What ChatGPT, Claude, Perplexity, and Gemini say about you when your buyer asks.
A free, 12 to 15 minute self-audit for B2B owners. Enter your company, what you sell, and your geography. The tool generates twelve copy-ready prompts, you run them in the AI tools your buyers use, and the page returns four scores plus a paste-ready headline finding. No login. No email gate. Fresh on every visit. Built by Tim Doelger, founder of Get 'er Done.
Most of the buying decision happens before your reps ever get a calendar invite.
B2B buyers complete the majority of their independent research before they will talk to a sales rep. Gartner has reported that B2B buyers spend only about 17% of their total purchase time meeting with potential suppliers, with the remainder spent on research, internal meetings, and supplier comparisons. (Gartner, "The B2B Buying Journey")
That research now runs through AI assistants as much as through Google. When a CFO asks ChatGPT for a vendor shortlist, your company is either named or it is not. When the AI describes what you do, the description is either accurate or it is invented. This tool tells you which one is happening, in fifteen minutes, with no email gate.
The output is four scores, each one a different finding. Discovery (does the AI know you exist), Shortlist (where do you sit on a comparison), Credibility (does the AI describe you accurately), and Risk (is the AI making things up). Each score points at a different fix.
Tell us three things. We generate the prompts.
This stays in your browser. Nothing is sent to a server. Refresh the page and it is gone.
Run each prompt in two AI tools. Score what comes back.
Use ChatGPT, Claude, Perplexity, or Gemini. The free versions are fine. Copy a prompt, paste it in, then mark the answer below.
Four possible findings. Each one points at a different fix.
Invisible (Discovery or Shortlist below 30)
The AI tools your buyers use do not return your company when asked who to talk to. This is a structured-data and entity-authority gap. Schema markup, founder and company disambiguation, and citation-attractive content patterns are missing or weak. The fix is implementation, not more marketing spend.
On the radar, not on the shortlist (Discovery 30 to 70, Shortlist below 50)
AI tools know you exist, but they do not put you in the top three when a buyer asks for a comparison. Competitor recommendations are stronger. The fix is competitive differentiation signals in your schema (slogan, alternateName, unique mechanism), plus contrast copy that explains how you differ from named alternatives.
Visible but misrepresented (Risk score below 60)
AI tools mention you but invent pricing, fabricate case studies, or describe customers you do not have. This is more dangerous than invisibility. A buyer reading wrong information about you arrives confused or skeptical. The fix is publishing canonical facts (pricing tiers, real case studies, named customers) in structured form on your own pages so AI tools have authoritative ground truth.
Winning the pre-contact phase (all four scores above 70)
AI tools name you in discovery queries, place you on the shortlist, describe you accurately, and do not invent facts. Hold the position. Re-run this audit in 90 days. AI search rankings shift faster than Google rankings, and competitors who are investing in structured data will move on the leaderboard quarterly.
Common questions about the check.
What does this tool actually do?
It generates twelve diagnostic prompts using your company name, what you sell, and your geography. You run each prompt in at least two AI tools (ChatGPT, Claude, Perplexity, or Gemini), then mark whether the AI mentioned you, what rank you held, and whether the answer was accurate. The page produces four scores in real time (Discovery, Shortlist, Credibility, Risk) plus a single paste-ready headline finding describing your biggest gap. No login. No email gate. Fresh on every visit.
How is this different from a Google SEO audit?
Traditional SEO audits measure whether Google shows your page when someone searches a keyword. This audit measures whether AI assistants name your company as a recommendation when a buyer asks a research question in plain English. Different audience, different signal layer. AI assistants read JSON-LD schema, entity graphs, and structured facts, not just keywords and backlinks. A site that ranks well on Google can still be invisible to ChatGPT and Claude if the entity-level signals are missing.
Why does the Risk Score weight hallucinations heaviest?
Invisibility is a missed opportunity. Misrepresentation is an active liability. If ChatGPT tells a prospect your pricing starts at one number and you actually charge a different number, that prospect arrives confused or angry, and the deal is harder to close than if they had never heard of you. Invented case studies and fabricated customer lists are worse, because they expose the buyer to information you cannot stand behind. The Risk Score isolates this so you know whether the priority is becoming visible or correcting what is already being said.
How often should I rerun this check?
Every 90 days. AI search rankings shift faster than Google rankings because models retrain, indexes refresh, and new competitors get cited monthly. A green result today can fade to yellow in a quarter if competitors invest in structured data and you do not. The check takes 15 minutes. Calendar it quarterly.
What if my scores are red? What is the next step?
A low Discovery or Shortlist score means the structured-data and entity-authority signals on your site are insufficient for AI tools to identify and recommend you. That is a content-architecture problem, not a marketing problem. The Agent Found service is the paid Get 'er Done engagement that implements schema markup, entity disambiguation, citation-attractive content patterns, and 90-day reassessment to fix it. A low Risk Score means AI tools are inventing information about your company; the fix there starts with publishing the canonical facts (pricing, case studies, customers) in structured form on your own pages so AI tools have authoritative ground truth to draw from.
Ran the check and the scores were mixed or low?
Discovery and Shortlist gaps are a structured-data and content-architecture problem. Risk gaps are a canonical-facts problem. Agent Found is the paid Get 'er Done engagement that fixes both. If your scores were high, run this audit again in 90 days. AI search rankings drift faster than Google rankings.