Voice, visibility and value: How PR becomes critical in LLM-dominated media
As artificial intelligence becomes more involved in how people find and consume information, the traditional role of PR is becoming more important. Citations, citations, citations… This is what we are hearing everywhere, and while as SEO we got used to getting links, now comes the time to get mentions as well. Welcome to the world of AI Search.
This article explains how large language models (LLMs) are influencing what people see, hear, and believe about brands. It focuses on how these changes affect visibility, trust, and influence. Three areas where PR plays a direct role.
Why are LLMs making public relations even more important
Large language models (LLMs) are a type of artificial intelligence trained on massive datasets. They process and generate human-like text based on patterns in online content. Examples include OpenAI’s ChatGPT, Google’s Gemini, and Meta’s LLaMA.
LLMs don’t return a list of links like search engines. Instead, they generate direct answers. These responses pull from news articles, blogs, social media, and forums. The content often reflects the tone and frequency of what they’ve learned.
The adoption of LLM-based tools is growing quickly. In 2023, ChatGPT reached 100 million users faster than any other consumer application in history. Google now includes AI Overviews in many search results, changing how people find information.
This creates new challenges for brands:
– Visibility challenge: If a brand isn’t mentioned in relevant, trustworthy content, it may not appear in AI-generated responses
– Context matters: LLMs rely on patterns across multiple sources rather than single webpages
– Zero-click information: Users get answers without visiting websites
For PR professionals, this means earned media and brand mentions are becoming even more valuable. The content that feeds AI responses often comes from the very sources PR has always targeted: news outlets, industry publications, and authoritative websites.
From traditional SEO to brand influence
Search engine optimisation (SEO) has traditionally focused on keywords, backlinks, and technical elements to help websites rank higher. These methods worked when search engines returned lists of links.
LLMs work differently. They examine patterns in language, context, and source credibility. They prioritise content from trusted sources rather than ranking pages based on keywords or backlinks.
This changes the game for digital visibility:
1. Mentions matter more than links
2. Authority of sources outweighs quantity of references
3. Context and relevance determine inclusion in AI responses
In this environment, earned media such as news articles, interviews, and expert commentary become more influential than technical SEO elements. A skincare brand featured in a health magazine article about sensitive skin is more likely to be included in an AI-generated response than a brand that only appears on its own well-optimised website.
How LLMs impact visibility and brand mentions
LLMs gather information through pattern recognition across large volumes of content. They learn from news, forums, academic sources, and other publicly available content.
Unlike traditional search engines, LLMs rarely show a list of websites. They generate answers directly in the interface, often called “zero-click search.” In this model, users don’t click through to websites. If a brand isn’t included in that output, it may not be seen at all.
This move means PR teams must focus on how widely and consistently a brand is referenced in external, credible contexts. The goal is no longer just to drive traffic to a website, but to ensure the brand is part of the conversation that LLMs draw upon when generating responses.
Optimising content for AI-driven search
Craft press releases with clear data
Press releases work better for LLMs when they include clear, factual information. Using plain language, short paragraphs, and bullet points for key data helps AI systems understand and extract information accurately.
Before:
Brand X, a leading innovator, is thrilled to announce a new product coming soon.
After:
Brand X will launch the Model Z1 Smartwatch on August 15. The product will be available at £299 in selected UK retailers and online. The Model Z1 offers fitness tracking, heart rate monitoring, and GPS navigation.
The second version provides specific details that LLMs can extract and use when answering questions about new smartwatches, pricing, or Brand X’s products.
Structure evergreen content with context
Evergreen content (material that remains relevant over time) works best for LLMs when it includes full explanations and clear definitions. This helps the model understand how topics connect to related ideas.
For luxury brands, this means creating content that explains concepts fully rather than assuming knowledge. Instead of simply writing “sustainable fashion,” include a sentence like “Sustainable fashion refers to clothing designed and produced with minimal environmental impact, often using recycled or organic materials.”
Embrace semantic keywords for LLMs
Semantic keywords are words and phrases related to a topic, not just exact match terms. LLMs interpret meaning based on the surrounding context and related terms.
For a B2B cybersecurity software company, relevant semantic keyword clusters might include:
- Threat detection, vulnerability assessment, and security protocols
- Enterprise-grade compliance standards, risk management
- Network monitoring, incident response, and data protection
Using these terms naturally in content helps LLMs associate your brand with relevant topics when generating responses about cybersecurity solutions, enterprise security, or specific threat mitigation needs.
Bridging PR and SEO for maximum reach
Public relations and search optimisation are becoming increasingly connected, maybe there is a world in the near future where PR agencies will merge with digital marketing agencies, who knows. Because when these functions work together, brands can create content that serves both human readers and AI systems.
Effective collaboration between PR and SEO teams includes:
– Sharing keyword research to inform press release language
– Aligning on messaging that supports both search visibility and brand positioning
– Creating content calendars that address both news cycles and search trends
This alignment ensures that press releases, blog content, and media pitches all support the same narrative and semantic signals, making it easier for LLMs to understand what the brand represents.
Tracking the new KPIs in an LLM world
Track brand mentions and backlinks
In LLM-driven search, brand mentions carry more weight than backlinks. LLMs analyse text patterns and frequency of mentions to determine relevance.
Tools like Brandwatch, Meltwater, and Talkwalker scan digital content for brand references. They log where and how often a brand is mentioned and in what context.
The focus changes from “how many links did we get?” to “where is our brand being mentioned, and in what context?”
Measure the share of voice in AI outputs
Share of voice in AI outputs refers to how often a brand appears in answers generated by tools like ChatGPT or Google AI Overviews. This is different from the traditional media share of voice.
To measure this, PR teams can:
– Enter relevant industry queries into multiple LLMs
– Record which brands are named in the responses
– Track frequency of inclusion and positioning within responses
Nonetheless, the above is quite manual, and there are tools out there that allow you to do this at scale. We use Otterly, and soon Accuranker will also give this information. If you’d like to know more about measurement, you can have a look at our LLM performance tracking solutions.
This provides insight into how visible a brand is in the AI-generated content that increasingly shapes consumer perceptions.
Evaluate sentiment in AI-generated responses
LLMs reflect how brands are discussed across the web. If the sentiment is mostly positive, AI responses will likely be positive too. If negative content dominates, AI outputs may reflect that tone.
Regular monitoring of AI responses can reveal sentiment trends and help PR teams address potential issues before they become widespread in AI-generated content.
Building trust and authority in AI responses
Provide reliable data sources
LLMs use public information to generate answers. They’re more likely to use content from accurate, consistent, and widely trusted sources.
To be recognised as reliable, you can:
– Publish data that’s clearly presented and easy to verify
– Cite sources for claims and statistics
– Use consistent terminology across all communications
– Provide dates, figures, and context
Content that avoids exaggeration and presents facts in a structured format is more likely to be included in an LLM’s response.
Maintain consistent messaging across channels
LLMs examine how topics are discussed across multiple platforms. If your brand’s messaging is inconsistent, using different claims or facts in various places, the model may not understand what the brand represents.
Consistency in messaging involves using the same key information, tone, and terminology across press releases, website content, social profiles, and interviews. This provides the model with a clearer understanding of the brand’s identity.
Driving real business value through PR
Public relations strategies that consider LLMs can support business outcomes in several ways:
1. Brand awareness: When a brand consistently appears in AI responses to relevant queries, it reaches consumers who might never have encountered it through traditional search
2. Trust building: Brands mentioned by AI systems as industry leaders or experts benefit from the perceived objectivity of the AI
3. Consideration: Detailed information about solutions or services in AI responses can help buyers evaluate options without visiting multiple vendor websites
For example, a cloud infrastructure company that consistently appears in trade publications about digital transformation is more likely to be cited when someone asks an AI, “What are the best enterprise cloud migration solutions?” The AI references what it has learned from multiple sources, and brands with strong thought leadership presence are more likely to be included.
Preparing your PR strategy for the future
As AI continues to change how people find information, PR teams can take several steps to adapt:
1. Review how your brand appears across high-authority platforms
2. Ensure press releases include clear data and context
3. Monitor AI-generated responses about your industry and brand
4. Build relationships with publications that influence AI training data
The goal is to ensure your brand is visible not just on search engine results pages but in the direct answers AI systems provide consumers.
At Passion Digital, we help ambitious brands navigate this changing landscape with data-driven strategies that ensure visibility in both traditional and AI-powered search environments.
FAQs about LLM-driven PR
- How can brands correct misinformation generated by AI?
To correct AI misinformation, publish accurate information across authoritative platforms like company websites, press releases, and reputable news sources. LLMs learn from patterns in public content, so consistent, corrected data in multiple locations will influence future outputs.
- How are LLMs changing the value of traditional media coverage?
Traditional media coverage remains valuable because LLMs rely heavily on content from trusted sources. Articles from established news outlets often serve as training data for AI-generated responses, making quality media placements even more important for brand visibility.
- How quickly do LLMs incorporate new information about brands?
Most commercial LLMs update their knowledge periodically rather than in real-time. Updates can take weeks or months to appear in responses, which means consistent, long-term PR strategies are more effective than one-off campaigns for influencing AI outputs.
- What types of content are most influential for LLM visibility?
Factual, clearly structured content from authoritative sources has the most influence on LLM outputs. This includes news articles, expert interviews, industry reports, and detailed product information published across multiple trusted platforms.