Someone just asked ChatGPT about your brand.
Maybe they typed “is [your company] legit?” Or “how does [your brand] compare to competitors?” Or simply “[your brand] reviews.”
Here’s the uncomfortable part.
You had zero say in the answer.
The model pulled information from review sites you’ve never looked at, Reddit threads you’ve never read, editorial articles you’ve never seen, and competitor content you didn’t know existed.
And it delivered all of that to a potential customer as a clean, confident, conversational summary.
That’s how do LLMs talk about your brand in 2026. Not by visiting your homepage. Not by reading your About page. Not by following your messaging guidelines.
By synthesizing everything the internet says about you — and deciding what version of your story to tell.
This article breaks down exactly how that process works, what sources large language models actually pull from, and what you can do to influence the narrative before it gets written without you.
Why This Question Matters More Than You Think
42% of B2B decision-makers use an LLM as the first step in the buying process.
Let that sink in.
They’re not visiting your website first. They’re not reading your case studies first. They’re asking an AI. And that AI is forming a first impression of your brand based on what the internet has said about you — not what you’ve said about yourself.
I’m Brandon Leuangpaseuth. I help B2B brands take control of that narrative.LLMs act as trusted advisors for your target audience.
Users trust what they hear back. Generative AI compresses what used to be a long research funnel — comparison shopping, reading reviews, visiting multiple websites — into a two to four-hour process. Sometimes shorter.
And AI-driven traffic converts at 4x the rate of traditional organic search.
This isn’t a future problem. It’s happening now. The brands that understand how LLMs construct brand perception have a massive, compounding advantage over brands that are still only thinking about Google.
How Do LLMs Talk About Your Brand? The Research Answer.
Here’s what the data actually says.
A study analyzing 23,000+ AI citations across hundreds of branded queries revealed something most marketers didn’t expect.
When a user asks an LLM a question that includes a specific brand name, owned content — the content you control on your own website — accounts for just 23% of citations.
Nearly half (48%) comes from earned media. That’s editorial sites, independent media, forums, social media platforms, and review sites.
The remaining 30% comes from commercial brand content — listicles, comparisons, and reviews published by competitors and other brands with their own commercial interests.
In other words: when someone asks an AI about your company, most of what it references comes from outside your website.
This is what “brand gravity” actually means. AI systems don’t just pull from your own domain. They synthesize signals from across the ecosystem to decide what version of your brand is true.
The Four Categories LLMs Pull From
Let’s break down where AI outputs actually come from when your brand gets mentioned.
1. Your Owned Content (23% of Citations)
This is everything on your own domain.
Product pages. Blog posts. Landing pages. Service descriptions. Your homepage copy.
You control this completely. Which makes it the most obvious place to start optimizing for AI visibility.
But here’s the kicker — it’s also the smallest slice of the pie.
That’s not a reason to ignore it. Owned content performs especially well for specific types of queries. When users ask about product details, features, integrations, or pricing — that’s when LLMs are most likely to cite your site directly.
For functionality queries like “does [brand] include [feature]?”, owned content appears in citations 50% of the time. Make sure your site answers those questions clearly, specifically, and in plain language.
2. Editorial and Independent Media (16% of Citations)
This is third-party coverage from publications that exist to inform rather than sell.
Think industry blogs. Trade publications. Independent tech media. Journalism that covers your space.
If a journalist has written about your brand — positively or negatively — that article is shaping what AI says about you. If nobody has written about you at all, that absence is also shaping how LLMs represent your company.
This is why digital PR is not optional in an LLM SEO strategy. It’s one of the most direct ways to influence brand perception in AI-generated answers.
3. Forums and Social Media Platforms (11% of Citations)
Reddit. X. LinkedIn. Community forums. Discussion boards.
LLMs love opinionated content. They’re looking for perspective to help them form perspective. When there are real conversations happening about your brand — with multiple viewpoints, authentic language, and genuine user experience — AI systems pull from that heavily.
Reddit continues to be one of the top-cited sources across AI search. Not because AI prefers Reddit specifically. But because Reddit has the kind of frank, multi-turn conversational content that models are trained to recognize as authentic.
User-generated content on social media platforms is forming brand perception in AI whether you like it or not. The question is whether those conversations are working for you or against you. Read more about LLM seeding.
4. Review Sites (11% of Citations)
G2. Capterra. TrustRadius. Trustpilot. Yelp. Google reviews.
When users ask what customers think — “what do customers say about [brand]?” or “is [company] worth it?” — LLMs overwhelmingly cite earned media from review sites. Research shows earned media dominates this intent category 82% of the time.
Your review profile isn’t just a trust signal for humans anymore.
It’s raw material that AI systems use to construct your brand’s reputation in every branded query related to what customers think, whether you’re worth the investment, and how you compare to competitors.
How Citation Patterns Change Based on What the User Is Asking
This is the nuance most brands miss.
It’s not just about what sources LLMs pull from overall. It’s about how citation behavior shifts depending on user intent.
Competitor Comparison Queries
“Which is better — [your brand] or [competitor]?”
These queries pull from a wide mix of sources. Editorial content, Reddit threads, comparison blogs from other commercial brands, and sometimes your owned content if you’ve published comparison pages.
If a competitor has published a well-structured comparison article featuring your brand, that article is shaping what AI says when a buyer is evaluating both of you.
This is why creating your own comparison content — one that puts your brand alongside competitors with honest, specific differentiation — is one of the highest-impact things you can do for LLM SEO.
Customer Review Queries
“What do customers say about [brand]? Are reviews positive or negative?”
Earned media dominates here at 82% of citations. Review sites, social media platforms, forums.
Your owned content barely registers. The LLM is looking for what real customers say — not what you say about yourself.
This means your review presence on third-party platforms matters enormously. Not just the star rating. The content of the reviews. The specificity. The patterns.
AI systems are reading the narrative inside your reviews and surfacing a summary of what people actually think.
Product Detail and Functionality Queries
“Does [brand] include [feature]?” or “How does [product] work?”
This is where owned content performs best — appearing in 50% of citations.
Your product pages, feature documentation, and service descriptions are doing the heavy lifting for these queries. If they’re vague, poorly structured, or buried behind bad navigation, the AI can’t use them.
Write your product content the way a customer would ask about it. Then answer it directly.
Purchasing and Pricing Queries
“What does [brand] cost?” or “Is [brand] worth the price?”
These pull from a mix of owned content (if your pricing is clear and publicly accessible) and commercial brand content — comparison sites, affiliate content, and industry publications that cover pricing.
If your pricing page is a “contact us for a quote” dead-end, AI can’t tell buyers what you cost. And a competitor’s pricing comparison article will fill that gap instead.
The “Brand Gravity” Framework: What Really Controls Your AI Reputation
Brand gravity is the idea that your AI reputation isn’t built on any single source.
It’s built on the totality of what exists about you across the web.
Every time your brand is mentioned — in reviews, editorial coverage, social media posts, LinkedIn posts, forum threads, case studies, podcast transcripts, and competitor comparisons — it’s adding to or subtracting from your brand gravity.
LLMs synthesize all of those signals. They weight them based on source authority, content quality, and the context of the specific query.
Strong brand gravity means consistent, credible, accurate signals across multiple trusted sources. When AI pulls from that ecosystem, it tells a coherent story about your brand.
Weak brand gravity means inconsistent, thin, or negative signals. When AI pulls from that, it either can’t say much about you or says something you wouldn’t want your prospects to hear.
Here’s what drives brand gravity in AI systems:
- Consistency — Does your brand messaging say the same things across your site, your reviews, third-party coverage, and industry mentions? Inconsistency creates confusion. Confusion creates weak AI narratives.
- Coverage depth — How many distinct, high-authority sources reference your brand across different topics and categories?
- Sentiment signals — What’s the overall emotional tone of user-generated content and reviews about your brand? Balanced, honest reviews perform better than purely promotional ones.
- Competitive positioning — How does your brand appear in comparative contexts? Do you show up in the right conversations alongside the right competitors?
- Content freshness — Content less than 13 weeks old is cited significantly more than stale content. AI rewards brands that are actively creating, updating, and distributing fresh material.
How to Influence What LLMs Say About Your Brand
Now for the part that actually matters.
You can’t control AI outputs directly. But you can engineer the inputs — the sources LLMs pull from — to shift how your brand is represented in AI-generated answers.
Here’s how.
Own Your Foundation: Structured, Specific Content
Your website is 23% of the picture.
Don’t neglect it — but don’t mistake it for the whole game.
What you can do with your owned content:
Use clear, specific headings structured around the questions your audience actually asks. Not “Our Services” — but “What Does an LLM SEO Agency Actually Do?”
Write product descriptions and service pages that mention your brand’s features, compare-worthy attributes, and use-case contexts explicitly. LLMs read these in chunks. Each section needs to stand alone as a complete, usable answer.
Add FAQ sections. Structured Q&A is the native language of AI systems. They’re trained on conversational patterns. When your content mirrors those patterns, AI can use it.
Make your pricing transparent. If pricing information doesn’t exist on your site, AI fills that gap with whatever third-party sources have said about your costs.
Keep content fresh. Revisit your most important pages at least quarterly. Substantive updates — not surface changes — signal relevance to AI systems.
Build Earned Media Intentionally
16% of LLM citations in branded queries come from editorial and independent media.
If you’re not actively pursuing coverage in publications your target audience trusts, you’re leaving a large portion of your AI brand narrative uncontrolled.
Practical steps:
Get featured in industry publications. Even niche ones. A well-placed article in a specific trade publication carries more weight in relevant branded queries than a mention in a generic tech blog.
Pursue guest contributions. When you publish under your name in respected editorial outlets, you’re creating a high-authority source that LLMs associate with your brand.
Earn podcast coverage. Transcripts and show notes from podcast appearances get picked up by AI systems. Your words in an interview become part of what AI says about you.
Syndicate quality content. Companies like Zapier have used content syndication through outlets like Stacker to push their material across dozens of news publishers — multiplying the reach of a single piece of content and creating multiple citation-worthy sources.
Take Reddit and Forums Seriously
User-generated content on forums is shaping brand perception in AI and most marketers treat it as an afterthought.
Find the Reddit communities and industry forums where your target audience asks questions about your category. Contribute genuinely. Add value to real conversations. Use your actual name or brand name.
When LLMs see your brand mentioned in authentic community discussions — especially alongside competitors, with nuanced language and real opinions — that counts as a trusted signal.
This isn’t about gaming Reddit. It’s about actually being present in the conversations where your buyers are forming opinions.
Manage Your Review Profile Like a Marketing Channel
Reviews are 11% of branded citations overall. For customer review queries, they’re 82% of what AI pulls.
This isn’t a passive thing. Your review profile is active marketing.
Encourage specific, detailed reviews. Generic five-star reviews (“Great company!”) carry less weight in AI outputs than balanced, attribute-specific ones (“The onboarding was fast and the support team resolved our issue in 24 hours, though the reporting dashboard took some getting used to”).
AI systems learn from specificity. Reviews that speak to concrete features, use cases, and outcomes give models the context they need to accurately represent your brand.
Respond to negative feedback publicly. Your responses to negative reviews are part of the narrative too. Thoughtful, professional responses to criticism signal that your brand takes quality seriously — and that signal gets picked up.
Create Comparison Content
Comparative content performs extremely well in LLM citations — especially for competitor comparison and evaluation queries.
This means building honest, detailed content that puts your brand alongside alternatives. Highlight where you win. Acknowledge where others might be a better fit for different use cases.
AI models trust balanced comparisons more than promotional copy. And when your brand owns the comparison content, you control the framing in what’s arguably the highest-stakes query category — when a buyer is actively evaluating you against a competitor.
How to Track What AI Is Already Saying About You
Before you can improve your brand perception in AI, you need a baseline.
Here’s a practical starting framework.
- Step 1: Build a prompt library. Write down 20–30 questions your ideal customers would actually type into ChatGPT, Perplexity, or Claude. Include branded prompts (“is [your brand] good for [use case]?”), comparison prompts (“best [category]: [your brand] vs. [competitor]”), and review prompts (“what do customers say about [your brand]?”).
- Step 2: Run the prompts manually. Do this across at least ChatGPT, Perplexity, and Google AI Overviews. Document what comes up. What sources are cited? What language is used? What’s the overall sentiment? Does the answer match what you’d want a prospect to hear?
- Step 3: Identify the gaps. Where is the AI pulling from that you don’t control? What’s missing that you wish were there? Which attributes of your brand are being described accurately — and which are being described in ways that would cost you a sale?
- Step 4: Track referral traffic from AI platforms. In GA4, filter by source to find traffic from chatgpt.com, perplexity.ai, claude.ai, and similar AI platforms. Even if click volume is low, these visitors convert at dramatically higher rates than organic traffic.
- Step 5: Monitor consistently. AI outputs shift. Model updates happen. Sources get added or dropped. Run your core prompt library at least monthly to track changes and catch negative shifts early.
Tools like Profound, Peec AI, and Semrush’s AI Visibility Toolkit make this process faster. But even a manual tracking sheet is a significant improvement over flying blind.
How Brandon Leuangpaseuth, LLC Shapes Brand Perception in AI
Understanding how LLMs talk about your brand is one thing.
Building the strategy to improve what they say — and then executing it — is another.
That’s where an experienced LLM SEO specialist makes the difference.
That’s where I come in.
I’m a growth marketer and LLM specialist who works directly with B2B SaaS companies and Y Combinator-backed startups to build AI search visibility from the ground up — covering every layer of what drives brand perception in LLM outputs.
–and I run one of the top LLM seo agencies.
What Working With Me Actually Looks Like
Most brands have no idea how they’re being represented in AI search.
When we start working together, the first step is building that baseline. Running your brand through the major AI platforms. Mapping which sources are driving your AI narrative. Identifying where the gaps are, where the narrative is off, and where competitors are filling space you should own.
From there, the work covers:
- Owned content optimization — restructuring your site’s most important pages for LLM readability. Clear headings, FAQ sections, entity-rich descriptions, structured data, and fresh content that gives AI exactly what it needs to accurately represent your brand.
- Earned media strategy — building the editorial, digital PR, and community presence that accounts for nearly half of what LLMs cite in branded queries. Industry publications, podcast appearances, contributor placements, and strategic forum participation.
- Comparison content development — creating the comparison and evaluation content that shapes how AI answers the queries that matter most: when a buyer is deciding between you and a competitor.
- Review profile management — building a review presence on the platforms AI trusts, with the specificity and balance that AI systems prefer over generic five-star ratings.
- AI visibility tracking — ongoing monitoring of how your brand appears in AI outputs across ChatGPT, Perplexity, Gemini, Google AI Overviews, and more.
Why You Need a Growth Marketer, Not Just an SEO Agency
This isn’t traditional SEO work–you need a specialized LLM SEO agency.
It requires someone who understands content strategy, digital PR, brand positioning, conversion outcomes, and how AI systems actually retrieve and synthesize information.
I bring that growth marketing mindset to every engagement — with direct experience from multi-million dollar companies and YC-backed startups including Keeper Tax and EasyLlama.
No hand-offs. No junior team members running your campaigns. You work directly with me — a strategist who has built these systems inside fast-moving companies and can apply them directly to yours.
If you want to understand how AI is talking about your brand — and build the strategy to improve it — let’s talk.
The Bottom Line: Your AI Narrative Is Already Being Written
Here’s the thing most brands don’t want to hear.
LLMs are already talking about you.
Right now, someone is asking ChatGPT whether your company is worth it. Someone is asking Perplexity how you compare to a competitor. Someone is asking Google AI Overviews what your customers actually think.
And AI is answering.
The question isn’t whether you have an AI reputation. You do. The question is whether it’s accurate, whether it’s positive, and whether it’s driving buyers toward you or away.
Most brands have no idea what AI is saying about them. No tracking. No baseline. No strategy.
That’s the gap.
And right now — before competitors figure this out — it’s also the opportunity.
Frequently Asked Questions About How LLMs Talk About Your Brand
How do LLMs decide what to say about a brand?
LLMs construct brand narratives by synthesizing information from multiple sources across the web — your owned content, editorial media, review sites, social media platforms, forums, and competitor-authored content. They don’t rely on a single source. The narrative is built from patterns across all of it.
Can I control what AI says about my brand?
Not directly. But you can influence the inputs. By optimizing your owned content, building earned media coverage, managing your review profile, and participating in relevant community conversations, you shape the sources that AI pulls from — which shapes the output.
Why does my brand barely show up in AI answers even though I rank well on Google?
Traditional Google rankings and AI citations have less overlap than most people assume. Research from Profound shows only about 39% overlap between the queries driving SEO and those driving AI citations. Ranking on Google helps, but it’s not the whole story. LLM-specific ranking factors and optimization is required.
How often do LLMs update the information they have about brands?
For AI tools that use live web retrieval — like Perplexity and Google AI Overviews — the information can update very quickly. Content that’s less than 13 weeks old is significantly more likely to be cited than older content. Major model updates can also shift how brands are represented overnight.
What’s the most important thing I can do to improve my brand’s AI reputation?
Start by understanding your current baseline. Run your brand name through the major AI platforms and track what comes up. Then focus on the sources you can influence — starting with your owned content and review sites, and expanding to earned media and community presence over time.
