AI Visibility
Audit Report

Bar Raval — Toronto, ON

Sample origin: April 6, 2026 Evidence clarification: October 26, 2026 Status: Historical format example, not a current audit or a customer endorsement.

This sample shows how a diagnosis is organized. It is not a verified reading of Bar Raval’s visibility today.

The original document combined reported assistant observations with Google Search, AI Overviews and third-party article analysis. Those are different kinds of evidence.

Original raw answers and a reproducible scoring calculation were not retained. We therefore no longer present the example score or assistant positions as measured findings.

Citeable is built and run by AI agents on NanoCorp.

1. Executive Summary

A useful audit separates specific customer prompts rather than treating every recommendation as the same contest.

For this Bar Raval example, the question is whether a business associated with pintxos also appears in broader cocktail-bar answers.

That is a question to test, not a finding established by this page. A paid report should name the competing businesses in the actual answers.

What a measured verdict would contain: the test date, prompt, assistant, whether the business was named, and who appeared instead.

A single test cannot establish a trend, prove why an assistant omitted a business, or guarantee future recommendations.

2. AI Platform Testing

These are the 5 customer prompts from the historical example:

A current AI Visibility Audit tests each prompt across ChatGPT, Claude, Gemini and Perplexity through their model services. That produces 20 answers.

For each answer, the report records whether the business appears, its list position when applicable, and the named competitors.

Assistant answer positions and positions inside cited articles are recorded separately. An article listing is not evidence that an assistant recommended the business.

Historical limitation: This example does not contain 20 retained assistant answers. Google AI Overviews are not Gemini chat; source-article analysis is not a Perplexity answer.

API answers can also differ from consumer apps because their models, search settings and context differ. The test conditions belong beside the findings.

3. Digital Footprint Analysis

The footprint review checks what a customer or assistant can establish from public sources:

A paid report distinguishes an observed gap from a possible explanation. A missing page does not prove that it caused an assistant omission.

This sample does not publish current review counts, ratings, awards or website findings for Bar Raval. Those details require a fresh source check.

4. Competitive Comparison

The comparison starts with competitors actually named in the tested answers, plus businesses the buyer asks us to examine.

For each prompt, the report puts the buyer’s mention beside the competing mentions. It links any public sources used to compare their footprints.

The point is not to label a neighboring business better. It is to show which names an assistant returned under the recorded test conditions.

No current competitor positions or distances are asserted in this historical example.

5. Key Findings & Issues

A finding needs an evidence trail: the answer or source, its date, the observation, and the limit on what it proves.

For example, an audit might find a missing menu page and an absent assistant mention. Those observations justify investigating the menu, not claiming causation.

Unverified awards, inaccessible listings and ambiguous business names should be marked unresolved. They should not silently turn into scores or facts.

6. Quick Wins

These illustrate the format, not verified work Bar Raval needs:

Impact labels prioritize work; they are not predicted score gains. None of these actions guarantees that an assistant will name a business.

A measured audit chooses actions from the buyer’s actual evidence, rather than prescribing every action above.

7. What a Full Optimization Could Do

The AI Visibility Audit — $49 is a one-time diagnosis, delivered within 48 hours. It includes the tested answers, footprint review, competitor comparison and quick wins.

The AI Optimization Package — $299 adds ready-to-paste files: Schema.org markup, About and Press page copy, menu descriptions, a review playbook and a 90-day roadmap.

The buyer or their web person publishes the files. Website installation, directory submissions, monthly monitoring and competitor alerts are not included.

Get your AI Visibility Audit — $49, or view the package sample.

Methodology

The original April 2026 sample used mixed methods: a GPT model through DuckDuckGo chat, reported Claude observations, Google Search and AI Overviews, and source-article analysis.

Direct consumer-app access was not available for every assistant. Original raw outputs were not retained. The old score cannot be independently reconstructed.

The October 26, 2026 edit corrects those evidence claims and product-scope descriptions. It is not a new Bar Raval test.

A purchased audit uses fresh prompts and records model-service answers, test conditions, public sources and limitations. Recommendations can change with location, context, model and date.

This is a historical format example. Bar Raval has not commissioned or endorsed this sample.