What "appearing in ChatGPT" actually means
It helps to separate two related but distinct ideas: ranking and citation.
Ranking refers to where a page sits within a list of organic results, governed by familiar SEO signals. Citation refers to whether an AI system references a brand inside the answer it generates. The two share foundations — authority, relevance, topical depth — but reward different work, and a brand can perform well in one while staying invisible in the other.
Citation itself takes two forms. The first is an unlinked brand mention, where an AI system references the company by name within its response. This contributes to brand salience and recall but doesn't generate direct traffic. The second is a sourced citation, where the AI references the brand's content as a source, usually with a link the user can follow. Both forms contribute to share of voice within AI search, but the sourced form compounds over time as users follow the link and reinforce the relationship between the brand and the question.
The principles governing citation are broadly consistent across ChatGPT, Gemini, Perplexity and Claude. Each platform uses its own retrieval system, but a brand that becomes citable on one is generally citable across the others, provided the underlying work was structural rather than platform-specific.
Why AI visibility sits across marketing and software
The most common reason brands underperform in AI search is not effort, but structure. AI citation depends on factors that span two disciplines usually run separately: marketing and software engineering.
The marketing side covers content quality, topic selection, the clarity and completeness with which questions are answered, third-party validation across the wider web, and the trust signals that influence source-level authority. The engineering side covers schema markup, information architecture, site performance, crawlability and the technical conditions that determine whether AI systems can read, extract and confidently cite from the content in the first place.
Most agencies operate cleanly on one side of that line. Marketing teams produce content the technical layer can't fully support. Development teams build sound foundations the content layer never fully exploits. The work that determines AI visibility falls into the gap between them, which neither side owns.
PQ Impact is structured to close that gap. Our strategic marketing and custom software development practices operate as a single integrated capability. Our team has built and scaled our own products from the ground up, which means the technical decisions that influence AI visibility — schema strategy, site architecture, content infrastructure, performance — are decisions we have implemented on our own platforms before recommending them to clients. That depth sits alongside our demand intelligence, marketing strategy and analytics work, not in a separate function.
The implication is straightforward. AI visibility is not a campaign. It is the cumulative outcome of how content and infrastructure decisions are coordinated over time.
How ChatGPT decides what to cite
AI systems evaluate sources somewhat independently of conventional ranking signals. Backlinks and domain authority still matter, but they carry less relative weight, and several other factors carry more. Understanding these factors is the first step toward influencing them.
Content clarity and structure. AI systems prefer content from which they can extract a clean, self-contained answer. Pages built around clear questions and unambiguous answers, with defined sections and discrete answer units, are easier to cite than pages where the information is buried in long, undifferentiated paragraphs. Clarity is editorial; execution depends on the technical layer presenting it in a structured form.
Semantic completeness and topical authority. AI prefers sources that answer a question fully rather than partially. A page that treats one topic thoroughly typically outperforms a page that touches on the same topic among several others. This is the principle of topical authority: the cumulative demonstration that a brand covers a category with enough breadth and depth to be treated as a reliable reference.
Authority signals beyond backlinks. Brand mentions across credible third-party sources, citations in industry publications and consistent positioning across multiple domains all contribute to the trust signals AI systems use when deciding what to reference. This is reputational work, but it is measurable.
Schema markup. Pages with comprehensive schema, particularly FAQ, Article and Organisation schema, are materially more likely to appear in AI-generated responses than pages without it. Schema gives AI systems machine-readable units they can extract and cite with confidence. It has shifted from a technical refinement to one of the more direct levers for AI visibility.
Information architecture and internal linking. AI systems use internal linking and topical clustering to identify which brand has genuine authority on a category, rather than just on individual pages. Coherent architecture signals breadth and depth; poor architecture leaves valuable content effectively invisible.
Performance and crawlability. AI systems depend on well-structured, performant, crawlable sites to identify what to cite. Poor performance, broken linking or inconsistent rendering quietly reduces citation in ways that are difficult to detect through conventional SEO metrics.
Freshness. AI systems weight recency more than traditional search does. Content that has not been reviewed or updated competes poorly against current sources covering the same ground, regardless of historical authority.
No single factor drives citation in isolation. The brands that perform well are typically strong across most of these areas, without a single weakness pulling down the overall signal.
Why demand intelligence is the input that holds everything together
AI visibility work is only as good as the strategic input that shapes it. The most common failure mode isn't editorial or technical, it's strategic: content built around topics chosen because they sound important internally, keywords inherited from previous campaigns, or briefs shaped by what the team finds interesting rather than what the market is asking.
In a traditional SEO environment, this was always inefficient. In an AI-powered search environment, it is materially more expensive. AI systems are particularly good at surfacing brands that have answered real buyer questions well, and particularly unforgiving toward content that hasn't. The cost of getting the input wrong is now visible in the output in a way it wasn't before.
Our demand intelligence work, part of the strategic marketing offer, addresses this directly. We use PAA (People Also Ask) analysis, search behaviour data, user question mapping and competitive intelligence to surface what buyers are actually asking, rather than what internal teams assume they're asking. The picture that emerges almost always differs meaningfully from the picture marketing teams arrive with. That evidence becomes the input for what gets written, how it is structured, and how performance is measured.
Without this step, AI visibility work tends to optimise the wrong content well.
The practical work: what to do, in order
Each stage below is a discipline in its own right, and rarely a one-off project. The order matters: doing the later steps before the earlier ones tends to produce work that has to be redone.
1. Establish a baseline. Start with how ChatGPT, Gemini and Perplexity currently represent the brand, where competitors sit within AI responses, and where the brand is absent, misrepresented or underrepresented. Our AI Brand Visibility Report is built for this: a visibility score, competitor benchmarking, gap analysis and a prioritised implementation roadmap. The strategy that follows is only as good as this baseline.
2. Surface real buyer questions. Replace inherited keyword lists with verified data on what the market is actually asking. The goal is a clear view of buyer-side demand, the questions competitors answer well or poorly, and the white space available to the brand.
3. Audit content for extractability. Restructure key pages around clear questions and complete answers. Tighten phrasing, remove ambiguity, break long sections into self-contained units. The goal is to make it as straightforward as possible for an AI system to lift a single, accurate, complete answer and cite the source confidently.
4. Implement comprehensive schema markup. Prioritise FAQ, Article and Organisation schema across the pages most likely to surface in AI answers. Avoid the common errors: inconsistent schema across templates, incomplete properties, schema that fails validation.
5. Strengthen the technical foundation. Performance, crawlability, structured data and information architecture all shape what AI systems can extract and cite. Where the work goes beyond standard SEO fixes, our custom software development capability addresses it directly.
6. Build topical authority across categories. Comprehensive category coverage outperforms isolated page optimisation. The goal is to be the source AI systems return to repeatedly on a subject, which requires breadth, internal linking and consistent quality across a connected set of pages.
7. Earn brand mentions across credible third-party sources. Trade publications, podcasts, partner content and earned coverage all contribute to source-level trust signals. This is patient work, but it compounds.
8. Set up measurement and monitoring. Citation rate, share of voice in AI answers, accuracy of brand description and competitor presence over time should be tracked the same way conventional metrics are. Our analytics and reporting frameworks are built to make that operational rather than ad hoc.
Common mistakes that keep brands invisible
A small number of patterns account for most underperformance in AI search.
- Treating AI search as either a passing trend or a complete replacement for SEO. Both readings are wrong. It is a structural addition to the search ecosystem, not a substitute, and not a fad.
- Running marketing and engineering as separate workstreams. Where the two functions are organisationally separated, the work falls into the gap between them.
- Optimising for keyword volume rather than verified buyer questions. AI systems reward content that answers real questions thoroughly.
- Publishing thin content that AI systems can't extract from confidently. Citation rewards completeness and clarity, not length for its own sake.
- Treating schema markup as a technical nice-to-have. It is one of the more direct levers for AI visibility, and one of the more reliably underused.
- Assuming Google ranking translates directly into AI citation. It doesn't. The competitive dynamics in AI search are genuinely different.
- Measuring only conventional SEO metrics. A brand can rank well and remain functionally invisible inside AI answers. Without measuring citation, that invisibility goes undetected.
How PQ Impact approaches AI visibility
Our work spans strategic marketing and custom software development because the visibility problem spans both. Treating it as a marketing problem alone, or as a technical problem alone, is the most common reason brands underinvest in the right areas and see disproportionately little return.
We have built and scaled our own products from the ground up. The technical decisions that determine whether AI systems can cite a brand cleanly — schema strategy, architecture, performance, content infrastructure — are decisions we have implemented and refined on our own platforms before recommending them to clients. That depth sits alongside our marketing strategy, demand intelligence, content and analytics capability, so AI visibility work is shaped by evidence on both sides of the discipline.
This integrated posture is what most search and content programmes are missing. It is the difference between a visibility strategy that compounds over time and one that simply continues without the underlying numbers improving.
Where to start
Appearing in ChatGPT answers is a distinct discipline from ranking on Google. It builds on the same foundations, but it rewards brands that bring marketing and software expertise together, and that work from evidence rather than assumption. The starting point is knowing where the brand currently stands.
Our AI Brand Visibility Report shows exactly that: how ChatGPT, Gemini and Perplexity currently describe your brand, how competitors appear in those same answers, where the gaps sit, and what to fix first. The broader strategy work follows from there.
Get in touch to discuss the report and how it would apply to your business.
