Share of model tracking for founders
Share of model is how often AI answers to your buying questions name you. Oetzi asks five engines, shows who got named instead, and lists what they cited.
By Max Gillespie, founder of Oetzi
Share of model is the fraction of AI engine answers to your category's buying questions that name you. Oetzi asks each question you track to ChatGPT, Perplexity, Gemini, Grok, and Google AI on a schedule, then shows the share, who was named instead of you, and the threads and pages those answers cited. Each question costs two credits across every engine.
What is share of model tracking?
It is rank tracking for the answer, not the results page. When someone asks an engine which tool to use for the thing you sell, the answer names three or four products, and share of model counts how often one of them is you.
The unit is a buying question, not a keyword. "What is the best social listening tool for a small startup" is a question. Your brand name is not, because nobody who is still choosing types your brand name. Oetzi drafts a first set of questions from your site during setup, and the first check runs alongside the seven day backfill, so you have a number on day one.
Which engines does the AI visibility check cover?
Five: ChatGPT, Perplexity, Gemini, Grok, and Google AI. The same question goes to all of them on each run, which is the only way the number means anything, since the engines disagree constantly. Being named by Perplexity and invisible in ChatGPT is a normal result.
Two credits covers one question across every engine, so a small set of real buying questions on a schedule stays cheap. A free account starts with 10 credits and no card. The pricing page has the rest.
Who gets named instead of you?
Every check records the products the answer did name. That list is often more useful than your own share, because it shows who the engines currently treat as the category, which is rarely the set you think you compete with.
Watching the list over a few weeks shows movement. A name appearing in answers it was absent from a month ago is doing something that worked. The competitor report puts that next to where the same names show up in public threads.
What do the answers cite?
Each check lists the threads and pages the answer pointed at. This is the part that turns a measurement into work. The engines are quoting somebody: a Reddit thread, a comparison post, a review page. Those sources are reachable.
That is where measuring stops being the product. An AI visibility tool that reports a score leaves you to guess the next move. Here, a cited thread is a conversation you can join, so it flows straight into social listening and a drafted reply in your voice, which you approve before anything posts. A cited comparison page is a gap your own posting plan can fill, drafted from your voice doc for one credit per run.
The honest caveat: these answers move. The same question asked twice in a week can return different products, which is exactly why the checks run on a schedule rather than once. A single reading is a snapshot and not a trend.
Questions
What is share of model?
Share of model is the share of AI engine answers to your category's buying questions that name your product. It is measured per question across ChatGPT, Perplexity, Gemini, Grok, and Google AI rather than on a single engine, because the engines return different products for the same question and a one engine reading will mislead you.
How is share of model different from share of voice?
Share of voice counts mentions across media and social. Share of model counts answers, specifically the answers engines give to people who are choosing something. A product can be mentioned constantly and still never appear in the recommendation, and the reverse happens too. They measure different moments, so track both if you like, but do not substitute one.
How much does a share of model check cost?
Two credits per question, and that covers all five engines on that run. Reading the results, the reports, and the cited sources costs nothing. A free account has 10 credits that never expire, and the paid plan is $49 a month with 100 credits included, so a modest set of tracked questions fits inside the monthly grant.
What do I do after I see the score?
Look at the citations, not the number. Each answer lists the threads and pages it drew from, and those are places you can actually act: a Reddit thread you can answer in public, a comparison page you can respond to with a post of your own. Oetzi drafts either one in your voice and holds it for your approval.
Oetzi is an AI marketer for founders who need users. It finds the threads where people ask for a product like yours on seven platforms, checks whether the AI engines recommend you, and drafts replies, posts, and short video in your voice. You approve everything. 100 credits included, 10 free to try with no card, 50 bonus credits with the first payment.
- Oetzi vs Am I Cited: two credit meters, one that draftsAm I Cited meters AI visibility tracking by credit and reports position. Oetzi meters too, but its credits also draft the reply and the post you approve.
- Oetzi vs AthenaHQ: an action center or a marketerAthenaHQ pairs AI visibility with an action center and web analytics. Oetzi measures the same answers, listens on seven platforms, and drafts in your voice.
- Oetzi vs LLMrefs: 11 engines tracked, or 5 engines acted on?LLMrefs tracks 500 prompts across 11 or more engines on one plan. Oetzi tracks five engines, then finds the cited threads, drafts the reply, and drafts the post.
- Oetzi vs Mangools AI Search Watcher: tracker or marketerMangools AI Search Watcher tracks LLM rankings inside an SEO suite. Oetzi measures the same, then finds the cited threads and drafts the reply for you.
- Oetzi vs Otterly.ai: AI visibility tracking, and then what?Otterly.ai tracks how often AI engines name you and stops there. Oetzi tracks the same, then finds the cited threads and drafts the reply. Which fits your week.
- Oetzi vs Peec AI: AI visibility reporting or acting on itPeec AI breaks AI visibility down by engine, country, and language for mid-market teams. Oetzi measures it, finds the cited threads, and drafts the reply.
- Oetzi vs Profound: enterprise AI visibility or a marketerProfound is the enterprise AI visibility platform, billed annually. Oetzi is a monthly plan that measures share of model, then drafts the reply and the post.
- Oetzi vs Promptwatch: crawler logs or the reply you sendPromptwatch adds AI crawler logs and visitor analytics to visibility tracking. Oetzi measures the same answers, then drafts the reply and the post.
- AI marketer for founders who need usersOetzi is an AI marketer for founders. Paste a URL, get scored conversations from seven platforms, replies and posts in your voice, and share of model.
- AI reply tool for Reddit and X, in your voiceTwo reply drafts per conversation in your own voice, held for your approval. Nothing posts by itself, because auto posted replies are what gets accounts banned.
- AI tool that writes posts in my voiceOetzi drafts one to three variants of one post on one platform from your own voice doc, with the day's trend research included, for a single credit.
- Brand voice document that makes AI write like meA voice doc built from your own posts and samples, corrected by sounds like me votes, and used by every reply, post, and video Oetzi drafts. Free.
- Daily trend scan: what to post about todayOetzi runs a trend scan every day and folds the day's research into every drafting run, so posts and video ideas arrive with an angle. The scan is free.