What Is an LLM Visibility Audit?

The short answer

An LLM Visibility Audit is a structured assessment of how a company appears — or fails to appear — when AI systems such as ChatGPT, Google's AI features, and Microsoft Copilot answer questions relevant to that company's products, services, and market. It identifies where a brand is mentioned, cited, absent, or inaccurately described, and translates those findings into a prioritized set of content and technical recommendations.

Why an audit is the right starting point

SEO tells you where your pages rank. It does not tell you whether your brand is part of the answer when a prospective customer asks an AI system a real question. An LLM Visibility Audit closes that gap: it identifies the questions your customers are actually asking, analyzes the answers AI systems are currently providing, and evaluates the digital signals surrounding your brand that likely influence those answers.

Prompt research and keyword research

The audit begins with prompt research: identifying the real, natural-language questions prospective customers ask at each stage of their decision, from broad category questions to specific eligibility and comparison questions. This is paired with keyword research to understand which topics carry commercial weight and search volume, so the prompt set reflects genuine business priority rather than an arbitrary sample.

Brand testing and competitor testing

Each prompt is run against major AI platforms, repeated across sessions to account for natural variation in generated answers. Brand testing records whether, how, and how accurately the company appears. Competitor testing records the same for named competitors, so a company can see not just whether it is present, but how its visibility compares to others being surfaced for the same questions.

Citation analysis and mention analysis

Citation analysis identifies which sources an AI system actually references when it answers a given question — company pages, news coverage, directories, review platforms — revealing which types of sources currently carry authority for that topic. Mention analysis distinguishes between a company being named directly, being cited with a link, or being entirely absent, since these are meaningfully different outcomes.

• Is the company mentioned by name in the answer?
• Is the company cited as a source with a link?
• Is the description of the company accurate and current?
• Which competitors appear for the same prompt, and how are they described?

Content gap analysis and technical website analysis

Content gap analysis compares the prompt set against the company's existing content to identify questions the company should be answering directly but currently is not. Technical website analysis checks the foundational requirements that make content retrievable at all: crawlability, indexing status, page speed, and structured data implementation.

Recommendations, content strategy, and ongoing measurement

The audit concludes with a prioritized set of recommendations, typically covering technical fixes, content to create or revise, and authority-building opportunities such as third-party mentions worth pursuing. Because AI-generated answers change as models and indexes are updated, ongoing measurement — retesting the same prompt set on a regular cadence — is necessary to know whether visibility is actually improving over time, rather than assuming a one-time fix is permanent.

What an audit does not do

An LLM Visibility Audit cannot guarantee that a company will be cited by ChatGPT, Google's AI features, or any other AI system. No credible methodology can make that guarantee, because no outside party controls how these models retrieve or generate answers. What the audit provides is an honest, evidence-based picture of current visibility and a clear set of priorities for improving it.