When shoppers ask AI what to buy, is your brand the answer?
The AI Shelf Snapshot is a free check of whether consumer brands show up in buyer-style prompts on the major AI assistants. We run real buyer-style questions through ChatGPT, Claude, Perplexity and Gemini for your category and send you a mini scorecard showing whether your brand is in the answers, and who is winning them instead.
In plain language: a buyer-style prompt is what a shopper actually types when they're deciding what to buy, like "what's the best prebiotic soda for someone cutting sugar?" It's not a search for your brand name. It's the question where an AI assistant picks two or three brands to recommend and leaves everyone else out. The Snapshot tells you, for free, whether you're one of the brands it picks.
How it works
- Tell us your brand. Your category and your top two competitors. Takes about two minutes.
- We ask AI like a shopper would. Five real buyer-style questions for your category, run through leading AI assistants like ChatGPT.
- You get the scorecard. A one-page traffic-light summary in your inbox within two business days: are you cited, where you rank, and who wins instead.
Four things you'll know
One line, plain: are you in the answers or not, and how good or bad it is.
You against the competitors AI actually names. Most founders meet a competitor they did not know they had.
What the assistants actually said in your category, from real runs. Screenshots kept on file.
What is driving the answers, and the first fixes we would make. The full audit turns these into a sequenced roadmap.
A note on method: AI answers move between models, sessions, and weeks, so one check proves little. We run every question cold with no brand context and score direction, not decimals. When we ran prebiotic soda in August, two brands owned 9 of 10 answers. Every category has a line like that, and the Snapshot shows you which side of it you're on.
Why it matters
AI answers usually name only two or three brands. Shoppers who ask ChatGPT, Claude, Perplexity, or Gemini what to buy never see the rest. The Snapshot tells you which side of that line you're on. It's the same testing method we use in the full AI Visibility Audit, scoped down to a free first look.
No sales call required, ever. You'll also get a short email series explaining your results and occasional, genuinely useful AI tips. Unsubscribe anytime.
The prompts we run, and what comes back
No mystery box. Here are the five buyer-style prompts we'd run for a prebiotic soda brand, and the shape of the scorecard that comes back two business days later. Yours uses your category and your competitors.
The five prompts
- What's the best prebiotic soda to buy right now?
- I'm cutting sugar but I miss soda. What should I drink?
- Olipop vs Poppi: which one should I buy?
- What prebiotic soda should I grab at Target?
- Is [your brand] any good? What do people say about it?
Four of the five never mention your brand. That's deliberate. If we named you in the prompt, the model would talk about you and you'd learn nothing about whether it recommends you on its own.
The scorecard you get back
Illustrative example using public brands, not a real client report.
Where it goes from here
The Snapshot is deliberately small: five prompts, one page, free, no call. It's enough to tell you whether you have a problem. It is not enough to fix one. Here's the honest progression.
AI Shelf Snapshot
5 prompts · 1 page · 2 business days
Tells you whether AI assistants recommend you in buyer-style prompts, and who they recommend instead. Direction, not decimals.
AI Visibility Audit
20-30 prompts · 4 models · ~1 week
The full diagnosis. Everything the Snapshot does at proper sample size, plus an AEO technical audit of your site and a prioritized fix roadmap your team keeps. $975 through Sept 1, then $1,950.
See the audit →Share of Model Tracker
Monthly · cancel any time
Proves the fixes worked and catches it when models change their minds. Requires a completed audit. Most brands take the audit and first three months together for $2,595.
See the tracker →Questions about the Snapshot
What AI assistants do you check?
ChatGPT, Claude, Perplexity, and Gemini. For the free Snapshot we run your five prompts across the leading assistants in your category rather than all four exhaustively, because the goal is a directional read in two business days. The paid AI Visibility Audit runs the full 20 to 30 prompt set across all four, in both default and web-browsing modes where the assistant offers both, and adds Google AI Overviews.
What counts as visibility?
Being named in the answer. We score four things. Citation rate: how many of the prompts name your brand at all. Position: whether you appear in the first three brands mentioned, since that is roughly where a shopper stops reading. Accuracy: whether what the assistant says about you is actually true, because being described wrongly can be worse than being absent. Competitor share: which brands take the answers you don't. A mention buried in a list of twelve is not visibility, and we don't score it as one.
Is the AI Shelf Snapshot really free?
Genuinely free, and no call is required at any point. The limit is capacity: five brands each month, because a human runs and reads every one. In exchange we ask for your email and we send a short series explaining your results, plus occasional AI tips. Unsubscribe any time and keep the scorecard.
How is the AI Shelf Snapshot different from an AI visibility tool?
Tools like Profound, AthenaHQ, Scrunch AI and Otterly.ai are monitoring platforms: you subscribe, connect your brand, and get a dashboard that tracks visibility continuously. They're good at watching a number over time. The Snapshot is a one-off read with a human interpreting it, aimed at answering a prior question, namely whether you have a problem worth instrumenting at all. If the answer is yes and you want continuous tracking, we'll tell you so, and we're happy to say when a subscription tool is the better buy for you.
Why only five prompts if one AI answer is close to a coin flip?
Fair challenge, and the honest answer is that five prompts give you direction, not a measurement. If you're named in zero of five or five of five, that's a real signal. If you land in the middle, the correct conclusion is "worth measuring properly," which is what the 20 to 30 prompt audit is for. We'd rather say that plainly than dress up a small sample as a verdict.
Do I have to be a big brand to get a Snapshot?
No. The Snapshot is built for challenger and mid-size consumer brands, roughly $5M to $500M in revenue, in food and beverage, beauty and personal care, household, and wellness. Smaller brands are welcome too; the test works the same way, and knowing early is worth more when you're still deciding where to put your effort.