Answer engines: getting named in AI answers, run as a measured practice
Answer engines
As you read this
Right now, a buyer is typing one of these questions.
There's no page two here, and no ad slot to buy your way into. The model answers with two or three names and the buyer moves on, so if yours isn't one of them, the conversation was over before you knew it was happening. Most brands have never had a way to find out where they stand. That's the first thing we fix.
The approach
Getting named in AI answers is a measured discipline.
Most teams treat answer engines as a tactic to bolt onto search. It is not. Being named in the answer is a different game from ranking: run as a practice, with an instrument on the table. We read what the engines your customers ask say about you today, find who they name instead and why, and close the specific gap. What follows shows both, not a promise about them.
How we keep the work honest
We hold the surface to the number your P&L actually feels: blended CAC, read as MER across the whole engine, not one channel’s self-reported byproduct. And we show that answer-engine visibility is genuinely bringing in new customers the only way contribution can be proven, by holdout and incrementality testing, not by trusting a platform’s scorecard. The instrument makes the surface visible; the testing makes the contribution defensible.
The evidence
Why don't AI answers obey the old search rules?
These answers are not assembled the way a results page was ranked. The published research is specific about how the engines choose, and it points somewhere brands rarely control. Command of that evidence is what separates running this surface from claiming to.
In Ahrefs’ study of 1.4 million ChatGPT prompts, the strongest citation signal was similarity between a page’s title and the question asked: the engines reward pages that answer it directly. And because most citations go to independent, third-party sources, what the web says about a brand now counts for more than what the brand says about itself.
“The question has changed. It’s no longer where you rank. It’s whether the machines your customers ask actually know you, trust you, and say your name.”
The full research behind it is laid out in what an AI-native growth agency is. This page is the room where we run it.
What reads the number
Pollymetric, our native Shopify app.
The work above only counts if the number moves, so we bring our own instrument to watch it. Pollymetric installs into your store, asks the major AI models a hundred-plus real buyer questions every week, and reports back what they said: how often you're named, who's named ahead of you, how you're described, and which sources the models leaned on. Your team doesn't run anything. It just reads.