What makes this different

The AI platforms are the data source. They answer for themselves.

Traditional SEO tools measure search rankings, traffic, and web content — none of which tell you what AI platforms actually say about your brand. This audit goes directly to the source. Each platform receives the same prompt via its official API and responds independently, in its own words, with no knowledge of what the others said. Then a separate analysis pass reads all four response sets and writes your report.

AI platforms self-report

ChatGPT, Claude, Gemini, and Perplexity each receive the same questionnaire simultaneously via their official APIs. They respond independently with no knowledge of each other's answers.

Responses captured and scored

Every response is recorded verbatim. Five dimension scores are extracted per platform. Nothing is approximated, inferred, or estimated from proxy signals.

Analysis AI writes your report

A separate AI analysis reads the complete set of responses from all four platforms, writes per-platform verdicts, surfaces cross-platform patterns, and generates your AEO action plan.

Branded vs. unbranded

Two types of visibility. Two completely different things to get right.

A brand can score well on one and fail completely on the other. Both matter. Most brands we audit have a gap between them.

Branded Presence

Your brand name is in the prompt

AI is asked specifically about you and must answer with what it knows. This reveals how accurately AI represents you to people who already know you exist, whether it knows your services, your location, your leadership, and whether the facts it states are correct.

"Is The Hargrove Group accepting new clients in Marin County?"

Branded search

Unbranded Presence

Your brand name never appears

AI has to surface you on its own, based on what it knows about your category and location. This is where real discovery happens, and where most brands have the biggest gaps. If you don't appear here, you're invisible to people who haven't heard of you yet.

"Who are the best luxury real estate agents in Marin County?"

Unbranded discovery

The question design

Seven areas. A complete picture of where you stand.

The questionnaire is structured around seven distinct areas of AI visibility. Each tests something specific. Together they reveal not just whether AI knows your brand, but how well, in what context, and where the gaps are.

Unbranded

Local and Niche Visibility

Questions that test whether your brand surfaces when location is part of the prompt: "best [category] in [city]," "who handles [specialty] near [area]." This is the hardest visibility signal to earn, because AI must independently associate your brand with a specific market, not just a category. For most regional businesses, it is where the largest gaps are found and where the most impactful improvements can be made.

Branded

Brand Recognition

Direct questions that test what AI knows about your brand when asked by name. This reveals whether AI can clearly describe who you are, what you do, and where you fit in your market.

Unbranded

Category Presence

Questions that ask AI to recommend providers or list top firms without mentioning your name. If you don't appear here, you're invisible to people who haven't heard of you yet.

Branded

Competitive Positioning

Questions that test how AI places your brand relative to competitors. This surfaces how you are framed alongside alternatives and whether your real differentiators come through.

Branded

Sentiment and Reputation

Questions that explore how AI characterizes your brand's reputation and the experience it associates with you. The narrative AI has built, even when no one asks a negative question.

Branded

Factual Accuracy

Questions that test specific details: your services, your leadership, your history, and your recognitions. Errors here are common and often invisible until you look for them directly.

Unbranded

Topical Authority

Questions that test whether your brand surfaces as a relevant resource for industry topics without your name in the prompt. Strong topical authority means AI sees you as a credible voice in your category.

The platforms

Four engines. Four different ways of knowing what they know.

Each platform is queried via its official API. They receive the same prompt independently, with no knowledge of what the others said. The reason their answers differ is not random. Each one sources its knowledge differently, and that directly affects what it knows about your brand.

ChatGPT Training + Live Search

OpenAI's model combines a large pre-training corpus with real-time web search via Bing, triggered selectively — not on every query. That split means ChatGPT's knowledge of your brand can vary depending on whether a given question triggers a search or draws from training data alone. It is the AI most people reach for first.

High usage platform. What it says about you shapes the first impression for the largest number of users.

Claude Training Data Only

When queried via API, Anthropic's model has no real-time web access. It relies entirely on what was published and indexed before its training cutoff. Claude is the clearest signal of long-standing brand presence: if your information is not embedded across the web, it shows here first.

The purest test of your historical web footprint. Low scores here often mean you have not accumulated enough indexed content over time.

Gemini Google Index (Live)

Google's model is built on the same infrastructure powering Google Search and AI Overviews. It queries Google's live search index in real time, which means your Gemini score is the closest available proxy for how Google's AI systems understand your brand — including your Google Business Profile and local citations.

Your Gemini score directly reflects your local SEO signals. A low score here often points to gaps in your Google Business Profile or local citation footprint.

Perplexity Live Web Retrieval

Every Perplexity query triggers a real-time web search. Answers are generated from those results with citations, with no reliance on a pre-training snapshot. Your Perplexity score is the most direct reflection of your current web presence: what is published, indexed, and findable right now, including directories, review platforms, and press coverage.

The fastest-moving signal. Changes to your web presence show up here before any other platform, often within days.

The report

One more pass turns platform data into your report.

After all four platforms respond, a cross-platform analysis reads the complete response set and generates the analytical sections of your report from what it finds. Scores feed the charts. The narrative is written fresh from the platform responses, every single cycle.

Scores parsed and graded

Five dimension scores are extracted per platform and averaged across all four. The result is your overall score out of 100, plus per-platform breakdowns.

Per-platform verdicts written

For each platform, a verdict is written: what it claims about your brand, how it characterizes you, and where the gaps are. You bring the ground truth. The verdict shows what each platform says.

Cross-platform patterns surface

With all four platforms in view, patterns emerge that no single platform can reveal. Where they agree, where they diverge, which claims are consistent, these signals reveal the true state of your footprint.

Gaps named, not buried

The report does not average your weaknesses into a number. The category breakdown shows which dimensions are underperforming and on which platforms. Gaps are stated directly and prioritized for action.

Gemini Perplexity ChatGPT Claude

Sample report data

Gemini Has the Richest Picture

Gemini’s 74/100 reflects the most complete and evenly shaped radar profile in this report, with dimension scores ranging from 13 to 19 across all five axes. This consistency indicates a solid but not dominant presence. The brand is recognizable and factually accurate within Gemini’s source set, but topical depth is not broad enough to drive strong category discovery. Closing the gap between Brand Recognition and Category Presence should be the near-term focus.

Brand Recognition Leads All Categories

Brand Recognition averages 16.25/20 across all four platforms, the strongest individual category in the report. This signals that the brand’s direct identity, name, location, and basic service description is reliably indexed across AI sources, driven primarily by the website, Google Business Profile, and directory listings. A solid foundation. The challenge is translating that recognizability into non-branded discovery contexts where the brand name does not appear in the prompt.

Topical Authority Is the Critical Gap

With an average of just 10.75/20, Topical Authority is the weakest dimension across all four platforms and the single highest-leverage improvement opportunity. AI platforms determine topical authority by whether a brand produces or is cited in expert-level content relevant to its category. Currently there is limited published educational content, third-party citations, or industry commentary establishing topical expertise to AI systems evaluating authority at scale.

Recurring audits

AI is not static. Neither is your score.

AI platforms update their models, re-index content, and shift behavior continuously. A score from 30 days ago reflects a different state of the internet, and potentially a different version of the AI, than today. Recurring tracking is what turns a snapshot into a signal you can act on.

Overall Visibility Score over 6 audits

Line chart showing AI visibility score rising from 41 to 79 across six weekly audits

+38 points over 6 weeks

Model updates change what AI knows

All four platforms update on rolling schedules. Those updates can shift how a brand is represented overnight, sometimes improving visibility, sometimes erasing it. Monthly tracking shows you when a model shift moves your score and why.

Your content keeps accumulating

Every article published, review earned, or mention accumulated online contributes to what AI understands about your brand. The effect is rarely immediate, but it is measurable over weeks. Recurring audits connect your content decisions to visibility outcomes.

Competitor activity reshapes your category

Category presence is relative. If a competitor earns a surge of coverage, AI may begin surfacing them ahead of you in the same unbranded prompts. That shift happens gradually, then suddenly. Recurring tracking gives you the signal before it becomes lost business.

We ask the questions your customers ask.
Then we show you how AI describes your brand.

From brand name to report. Here is what happens in between.

The AI platforms are the data source. They answer for themselves.

Two types of visibility. Two completely different things to get right.

Not sure how AI sees your brand right now?

Seven areas. A complete picture of where you stand.

Four engines. Four different ways of knowing what they know.