As Australians’ online searching habits have changed, AI has become a priority channel for organisations looking to maintain or improve their Share of Voice in the market.
Traditional search behaviour, which might have begun by typing a string of keywords like “low-interest rate credit cards” together into a search engine has been traded in for more complex, conversational research where users are looking for rankings, recommendations, pros and cons, and up-to-date context.
Large Language Models (LLMs) like ChatGPT and Gemini provide these conclusions in seconds, and are part of everyday life for 77% of the 16+ population in Australia. That proportion is growing, specifically by 6 million from June 2025 to May 2026. Even as trust lags, usage is booming.
As a result, PR, marketing and communications teams are increasingly being asked questions from the top like “Are we being mentioned by AI more than our competitors?”
But while ‘AI Visibility’ has become a recognisable buzzword in the PR and communications world, teams are struggling to establish a workflow that actually helps better work get done. In this article, our team explains what AI Visibility is, what low visibility can cost an organisation, how to conduct an audit, and how to start building your brand’s presence.
When you search for a product or service on an AI platform like ChatGPT, which is the most popular AI tool in Australia, the brands that are mentioned in the response have earned visibility. However, there are different aspects of that visibility.
Where is information being sourced from? Was there something negative mentioned about the brand? Altogether, AI Visibility is the extent to which a brand appears in AI-generated answers across different platforms.
There are three dimensions to consider:
| Dimension | Question |
| Presence | Does the brand appear at all? |
| Prominence | Is it mentioned more or less than its competitors? |
| Framing | Is it described positively, neutrally, or negatively? |
Measuring AI Visibility is not simply asking the question of “Are we visible?” Similar to reputation management, the question that’s actually being asked is “What version of our brand is being presented to AI?”
When LLMs search for information to answer a question, they may draw on:
The exact mixture you end up with varies by platform and query, with domains prioritised based on common qualities like high domain authority, content freshness, and clarity of structure.
Each brand will have their own source ecosystem for different AI platforms. For ChatGPT responses about your brand, you may see your owned website as the top-cited domain. For Claude, it might be an article from ABC News.
Regardless, when you measure AI Visibility and drill down into how AI is sourcing information about your brand, you’re testing the strength and breadth of the different media types you’ve invested into within your communications strategy.
As with traditional search rankings, brand performance in AI-generated answers has real business consequences. This might look like website traffic loss where your owned website could have been cited, or even churn with current clients researching their options. To move away from an abstract idea of ‘AI Visibility’ and into the costs of avoiding it, there are four risks we can look to.
If your brand is absent from category and comparison responses from AI, your competitors become the default recommendation.
For recommendation-based queries, this means missing out potential business, and the risk here applies to every stage of the customer journey. Whether a searcher is new to your business, a current customer considering alternatives, or a lost customer, LLMs are being used to synthesise the most up-to-date information about your brand and its competitors.
Because each AI platform has a different method of prioritising sources, your brand can be represented differently across ChatGPT, Gemini, and others. As a result, high visibility and positive sentiment from one LLM doesn’t necessarily apply to another.
While overall low visibility exposes your brand to the most risk, isolated visibility issues on specific platforms also problematically segments your audience depending on what tools are being used. Because AI adoption still has plenty of room to diversify, this risk is also a volatile one.
Without knowing the sources that LLMs are using to describe your brand to customers, your organisation may be unaware of years-old media that continues to be surfaced in AI responses.
Outdated earned media can reproduce false information on current leadership teams, pricing, discontinued products, historic controversies, among other things.
In Australia and New Zealand, 41% of consumers use AI to search for personalised product recommendations. This means your most important commercial touchpoint may be happening before a customer reaches your website, within an LLM. Missing visibility on this channel means missing out on speaking to highly engaged members of your target audience, likely close to a purchase decision.
If your team is starting to look into AI Visibility as a priority metric, you'll want to run a quick audit. After, you'll have an idea of how your brand appears to different LLMs, and how information is being sourced.
Start with a comprehensive list of sample prompts, or queries, about your industry and/or brand that someone might ask an LLM. For example:
You can choose to include or exclude mentions of your brand specifically. However, by excluding your brand’s exact name, you’ll be able to see a more accurate Share of Voice result.
Within a tool like Streem’s AI Visibility Dashboards, aim to compare results between at least ChatGPT and Gemini, which earn most LLM traffic globally. Then, add additional LLMs to track depending on your audience’s demographics and potential search behaviour. In our platform, we measure performance on:
Use a simple classification system to drill down into what’s driving your AI Visibility, and what may be holding it back.
| Finding | Likely Underlying Issues |
| Brand is absent | Low-authority, weak category coverage or limited third-party, earned media discussion |
| Brand present but with negative framing | Negative source concentration from high-authority sources |
| Brand present but cited using old earned media | Old, potentially outdated sources have yet to be replaced by newer, more authoritative media that's been reinforced by earned and shared channels |
| Strong visibility on some platforms, weak on others | Different source ecosystems or recency preferences to interrogate |
| Visibility is surpassed by competitors | Lack of comparison-worthy proof points, customer evidence and independent reviews |
Compare how often and where your brand appears on leading LLMs compared to its direct and tangential competitors. Is there a clear leader? Even distribution? Are there differences between platforms? Depending on your industry, this ranking may change from day-to-day, so treat this audit as a directional research method that can be done periodically, not a once-off process.
Most marketers already know the PESO model: Paid, Earned, Shared, and Owned media. It's been a useful way to organise marketing activity for over a decade, originally as a way to think about integrated communications and, more recently, SEO.
Paid media is what you pay for directly. Earned media is coverage and mentions you didn't pay for. Shared media is the conversation happening across social platforms and communities. Owned media is everything published on channels you control.
What's changed is why this framework matters. To AI, a brand that only publishes polished owned content, with no independent validation and no conversation happening around it, looks thin to a system that's trying to establish whether a claim is trustworthy. A brand with great press coverage but a confusing or outdated website gives the same system conflicting signals to reconcile.
Getting visible and staying credible in AI-generated answers means treating PESO as one connected system rather than four separate workstreams. Here's what each part is actually doing in that system.
Paid media is usually an indirect contributor to AI visibility, not a direct one. Its job is to reinforce consistent messaging at scale. For example:
Earned media is where most AI citations actually originate, providing third-party, authoritative credibility. In some studies, earned media coverage has been seen to dominate over 90% of citations. Chen et al. (2025) called it a “systematic and overwhelming bias” over brand-owned and social content.
A claim repeated across several independent outlets carries more weight than the same claim made once, by you, about yourself.
Shared media plays a much bigger role in AI outputs than most brands assume, and its role often differs sharply by platform. In top Google AI Overview citations, for example, the most-referenced domains tend to be YouTube, Reddit, and Facebook.
Similar to earned media, shared attention builds attention and credibility. For AI models, engagement on social media, forums, review sites, and otherwise, signals relevance and usefulness.
This is the source material everything else points back to. The most effective format for owned content in an AI-visibility context is one that answers specific questions plainly:
Although owned media rarely gets cited at the volume that earned media does, this type of content allows you to control the narrative at the source, strengthens brand recognition, and can work to establish your brand as a credible, citable voice, particularly when content is backed by data or methodology.
Altogether, AI visibility shows up where these four media types overlap: owned content that's clear enough to be cited, earned coverage that validates it, shared conversation that reflects real usage and honest opinion, and paid media that gives all of it a bigger audience to be discovered by.
Instead of “We got 18 more AI citations this quarter,” you’ll want to be able to find insights that read like:
“Across our priority ANZ prompts, brand mention share increased by 10% this quarter while our competitors’ share only grew 5%. Positive sentiment across all focus LLMs also grew 4%. Three of our Tier 1 media placements were picked up by ChatGPT and Gemini, contributing to this trend.”
The areas your team will need to stay across include:
AI visibility is more than the next channel your communications team needs to own. It also functions as a mirror that shows how well your existing channels are working together.
The best way to see this is in a dashboard that isn't a standalone AI Visibility silo, one that brings media monitoring, outreach, social listening, and AI Visibility together in one place, so you can see what’s working.
Brand conversations are already happening inside AI searches and answers, with models increasingly shaping which brands audiences see, and how they stack up against competitors.
With Streem AI Visibility Dashboards, your team can now monitor all leading LLMs, find their top-cited sources, and refine your media strategy to fill gaps, all inside the same platform you already use for media monitoring, social listening, and outreach.