing Capterra, TrustRadius, and Trustpilot.

These platforms are especially influential at the bottom of the funnel. When someone asks an AI assistant "What's the best analytics tool for small businesses?", the AI draws heavily on structured review data to formulate its recommendation. If you want to understand how your brand performs in this landscape, tracking the right GA4 metrics alongside your co-citation data gives you the full picture.

News, blogs, and industry publications

Traditional media and industry publications remain important co-citation sources. Being featured in a TechCrunch roundup, a niche industry comparison, or a respected blog's "tools we use" post creates the kind of authoritative mention that AI models trust.

The key insight from our article on entity clarity applies here: your brand needs to be mentioned in ways that clearly communicate what you do. A passing mention in an unrelated article carries less weight than a contextual mention in a relevant comparison.

Case study: how Tally.so made ChatGPT their #1 referral channel

Tally.so's success is one of the clearest real-world examples of co-citations driving AI referral traffic at scale. According to SimilarWeb data analysed by Qwairy:

  • Tally.so receives approximately 9 million visits worldwide
  • 19% of visits (~2 million) come from referrals
  • ChatGPT alone accounts for 10% of all referral traffic, sending over 3,000 qualified leads per week

How did they achieve this? Not through a massive backlink campaign, but through pervasive brand mentions:

  1. Consistent presence in "best form builder" roundups across dozens of publications
  2. Active Reddit discussions where real users recommend Tally in relevant threads
  3. YouTube tutorials and comparisons featuring Tally alongside Typeform and Google Forms
  4. G2 and Product Hunt reviews establishing credibility and structured data
  5. A clear, differentiated value proposition ("free form builder") that AI models can easily categorise

Tally didn't set out to "optimise for AI." They built a brand presence so thorough that AI engines couldn't ignore them. The lesson: co-citations at scale, across diverse platforms, compound into AI visibility.

If you're tracking traffic with Meaning, you'd notice this shift as a growing "referral" segment from AI domains, a clear signal that your brand mention strategy is working. Think of it as ChatGPT for Google Analytics: just ask what's changed in your referral traffic and get a plain-English answer.

The co-citation audit framework

Before building new co-citations, you need to understand your current position. A structured audit reveals where you stand relative to competitors and highlights the gaps worth closing first.

Step 1: map your current co-citation network

Search for your brand name (in quotes) across:

  • Google: "your brand" + "best [category]", count how many roundups include you
  • YouTube: Search "your brand" vs or "your brand" review, note comparison videos
  • Reddit: Search your brand name on Reddit, identify threads where you're discussed
  • G2/Capterra/Trustpilot: Check your listing completeness and review count
  • News/blogs: Use Google News search for recent brand mentions

Step 2: identify your co-citation peers

List every brand that appears alongside yours in these contexts. These are your co-citation peers. Note:

  • Which competitors appear more frequently than you?
  • Which contexts (roundups, reviews, discussions) are you missing from?
  • Are there adjacent categories where you should appear but don't?

Step 3: audit mention quality

Not all mentions are equal. Evaluate each mention for:

  • Context relevance: Is the mention in a relevant category discussion?
  • Sentiment: Is your brand discussed positively, neutrally, or negatively?
  • Recency: AI models weight recent content more heavily
  • Source authority: Is the mention on a high-authority platform?
  • Entity clarity: Does the mention clearly explain what your brand does? (See our entity clarity deep-dive for why this matters.)

Step 4: score and benchmark

Create a simple scorecard:

  • Total brand mentions (Google: "your brand" -site:yourdomain.com)
  • Roundup inclusions (count of "best of" / comparison articles featuring you)
  • Review platform presence (total reviews across G2, Capterra, Trustpilot)
  • Reddit mention count (unique threads mentioning your brand)
  • YouTube mention count (videos mentioning your brand by others)

Compare these numbers against your top 3 competitors. The gaps reveal your priorities.

The co-citation building playbook

Earning co-citations requires a systematic, multi-channel approach. The six strategies below are ordered by impact, starting with the highest-leverage activities.

1. earn roundup inclusions

"Best of" and comparison articles are co-citation gold. To earn inclusions:

  • Identify existing roundups in your category that don't include you
  • Reach out to authors with a genuine pitch explaining your differentiation
  • Create comparison content on your own site (e.g., "Brand X vs Brand Y vs Us"), as this often triggers inclusion in others' comparisons
  • Monitor new roundups being published and pitch for inclusion early

2. build your YouTube presence

Given YouTube's 0.737 correlation with AI visibility:

  • Create tutorial and demo content: instructional videos are cited significantly more frequently, according to research by Neil Patel
  • Encourage reviews by sending product access to YouTube creators in your niche
  • Participate in comparison videos: even being featured as one option among five builds co-citation
  • Optimise video titles and descriptions with clear brand + category keywords

3. cultivate Reddit authentically

Reddit discussions carry weight precisely because they're authentic. Don't spam:

  • Participate genuinely in subreddits relevant to your category
  • Answer questions where your product is a legitimate recommendation
  • Encourage satisfied users to share their experience (without incentivising)
  • Monitor mentions and engage constructively with criticism

4. invest in review platforms

For B2B brands especially, review platform presence is non-negotiable:

  • Claim and complete your G2, Capterra, and Trustpilot profiles
  • Actively solicit reviews from happy customers (G2's data shows a direct correlation between review volume and AI citations)
  • Respond to reviews: engagement signals legitimacy
  • Keep profiles updated with current features and screenshots

5. pursue strategic PR and guest content

Earned media creates high-authority co-citations:

  • Pitch industry publications with data, insights, or unique angles
  • Contribute expert commentary to journalists covering your space
  • Write guest posts for respected blogs in your niche
  • Participate in podcasts: transcripts become searchable co-citations

6. monitor and iterate

Co-citation building isn't a one-off project. Set up ongoing monitoring:

  • Google Alerts for your brand name and key competitors
  • Monthly audit using the framework above
  • Track AI referral traffic: Meaning can help you spot when AI-driven organic traffic shifts, showing whether your co-citation efforts are translating into actual visits
  • Quarterly competitive benchmarking to measure progress against peers

For a deeper look at which metrics to track for generative search, see our GEO performance metrics guide.

What this means for your analytics

Co-citation strategies change your traffic patterns in measurable ways. As AI search reshapes how users discover brands, traditional analytics tell an incomplete story. You might see:

  • Declining direct Google organic clicks (users getting answers from AI Overviews without clicking through)
  • Growing referral traffic from AI domains (chat.openai.com, perplexity.ai)
  • Increased branded search volume (users hearing about you from AI, then searching directly)
  • Shifts in conversion patterns (AI-referred users may convert differently)

This is where having conversational access to your analytics data becomes valuable. Rather than digging through dashboards to correlate these signals, you could ask Meaning a natural language question like "How has my referral traffic from AI sources changed this quarter?" and get an immediate, contextual answer. It's Google Analytics 4 made easy: ask a question, get an insight, act on it.

The brands that win in AI search won't just optimise. They'll measure, learn, and adapt. As we explored in our nine proven GEO techniques article, the discipline is still young, and the feedback loop between strategy and measurement is critical.

Frequently asked questions

The data strongly suggests so. Ahrefs' study of 75,000 brands found branded web mentions have a 0.664 correlation with AI Overview visibility, compared to just 0.218 for backlinks. This doesn't mean backlinks are worthless. They still matter for traditional search rankings, which indirectly feed AI visibility. But for direct AI citation, brand mentions carry roughly three times the weight.

How do I know if AI engines are already recommending my brand?

The simplest test: ask ChatGPT, Perplexity, and Google's AI mode questions your ideal customers would ask (e.g., "What's the best [your category] tool?"). Note whether your brand appears. For systematic tracking, tools like Ahrefs Brand Radar or Otterly.ai monitor AI mentions across platforms. On the traffic side, a GA4 AI chatbot like Meaning lets you chat with Google Analytics and surface AI referral trends instantly without building custom reports.

Does negative sentiment on Reddit or review sites hurt my AI visibility?

It can. Profound's research shows AI engines pull from Reddit for both positive and negative brand sentiment at roughly equal rates. If negative discussions dominate, AI models may reference those concerns when recommending alternatives. The solution isn't to suppress criticism but to ensure the volume of positive, authentic discussion outweighs it.

How long does it take for new brand mentions to influence AI recommendations?

There's no fixed timeline, but most AI models update their training data periodically and supplement with real-time retrieval (RAG). Mentions on high-authority, frequently crawled sites (like YouTube, Reddit, or major publications) tend to surface faster, often within weeks. Building a consistent co-citation presence over three to six months is a realistic horizon for seeing measurable shifts in AI visibility.

Should I focus on one platform (e.g., YouTube) or spread efforts across many?

Both matter, but diversification is more defensible. YouTube shows the strongest single-platform correlation (0.737), so it deserves priority investment. However, AI models synthesise across sources. A brand mentioned on YouTube, Reddit, G2, and industry blogs presents a more robust signal than one dominating a single platform. Start where you're weakest, then build breadth.

Traditional link building asks: "Will you link to my site?" Co-citation building asks: "Will you mention my brand in context?" The distinction matters because AI engines parse mentions and context, not just hyperlinks. A roundup article that lists your brand name alongside competitors, even without a link, still builds your co-citation profile. This shifts outreach from requesting links to requesting inclusion in relevant conversations.

What's next in this series

This article is part five of our seven-part series on Generative Engine Optimisation:

  1. What Is GEO?
  2. 9 Techniques That Boost AI Search Visibility
  3. How to Structure Content So AI Engines Actually Cite You
  4. Entity Clarity: Why AI Can't Recommend You If It Doesn't Understand What You Do
  5. Co-Citations and Brand Mentions: The New Backlinks for AI Search (you are here)
  6. Measuring GEO: The Metrics That Actually Matter
  7. The Complete GEO Playbook: Putting It All Together (coming soon)

Meaning is an AI-powered Google Analytics chatbot by Hivory. Ask your analytics data questions in plain English and get instant, actionable answers. Try Meaning →