Ask ChatGPT Who to Hire. Then Look for Your Name.
Imagine a buyer asking ChatGPT, “Which company should we hire for this?” They might be looking for a software provider, a manufacturer, an agency, or a…
specialist in your category.
The answer may name a handful of companies and explain why each one fits. Would yours be there?
That question is worth testing. It is also easy to test badly.
Start with the question your buyer would ask
Leave your company name out of the prompt. Include the details that would actually shape the decision: the buyer’s industry, the problem, geography, company size, and any constraints that matter.
For example: “Which firms should a Chicago-based manufacturing company consider for [the specific need]?” The useful wording will be different for every business. Your sales team probably hears better versions of this question every week.
Ask it in ChatGPT, then try the same question in a few other AI services. Save the complete answers. Note who appears, how each company is described, and which sources the service shows.
Look beyond whether you are named
There are at least three different outcomes to watch for:
Absent: Your company does not appear in the answers you reviewed.
Mentioned: It appears, but the answer does not explain why a buyer should consider it.
Recommended: The answer presents it as a fit for the buyer’s specific need.
If you are named, read the description. Is it accurate? Does it reflect what makes you different? If a competitor is named instead, what does the answer say about them, and what evidence does it point to?
A mention, a recommendation, and a citation are different signals. None is a sale. Keeping them separate helps your team decide what to work on.
One answer is a starting point
AI answers can vary across services and runs. A single prompt is useful for spotting a question to investigate; it is not a reliable market-wide visibility score.
For a clearer baseline, choose a small set of real buyer questions, run them repeatedly across the services that matter to your audience, and keep the questions and testing conditions consistent. Record the answers and sources so someone else on your team can check the finding.
That is especially important when you make a change. If your company appears more often on a later retest, you can report the observed difference. You cannot assume that one page edit caused it.
Turn the finding into work your team can use
If your company is missing or misrepresented, start with the pages and sources shaping those answers. Your website should plainly explain who you serve, the problems you solve, and why a buyer would choose you. Product pages, comparisons, customer proof, directories, and third-party coverage may also matter, depending on what the answers actually cite.
The point is not to replace your current marketing plan. It is to see where buyers may be forming an impression before they reach your site, then make the most useful changes within the work your team already does.
At Signal Harbor, we use buyer questions to build an AI Visibility Snapshot, review the answers and sources, and turn the gaps into a campaign plan your team can inspect in the portal. We then retest the same questions to see what changed within that scope.
Try the simple version yourself this week: write down five questions a real buyer would ask, leave your brand name out, and look closely at the answers. If you want help going deeper, book a complimentary Snapshot and portal demo.
Sebastian Miller
Co-Founder, Signal Harbor
signalharborai.com · LinkedIn
This article first appeared in Signal Harbor Weekly, the Signal Harbor newsletter on AI visibility.
Keep reading.
All articlesHow to Measure AI Visibility When the Answers Keep Changing
A marketing team asks an AI service which providers a buyer should consider. Their company appears.
AI Visibility Is Not Traffic
Why the influence of an AI recommendation may appear somewhere else. Imagine a software buyer looking for a new competitive intelligence platform.
Signal Harbor, in Plain Answers
If you are looking at Signal Harbor, these are the questions worth starting with.
