What personal AI visibility measures
Define a time-bound observation of answers about a person’s expertise.
On this page
Personal AI visibility is a dated observation of how selected AI engines answer selected questions about the expertise you provide. It is evidence from a defined test, not a verdict on your ability or future work.
A person can do excellent work and fail to appear in a small set of answers. A person can also appear in an answer that is vague, mistaken, or unrelated to the service they want to sell. Ask what a particular engine returned when it received a particular client-like question at a particular time.
Gistful currently checks ChatGPT, Gemini, Claude, and Perplexity. A scan keeps the saved question, engine, date, profile scope, returned answer, and cited public sources when they are available. It also keeps missing or failed checks separate from negative findings. Gistful’s methodology explains the boundaries of that record.
Start with the observation
Suppose you advise B2B software companies on pricing. A useful question might be: “Which independent pricing strategists help B2B software teams in the United States?” A scan records what each included engine says in response to that question. It does not test every variation a buyer might use, every country, every account state, or every future answer.
That scope gives the result its meaning. Without the question, date, and engine, “I showed up” or “I did not show up” has little value. A scan becomes reviewable when someone can see what was asked and what came back.
The same principle applies to an apparent recommendation. An engine may list a name, describe a person, cite a page, ask for more context, or fail to return an answer. Those are different observations. Treating them as one number erases the detail you need to decide what to check next.
Visibility is not competence
Visibility measures the returned sample. It does not measure whether you are good at your work, whether clients trust you, or whether a potential client will hire you.
An answer can omit a qualified person because the question was too broad, the stated geography changed the answer, the engine had different available information, or the response did not produce candidates at all. A mention can also be weak evidence. It may identify the wrong person, name someone without a reason, or refer to work you no longer offer.
Google gives a similar warning from the search side. Its guidance says eligibility and adherence to best practices do not guarantee that content will be crawled, indexed, or served. It also warns against producing large volumes of pages merely to manipulate generative results. The durable work is clear, useful, original material for a real reader, not manufactured mentions. Google’s guidance is useful here because it describes its own search features, not a promise about the other engines in a Gistful scan.
Record the fields that make the result usable
Before you interpret a scan, write down its scope. Use this short record:
| Field | What to record | Why it matters |
|---|---|---|
| Question | The exact saved wording | Small wording changes can change what an engine answers. |
| Intent | The client decision behind the question | A name-recognition question is different from a recommendation question. |
| Engine | ChatGPT, Gemini, Claude, or Perplexity | Engine results should remain visible rather than blended into an unnamed total. |
| Date | When the answer was collected | Answers can change over time. |
| Locale and profile scope | Requested geography, language, and facts supplied | These details can shape the answer. |
| Answer state | Returned; missing or failed; or identity review needed | Missing or failed evidence is unmeasured. |
| Evidence | The answer and cited links, where available | A result needs inspection before it supports a decision. |
The record prevents a later rescan from being framed as an improvement when the only change was the questions, locale, or engine coverage.
An illustrative reading
One returned answer might name you and link to your service page. Open it: does the page describe the service accurately, and does the answer connect you to the right work? Another returned answer might say it needs more detail before it can help. It belongs in the returned-answer record, but it does not establish a supported mention. Keep missing or failed checks unmeasured. A third answer might list your name while the surrounding text refers to someone else with the same name. Review it before treating it as identity-supported evidence.
None of these examples proves why the answer took that form. A single observation cannot identify a ranking factor or prove that publishing a page caused a later change.
Use the result to choose a bounded next step
Read the answer before you change anything public. If the questions are too broad, improve the saved question set and label the next scan as a new baseline. If your public profile gives an old service description, correct the profile because readers need accurate information, not because a scan guarantees a payoff. If an answer names you for the right work, check the cited page and preserve the evidence.
Then keep the original question set when you want a comparable rescan. Gistful compares scans only when the saved questions, methodology, and relevant scope match. A changed scope can still be useful. It just answers a different question.
The practical artifact is the scope record above. Save one copy with every scan. When you revisit the result, it will tell you what you measured and what you did not.