AI VISIBILITY FOR BUSINESS
The Orca AI Recommendation Test
Can AI find your business? Does it understand you? Will it recommend you?
The Orca AI Recommendation Test is a practical framework for examining how your business appears across AI-powered search and recommendation platforms.
It looks beyond a single ChatGPT response to assess how consistently your business is discovered, accurately represented and selected across the different questions your customers may ask.
How the test works →
FIND. UNDERSTAND. RECOMMEND.
THE PROBLEM WITH QUICK CHECKS
One answer isn’t a measurement.
Asking ChatGPT to recommend a business like yours can be a useful place to start. It may reveal that your business is missing, described inaccurately or being overlooked in favour of competitors.
Small changes in wording can alter which businesses are considered relevant.
01
The question asked.
A business may appear for a specific requirement but remain absent from broader category searches.
02
The customer’s needs.
A user’s location, previous interactions and available context can influence the answer they receive.
03
Location and personalisation.
ChatGPT, Google’s AI features, Claude.AI and Perplexity may use different sources and produce different recommendations.
04
The platform used.
AI platforms can only work with the business information, content, reviews and third-party evidence they can find and interpret.
05
The information available.
A ONE-OFF SEARCH ASKS:
“Did my business appear this time?”
A MEANINGFUL AI VISIBILITY TEST ASKS SOMETHING MORE USEFUL:
“How consistently is my business found, understood and recommended across the searches that could influence a customer’s decision?”
A three-part test of aI visibility
APPEARING IN ONE AI-GENERATED ANSWER IS NOT THE SAME AS HAVING A STRONG, RELIABLE PRESENCE.
The Orca AI Recommendation Test examines three distinct layers of visibility. Together, they reveal whether AI platforms can access meaningful information about your business, interpret it accurately and consider it relevant when someone asks for a recommendation.
A repeatable test, not a screenshot.
The Orca AI Recommendation Test uses a structured set of prompts and observations to build a more reliable picture of how a business appears across AI-powered search.
It does not assign a misleading “ChatGPT ranking.” Instead, it examines patterns across the questions, platforms and recommendation contexts most relevant to the business.
-
We begin with the questions real customers may ask while discovering, comparing and choosing businesses.
These can include:
Broad category searches
Location-based recommendations
Problem and need-based searches
Searches involving specific attributes or requirements
Comparison and evaluation questions
Branded questions used to assess accuracy
The prompt set is shaped around the business, its customers and its commercial priorities rather than copied from a generic keyword list.
-
Location, personalisation and previous conversations can influence AI-generated answers.
Tests are therefore conducted using clearly defined prompts and conditions. Location is stated directly, new conversations are used where appropriate and branded searches are separated from unbranded recommendation searches.
This reduces avoidable distortion and makes results easier to compare.
-
The same core prompt groups are tested across selected AI search and recommendation platforms.
Results are observed over more than one prompt or isolated response, helping identify recurring patterns rather than treating a single answer as definitive.
-
The test captures more than whether the business is mentioned.
It examines:
Recommendation inclusion
Accuracy of the business description
Services, locations and attributes associated with the business
Competitors appearing in the same answers
Sources and pages used to support responses
Missing, outdated or conflicting information
The prompts under which visibility strengthens or disappears
-
The findings are compared with the business’s website, content, authority signals and wider digital footprint.
This helps reveal whether the primary issue is discoverability, understanding, credibility, relevance or a combination of several factors.
The outcome is not simply a visibility score. It is a clearer set of priorities for strengthening how the business is represented and considered across AI-powered search.
WHAT THE GAPS CAN REVEAL
The gap isn’t always where you think it is.
When a business is missing from AI-generated recommendations, the immediate assumption is often that it needs more content, more keywords or more technical optimisation.
Sometimes it does. But the real issue may sit somewhere else entirely.
The Orca AI Recommendation Test helps distinguish between different types of visibility and recommendation gaps.
A discovery gap
Important information about the business is difficult to find, inaccessible to relevant crawlers or absent from the sources AI platforms consult.
A relevance gap
AI platforms can identify the business, but cannot confidently connect it with the specific needs, attributes or questions contained in the prompt.
A clarity gap
The business exists online, but its services, locations, audience or points of difference are not communicated clearly enough.
A consistency gap
Business details, positioning or service information vary across the website, profiles, directories and other sources.
An evidence gap
The website makes strong claims, but there is insufficient independent information, authority or third-party corroboration to support them.
A competitive gap
Competing businesses have stronger or more accessible signals connecting them with the customer’s request.
The value of the test is not simply identifying that visibility is weak. It is understanding what may be limiting it and where focused work is most likely to make a meaningful difference.
Different gaps require different responses. More website content will not correct inconsistent business information. Schema alone will not create authority. Better rankings will not necessarily ensure that AI platforms understand why a business should be recommended for a particular need.
Run a quick AI visibility check
You don’t need specialist software to uncover the first signs of an AI visibility problem.
This quick check will not measure your visibility conclusively, but it can reveal whether AI platforms can identify your business, describe it accurately and connect it with relevant customer needs.
BEFORE YOU BEGIN
Open a new conversation and state the location directly rather than assuming the platform knows where you are.
Use the same prompts across ChatGPT, Gemini and Perplexity where possible. Record the date and results so you can compare what each platform returns.
-
“Which [type of business] in [location] would you recommend? Explain why and cite your sources.”
This tests whether your business enters the conversation without being named or prompted.
-
“I’m looking for a [type of business] in [location] that [specific requirement or need]. Which options would you recommend and why?”
Choose a genuine need or attribute that should make your business relevant.
-
“What are the best options for [service or need] in [location]? Compare their strengths and who each would suit.”
This can reveal which qualities AI associates with your business and its competitors.
-
“What can you tell me about [business name], including what it offers, where it operates, who it serves and what distinguishes it? Cite your sources.”
This does not test unprompted discovery, but it can expose inaccurate, incomplete or outdated information.
-
For each response, record:
Whether your business appeared
How accurately it was described
Which services and attributes were associated with it
Which competitors appeared
Which sources were cited
Any missing, outdated or conflicting information
What the results mean.
Appearing once does not prove strong AI visibility, and being absent once does not prove that your business is entirely invisible.
Look for patterns. Does your business appear only when named? Is it understood accurately but rarely recommended? Does it surface for specific needs but disappear from broader searches? Are competitors supported by sources your business lacks?
Those patterns provide the first clues about where your recommendation gap may sit.
Download the AI-visibility quick-check scorecard →
WHEN A QUICK CHECK ISN’T ENOUGH
A self-test can show the problem. An audit investigates why.
A self-test can expose obvious visibility issues. It may show that your business is missing from recommendations, being described inaccurately or appearing only under very specific conditions.
What it cannot easily explain is why.
The Orca AI Recommendation Audit
The Orca AI Recommendation Gap Audit is a tailored AI visibility audit designed to examine how your business is found, understood and considered across AI-powered search.
It combines structured prompt testing with an analysis of your website, content, competitors, authority signals and wider digital footprint.
Depending on the business and agreed scope, the audit may include:
Multi-platform testing
Structured testing across selected AI search and recommendation platforms to identify recurring patterns and meaningful differences.
Brand accuracy review
An assessment of how accurately AI platforms describe your business, services, locations, expertise and points of difference.
Competitor Recommendation Analysis
Insight into which competitors are being selected, the contexts in which they appear and the signals that may be supporting their visibility.
Recommendation Gap Diagnosis
Identification of the discovery, clarity, relevance, evidence, consistency and competitive gaps most likely to be affecting your visibility.
Tailored Prompt Landscape
A set of recommendation, comparison, location and customer-need prompts shaped around your services, audience and commercial priorities.
Multi-platform testing
Structured testing across selected AI search and recommendation platforms to identify recurring patterns and meaningful differences.
Source & Citation Review
Analysis of the websites, pages and third-party sources being used to support AI-generated answers.
Website & Content Review
A review of whether your website clearly communicates the information AI platforms and potential customers need to understand your business.
Prioritised action Plan
Clear recommendations ordered by relevance, potential impact and the role each action plays in strengthening your wider search presence.
The audit does not promise a fixed “ranking” or guaranteed inclusion in AI-generated recommendations.
It provides something more useful: a clearer understanding of your current position, the evidence shaping it and the work most likely to strengthen your visibility over time.