Sample match report
Which questions show an AirOps ad in an AI chat?
In Prism's own test on October 6, 2026, 1 of 10 visitor questions cleared the 0.65 match floor and showed the AirOps ad as an Affiliate ad card. The other 9 did not, so a card might not surface for them. These are Prism's own scores and say nothing about how any other platform behaves.
How did Prism score each question?
Prism turns the visitor's question into a bag of words and compares it with the words in the ad's headline, description and keywords. The result is a cosine score between 0 and 1. A card shows only when the score reaches 0.65. The ad's topic label is not part of the score.
The same code runs in the live chat and in this test, and the ten questions are a fixed set kept in one file. A message that touches a sensitive subject gets no card at all, whatever it scores. This table scores the ad on its own against Prism's affiliate library. In the live chat, an approved paid ad that clears the floor is shown ahead of any affiliate ad.
What did each of the 10 questions score?
| # | Visitor question | Shown | Score | Floor |
|---|---|---|---|---|
| 1 | How do I get my content cited by ChatGPT? | Yes | 0.737 | 0.65 |
| 2 | Why does AI search never cite my blog posts? | No | 0.089 | 0.65 |
| 3 | How can I track whether AI engines mention my brand? | No | 0.288 | 0.65 |
| 4 | What makes content more likely to be cited in AI answers? | No | 0.433 | 0.65 |
| 5 | Is there a tool to help my articles get cited by Perplexity? | No | 0.384 | 0.65 |
| 6 | How do I write content that AI search engines will quote? | No | 0.356 | 0.65 |
| 7 | What is generative engine optimization? | No | 0.136 | 0.65 |
| 8 | Can I measure my AI search visibility? | No | 0.136 | 0.65 |
| 9 | What is a good subject line for a newsletter? | No | 0.000 | 0.65 |
| 10 | How much protein should I eat after a workout? | No | 0.044 | 0.65 |
1 of 10 questions showed a card. Questions that are unrelated to the ad score near zero, which is the intended result.
Which questions came closest to missing?
The narrowest pass was "How do I get my content cited by ChatGPT?" at 0.737. Questions that missed by the smallest margins might not surface a card, even though a person would see them as related to the ad:
- "What makes content more likely to be cited in AI answers?" scored 0.433, 0.216 below the floor.
- "Is there a tool to help my articles get cited by Perplexity?" scored 0.384, 0.265 below the floor.
- "How do I write content that AI search engines will quote?" scored 0.356, 0.293 below the floor.
A question scores higher when its words overlap the ad's own wording. Different words for the same idea score lower in Prism's engine, and that is a property of word-overlap scoring, not a verdict on the product.
What ad did Prism test?
This is the ad exactly as stored in Prism's library. Prism did not write or change the wording.
- Label
- Affiliate ad
- Headline
- Stop guessing. Start getting cited.
- Description
- AirOps tracks your brand across AI engines and helps your content get cited.
- Button
- See AirOps
- Keywords
- how do I get my content cited by chatgpt, get cited by ai search engines, ai content visibility tracking, why is my content not cited by ai
- Link
- AirOps (affiliate link)
Frequently asked questions
Is this a real ad?
The ad text on this page is the AirOps affiliate ad as it is stored in Prism's ad library. Prism earns a commission if a visitor buys through the link, and every card is labelled "Affiliate ad".
Does a high score mean the ad will show on ChatGPT or Google?
No. The scores come only from Prism's own matching engine. Prism has not measured any other platform and makes no claim about them.
Can I test my own ad the same way?
Yes. Paste your ad into the chat on the Prism homepage and ask questions your customers would ask. Testing is free.
Some links on this page are affiliate links. Prism Publication may earn a commission at no extra cost to you. See the other sample reports or test your own ad.