Glossary
LLM visibility
LLM visibility is how often large language models such as ChatGPT name or cite your brand when people ask questions in your category. It is measured by asking a fixed set of prompts and counting mentions and citations, not by a rank.
Updated
Ask ChatGPT "what are the best linen sheets for hot sleepers?" and read the reply. If your brand is named, or your page is one of the sources, that answer counts towards your LLM visibility. If not, it counts against. LLM visibility is the share of such answers, across a set of questions buyers really ask, in which a large language model mentions or cites you.
It has no rank and no single number that stays still. Answers change with the wording, the model, whether search is on, and the day. So the measurement is a sample: a fixed list of prompts, asked more than once, with a dated record of who was named and which pages were cited. The term is popular with tool vendors because it names what they sell, but the idea is simple enough to run by hand in a spreadsheet before you pay for anything.
LLM visibility vs AI visibility vs SEO visibility
AI visibility and LLM visibility are used for the same measurement: whether assistants name you when buyers ask. Vendors who say "LLM" tend to stress the model; vendors who say "AI" often include surfaces such as Google AI Overviews that sit inside a results page. When you compare tools, ask which surfaces each one actually queries rather than reading meaning into the label.
SEO visibility is older and different. It is usually a score built from where your pages rank for a set of keywords, weighted by search volume and an estimated click-through rate, so it describes a list of links. LLM visibility describes a written answer, which may name you without linking you, link you without naming you, or leave you out even when you rank. The two are related, because assistants with search on usually read pages that already rank, but a good SEO visibility score does not tell you what an assistant said.
How LLM visibility is measured
Four pieces make a measurement worth reading.
Prompt set. A fixed list of questions written the way buyers ask them, without your brand name, covering the category, the problem and comparisons. Keep it stable, and change it only with a dated note, or the trend line means nothing.
Mentions and citations. A mention is your brand name in the answer text. A citation is a source link the assistant shows. Record them separately: an answer can name you while citing a review site, or cite your blog post while recommending someone else.
Share of voice. Of all the brand mentions across the prompt set, the share that are yours. It shows where you stand against the competitors the assistant names, which a raw mention count cannot.
Sampling variance. The same prompt asked twice can return different brands. Ask each prompt more than once per run, report results as "named in X of Y valid answers" with the counts visible, and record a failed or refused answer as unavailable rather than as a miss. A change of one or two answers in a small panel is noise until it repeats.
What an LLM visibility tool does
An LLM visibility tool automates that loop. It sends your prompt set to one or more assistants on a schedule, stores the answers, detects which brands and domains appear, and charts mentions, citations and share of voice over time. Some add sentiment, the specific pages cited, or content suggestions.
What to look for before paying: which assistants it queries, and whether it uses an API or the consumer app, since API answers can differ from what a shopper sees; whether you can write and export your own prompts; whether raw answers are kept so you can read them, not only scores; how many times each prompt is asked per run; how failed answers are counted; and what the score's denominator is. The best LLM SEO tools guide and the AI Overview trackers guide compare current products, and the AI visibility tool evaluation guide turns these questions into a checklist.
AI brand visibility for a store
"AI brand visibility" is the same idea seen from the brand's side, and for a store it splits into two questions. Discovery: when a shopper asks an unbranded question, such as "best refillable deodorant", are you among the brands named? Accuracy: when a shopper asks about you by name, does the assistant get your products, prices, materials and policies right? Measure them with separate prompts, because a store can be described well when asked about directly and never be named otherwise.
What moves either is mostly ordinary work. The assistant's search crawler, such as OAI-SearchBot, has to be allowed to fetch your pages. Pages need to answer one buyer question plainly near the top, with product facts that match the catalogue everywhere they appear. And other sites, such as reviews, press and community threads, have to mention you, because assistants repeat what the wider web says.
How to check LLM visibility for free
Start by hand. Write ten buyer questions without your brand name, ask them in ChatGPT with search on, ask each one twice, and log the date, the question, whether you were named and which pages were cited.
Two free Citelift tools help. The free check needs no sign-in and asks ChatGPT three buyer questions for your category, separating brand mentions from cited links. The manual checker prepares templated questions and analyses an answer you paste, in your browser, without calling an AI model. The AI visibility measurement guide shows how to turn the log into a repeatable panel.
What LLM visibility cannot tell you
It is a sample, not a census. No tool sees every conversation, API answers can differ from what the consumer apps show, and one day's answer says little about the next. It does not measure traffic or sales: a mention is not a visit, and a visit is not an order, so keep referral sessions and attributed orders in their own columns. It cannot show what is in a model's training data. And no tool can promise a mention, because the assistant chooses its sources; treat a guarantee of citations as a warning sign.
Questions.
What is LLM visibility?
How often large language models such as ChatGPT name or cite your brand when people ask questions in your category, measured across a fixed set of prompts rather than as a rank.
Is LLM visibility the same as AI visibility?
In practice, yes. Both describe whether assistants name you. Check which assistants and surfaces a tool queries rather than relying on the label.
What is a good LLM visibility score?
There is no standard scale. Compare yourself with your own earlier runs and with the competitors named in the same prompts, and only read a score with its prompt set, run count and denominator beside it.
Can I check LLM visibility for free?
Yes. Ask ten unbranded buyer questions in ChatGPT, twice each, and log the results. Citelift's free check at /check asks ChatGPT three buyer questions for your category without a sign-in.
Citelift is listed on the Shopify App Store: Citelift on the Shopify App Store.