Questions people actually ask about AI visibility
What is generative engine optimization (GEO)?
GEO is the practice of getting a company named and cited in the answers AI assistants give to buyer questions. Traditional SEO optimizes for a ranked list of links. GEO optimizes for being one of the three to five companies an assistant names in a sentence.
How is GEO different from SEO?
SEO competes for position in a list of ten blue links, while GEO competes to be named in a single generated answer where only a handful of sources get cited. Ranking well on Google does not mean being cited by an AI assistant — most of the companies we audit rank fine and are invisible in AI answers.
How do I check whether ChatGPT knows my company exists?
Ask it the question your buyer would ask, such as "best commercial HVAC contractor in Phoenix", not your company name. Searching your own name tells you nothing, because the assistant will find you either way. Run it several times, because answers vary between runs — that variance is why a single screenshot proves nothing.
Why does a competitor smaller than us get recommended by AI?
Usually because they appear in the sources the assistant reads — industry directories, review platforms, roundup articles, dated posts — and you do not. Size does not determine citation. Presence in citable sources does.
Is AI visibility actually costing us deals, or is it a vanity metric?
It costs deals when your buyers use AI to build their shortlist. If they narrow to three firms before contacting anyone, being absent means you never enter the process, and you never see the loss because there is no lost RFP to point at.
How long does it take to show up in AI answers?
Measurable movement typically takes 60 to 90 days, because it depends on external sources being updated and re-crawled. Anyone promising results in two weeks is guessing.
What does an AI visibility audit include?
A 0-100 AI Visibility Score built from at least thirty grounded runs across your buyers' real questions, a side-by-side comparison against the competitor being recommended instead of you on the identical runs, a question-by-question scorecard marking each answer Cited, Mentioned, or Absent, a diagnosis of the specific causes, and a prioritized 90-day roadmap. The audit is yours to keep whether or not you continue.
Which AI engines do you measure?
Three: Perplexity, OpenAI, and Gemini. Every question set is run ten times — three runs on Perplexity, three on OpenAI, four on Gemini — and the score is the number of those ten runs that named you. A score is never reported for an engine that was not measured.
Why ten runs instead of one check?
Because AI answers are not deterministic. Ask the same question twice and you can get two different shortlists, so a single check measures nothing but luck. Ten runs turns an anecdote into a rate you can track, and it is the same ten runs your competitor is scored on.
Why does grounding matter?
A grounded run makes the engine search the live web before answering. An ungrounded one answers from training memory, which tells you whether the model absorbed your company months ago — not whether it recommends you today. Measuring the ungrounded version is the most common way an AI visibility report flatters a company that is actually invisible.
How much does GEO cost?
AmppScaling starts with a one-time AI Visibility Audit at $497. Ongoing GEO work is a flat monthly engagement, scoped on a 30-minute call, and the audit fee is credited toward your first month if you continue. There are no long-term contracts.
Can you guarantee results?
On the monthly engagement, yes: a measurable increase in your AI citation share within 90 days, tracked live in your dashboard — or you do not pay for month four. We do not guarantee a specific ranking or a specific number of citations, because AI answers vary between runs and nobody controls them directly.