Buyers ask ChatGPT and Gemini for a shortlist before they ever visit your site. MentionX shows whether you're on it, who is instead, and what your team is doing about it.
AI answers now shape which vendors buyers consider. MentionX measures how often AI engines recommend your brand for real buyer questions, who they recommend instead, and how that changes week to week, in a deck you can take straight to the board.
Recommendation Rate: the share of AI answers that recommend you, across every engine in your plan.
Recommendation Rate →“Who is winning instead?”Share of voice by competitor, with the prompts and engines where each one beats you.
Competitor Tracking →“Is the team's work moving it?”Weekly scans and a Boardroom deck that links every chart to the stored answers behind it.
Boardroom Deck →Competitors keep getting named by AI. Are we on that shortlist or not?
34% this quarter, up from 22%. Kestrel still leads at 41%. Here are the three gaps we close next.
Example conversation. Invented people and numbers.
One scan shows your Recommendation Rate, rivals and the sources AI trusts.
The Action Board turns gaps into owned actions for SEO, content and PR.
Weekly scans on the same prompts show real movement, not noise.
A Boardroom deck for the quarterly review, with evidence on every slide.
“We think we're doing fine in AI search. Traffic from AI referrals is small.”
“AI recommends us in 34% of buyer answers, up from 22% last quarter. Kestrel still leads on compare prompts; here's the plan.”
Example only. Invented numbers to show the format.
4 engines, 200 prompts per scan, Boardroom decks and Mentionie for the team: enough coverage to answer the shortlist question for a whole category.
Measurement, reporting and how much team time it takes.
Start with Recommendation Rate: how often AI engines recommend your brand for real buyer questions. Read it with share of voice against competitors and the sources engines cite.
Your first scan gives you a baseline as soon as it finishes; how long that takes depends on the engines and prompts in your plan. Movement from content and source work then shows up in weekly scans on the same prompts.
Yes. Boardroom exports a ready-to-present deck from your scan, and every slide traces back to stored AI answers.
No. The formulas and weights are published on the methodology page and calculated in code, not by an AI model.
No. Seats are unlimited from Spark (Starter includes 3), and the Action Board ranks what to do first so a small team can focus on the changes that move the most answers.
Enter your URL and market. Your first scan answers it on every engine in your plan.