One Brand’s AI Visibility Tracking Ranged From 15.5% to 59.5% Depending on the AI Engine

• In Zen Media's 2026 corporate event planning campaign, one brand appeared in 15.5% to 59.5% of buyer-intent prompts depending on the AI engine.
• In an oilfield supply campaign, two engines with similar visibility produced first-place rates of 57.6% and 92.5%.
• In the enterprise AI readiness campaign, a GlobeNewswire release was cited in 47 prompts, slightly ahead of the client's own website at 46.
• In the two campaigns that tracked branded and non-branded discovery, 85% to 87% of the prompts that surfaced the client did not mention the company by name.

Miami, FL, Sept. 25, 2026 (GLOBE NEWSWIRE) -- Zen Media's AI visibility tracking in three 2026 campaigns covering 900 buyer-intent prompts showed that the same company can perform very differently across ChatGPT, Gemini, Claude, Perplexity and Grok, even on identical prompt sets.

In the corporate event planning campaign, the client appeared in 15.5% of prompts on the weakest engine and 59.5% on the strongest. The five-engine average was 45.6%.

Looking only at that average would have missed the weakest platform.

"Clients usually want one number, but one number can hide the thing they actually need to fix," said Sarah Evans, CEO of Zen Media. "If you're strong in four engines and barely showing up in the fifth, averaging everything together doesn't help you decide what to do next. We need to know where you're showing up, where you're leading, what sources are being cited and who is appearing instead of you."

Showing Up and Leading Are Different Results

Zen Media saw that clearly in its oilfield supply campaign.

Two engines surfaced the client at similar rates: 45.4% and 40.6%. One placed the client first in 57.6% of appearances. The other did so 92.5% of the time.

For the client, those results point to different problems. One engine may need more content that helps the brand enter relevant answers. Another may already recognize the brand as relevant but does not consistently place it near the top.

Combining both into a single visibility score makes that harder to see.

Most of the Visibility Came From Prompts That Did Not Name the Brand

Zen Media also looked at whether buyers had to mention the company by name for it to appear.

In the corporate event planning campaign, 152 of the 175 prompts that surfaced the client were non-branded, or 87%.

In the enterprise AI readiness campaign, 110 of 130 were non-branded, or 85%.

A prompt that contains a company's name starts with brand awareness. A buyer asking who can solve a problem, which provider to consider or how to handle a specific situation gives the engine a choice about which companies belong in the answer.

For brands trying to reach buyers before a shortlist has formed, that second type of visibility is especially important.

A Press Release Became the Most-Cited Source in One Campaign

The enterprise AI readiness campaign also produced a result that changed the content strategy.

A GlobeNewswire release was cited across 47 measured prompts, making it the most frequently cited source in the campaign. The client's own domain appeared in 46. Another owned domain appeared in 41.

The oilfield campaign looked different. There, the client's own website led the citation set ahead of the distributed release.

The source carrying a brand into AI answers can change from one campaign to another.

"Publishing more pages on your own website is not automatically the answer," Evans said. "In one campaign, the strongest citation asset was the distributed story. That changes what we publish next and where we put it. If you know which sources are actually carrying the brand into answers, you can make a much better content decision."

Zen Media uses both owned content and third-party distribution in its AI visibility programs for that reason.

Competitive Visibility Needs the Same Prompt Set

In the enterprise AI readiness campaign, Zen Media measured the client and competing firms against the same group of buyer prompts.

The client recorded 1,026 mentions. The next three named firms recorded 631, 472 and 432.

Those figures are useful because the companies were tested against the same questions during the same period. They are not universal scores and should not be compared with raw mention totals from another campaign using a different prompt set.

The same applies when comparing AI engines. The prompts, measurement window and competing answers need to stay consistent.

What an AI Visibility Report Should Show

Zen Media looks at five separate signals:

Signal	What it tells you
Engine coverage	How often the brand appears on each platform
Position	Where the brand lands when it appears
Non-branded discovery	Whether the brand surfaces when the buyer has not named it
Citation source	Which pages, releases or publications are being used in the answer
Competitive visibility	Which other companies appear for the same questions

The prompt set and measurement dates sit behind all five.

Without that context, a percentage or mention count says very little about what changed or what should happen next.

Measurement Should Change What Gets Published

Zen Media uses this engine-level analysis inside Published Monthly™ to decide what content to create and where to place it.

The work starts with buyer questions rather than a list of branded keywords. Those prompts show where the client already appears, where competitors are taking its place and which sources are showing up in the answers.

From there, Zen Media builds owned and distributed content around the prompt clusters where there is room to improve, then measures what happens after publication.

Zen Media calls the brand's presence across those measured answers Answer Share™: how much of the relevant answer set the brand occupies compared with the category and competitors being tracked.

The content plan changes depending on what the data shows. In one campaign, distributed coverage became the strongest citation source. In another, the client's own site did.

The reporting is meant to make the next publishing decision clearer.

How the Campaigns Were Measured

Zen Media analyzed three 2026 AI Visibility Optimization campaigns in corporate event planning, enterprise AI readiness and oilfield supply.

Together, the campaigns covered 900 buyer-intent prompts across ChatGPT, Claude, Gemini, Perplexity and Grok. Measurement windows ranged from seven to 21 days.

Coverage was calculated as the percentage of prompts in which the client appeared in an answer. Position was tracked separately.

Each campaign used its own prompt set and measurement window, so raw totals from one campaign were not treated as direct comparisons with another.

No AI platform participated in, reviewed or funded the analysis, and no platform was notified before the data was collected.

Frequently Asked Questions

Question: Why does my company show up in ChatGPT but not Gemini?

Answer: Different AI engines can return different companies for the same buyer question. In Zen Media's corporate event planning campaign, the gap between the strongest and weakest engine reached 44 percentage points on the same prompt set. The data shows where the difference exists, but it does not establish the internal reason one platform ranked a company differently from another.

Question: How do you measure AI visibility across ChatGPT, Gemini, Claude, Perplexity and Grok?

Answer: Use the same buyer-intent prompts across each engine and report the results separately. Zen Media tracks whether the brand appears, where it appears, whether the company surfaced without being named, which sources were cited and which competitors appeared for the same questions.

Question: Is a single AI visibility score accurate?

Answer: It can summarize performance, but it can also hide weak results on an individual platform. In Zen Media's corporate event planning campaign, average coverage across the five engines was 45.6%, while the weakest engine returned the client in only 15.5% of prompts.

Question: Do press releases get cited by AI engines?

Answer: They can. In Zen Media's enterprise AI readiness campaign, a GlobeNewswire release appeared as a cited source across 47 measured prompts, slightly ahead of the client's own website at 46. In the oilfield campaign, the client's website led instead.

Question: How do you measure non-branded AI visibility?

Answer: Test prompts that do not contain the company's name and see whether the brand still appears for questions about the buyer's problem, category or purchase decision. In two Zen Media campaigns, 85% to 87% of the prompts that surfaced the client were non-branded.

Question: How do you track brand position in AI answers?

Answer: First determine whether the brand appeared, then track where it was placed. Coverage and position should be kept separate because a company can appear frequently without leading the answer. In Zen Media's oilfield supply campaign, two engines with similar coverage produced first-place rates of 57.6% and 92.5%.

Question: What should an AI visibility report include?

Answer: It should show coverage and position by engine, performance on non-branded prompts, citation sources, competitor visibility, the prompts being measured and the dates of the measurement window.

Question: Can an agency guarantee visibility on a specific AI engine?

Answer: No. Results change by engine, prompt set, category and measurement window. What can be controlled is how the work is measured, what gets published in response to the findings and whether the same methodology is used to track movement over time.

About Zen Media

Zen Media is a marketing and communications agency that helps brands build visibility, authority and influence across media, search and AI-generated answers. Its AI visibility programs combine buyer-prompt research, owned content, third-party distribution, citation analysis and measurement across ChatGPT, Gemini, Claude, Perplexity and Grok.

Sarah Evans, CEO
Zen Media
sarah@zenmedia.com