Questions and answers.

When an AI answers "what's the best [category]," it names a few brands in order. That shortlist is the shelf. Your shelf space is how often, and how high, you appear on it, measured across hundreds of answers on each AI assistant.

ChatGPT, Google Gemini, Google AI Overviews, Google AI Mode, Microsoft Copilot, Perplexity and Claude. Exact model versions are re-verified every edition and pinned in the methodology change log.

On the roadmap. We add an assistant when we can sample it rigorously: repeatedly, at scale, as shoppers see it.

Because AI answers are probabilistic. Independent research found fewer than 1% of identical prompts return identical brand lists, and it takes on the order of 60 to 100 runs to get a stable picture. One-run tools are reporting noise. We sample every shelf hundreds of times on every assistant. Details in the methodology.

The public Index publishes a new edition every quarter, and every edition states the week its data was collected. Subscribers' shelves are re-sampled weekly, and alerts fire on confirmed movement, not on a calendar.

Three ways. They're horizontal (any brand, any industry) while AI Shelf Space tracks one vertical, supplements, with every question hand-curated per shelf. They mostly sample each prompt once; we sample every shelf hundreds of times per assistant, because single runs are noise. And they report abstract scores; we report rank on a shelf against named competitors, with your products on the rows. If a horizontal tool fits your team better, use one. But check its sample size first.

The category calls it AI visibility, Answer Engine Optimization (AEO) or Generative Engine Optimization (GEO) tracking. We call it shelf measurement: same problem, stricter method.

The same place it is with Nielsen, SPINS or Ahrefs: everyone can buy the ruler, and the edge goes to whoever reads it first and moves first. Our alerts exist precisely so that's you. The shelf gets re-ranked whether anyone subscribes or not. The risk isn't sharing the ruler; the risk is being the brand that never looks at it.

Don't trust us. Check the receipts instead. The Index is public. The methodology and its denominators are public. The checker runs on your own brand. And the first 30 days are covered by a full money-back guarantee.

We map your catalog to our shelves and your dashboard is live within two business days, with the current edition's data in it.

No. There is the free checker, the public Index, and a 30-day money-back guarantee on both plans.

Both come with a 30-day money-back guarantee: email us within 30 days and we refund you in full. After that, Yearly Intelligence is a yearly plan: cancelling means not renewing, and access continues to the end of the paid year. See the refund policy.

Write to us with your questions and you get a written answer within a business day, from the person who built the measurement.

No. Subscriber dashboards are private. The public Index shows only category-level rankings computed from our own sampling, data that exists whether or not anyone subscribes.

No. AI Shelf Space measures the public AI assistants, not your internals. The only thing we take from you is your product catalog (public anyway) so we can map your products to shelves.

From our own sampling, run for every edition. No vendor-supplied numbers, no self-reported data, no pay-for-placement. Method, assistants, sample sizes and a change log are in the methodology.