Key Takeaways
- Test whether the answer names anyone before you spend a month on it. Some questions return advice and never a business, no matter who asks. Those cannot be won by anybody, so tracking your absence from them measures nothing.
- Treat your own name as the least useful question on the list. Only people who already know you type it. Getting named there proves the assistant can read your site, which is worth knowing once and never worth optimizing for.
- Prefer the questions where somebody is about to spend money. Cost questions, comparison questions, and who-should-I-hire questions sit closest to a decision, and they are the ones that force an answer to name specific businesses.
- Drop any question you could not honestly win. If there is no real reason a fair answer would name you over the twenty others, no amount of publishing fixes it. Narrow the question until there is a reason.
- Keep the list short enough to actually re-run. A list you check quarterly beats a list of two hundred you check once. The measurement only works if the same questions get asked the same way again.
The first time somebody runs their own questions through ChatGPT, the list they write is almost always wrong. Not badly written. Wrong in a specific and predictable way: it is a list of things they wish they were known for.
So the list contains their methodology, their service names, their own business name, and the word best a lot. Those questions return pleasant answers and mean almost nothing.
Sorting the list is worth more than lengthening it. Most of the work of getting named happens on a handful of questions, and knowing which handful is the whole job.
Which AI questions are worth winning?
The ones a real buyer would type, asked near the moment they decide to spend money, where the answer already names specific businesses. All three conditions have to hold. Miss any one and the question is either unwinnable, unreachable, or not worth the win.
That rules out more of a typical list than people expect, usually somewhere around half.
It also means the sorting can be done in an afternoon, before any publishing work starts, which is the cheapest hour in this entire exercise.
The four kinds of vanity question
Vanity questions are not silly. They are plausible, and that is what makes them expensive. Four shapes account for nearly all of them.
Your own name. Covered properly further down, because it is the one everybody starts with and the one that misleads hardest.
Your service wording. If you sell something you call a clarity intensive, nobody is searching for a clarity intensive. Buyers describe problems, not products, and a question built from your product names is a question only you would ask.
Your methodology. The framework you named, the system you built, the acronym on your slides. Real buyers meet these after they hire you, not before.
Superlatives with no boundary. Best coach, top consultant, leading expert. These return directories and magazines, because the field is too large for an answer to responsibly name individuals. Why directories dominate that shape of question is worth understanding before you fight it.
Unwinnable by design
A buyer asks it, but the answer never names anyone
Useful to write for. Pointless to measure.
Worth winning
A buyer asks it, and the answer names businesses
Spend everything here.
Neither
Nobody asks it, and nothing gets named
Leave it alone entirely.
Vanity
Nobody asks it, but the answer does name businesses
Feels like winning. Reaches nobody.
Only the top right box earns publishing time. The top left is worth writing for and pointless to score yourself against, and both bottom boxes are noise however good the answers look.
The top left box is the one most people get wrong in the other direction. Those are excellent questions to write content for, because buyers genuinely ask them, and they are useless as a scoreboard because nobody is ever named.
The four tests that sort your list
Run each question on your list through these four in order. The first failure ends it; there is no need to run the rest.
| Test | The question behind it | How to check | If it fails |
|---|---|---|---|
| Buyer language | Would somebody outside your business type this? | Read it aloud to a past client and watch their face | Rewrite it in their words, or drop it |
| Nameability | Does the answer name businesses at all? | Ask it three times and count the names that appear | Keep it for content, remove it from your scoreboard |
| Distance to a decision | Is this person close to spending money? | Ask what they would do in the next ten minutes | Deprioritize it behind the questions that are closer |
| Honest winnability | Is there a real reason you would be the answer? | Name the reason in one sentence, out loud | Narrow the question until a reason exists |
The fourth test is the uncomfortable one and it saves the most time. If you cannot say in one sentence why a fair answer would name you rather than the other twenty, publishing more pages will not supply the reason.
Why your own name is the most misleading question
Because it is the one question you are guaranteed to eventually win, and winning it tells you almost nothing.
Only two groups type your business name: people who already know you, and you. When an assistant describes you accurately, what you have learned is that your site is reachable and readable. That is genuinely worth confirming once, because a failure there means something is broken, and it is worth nothing as an ongoing measurement.
The reason it feels like progress is that it is the only question where you can see yourself improve quickly. Systems that fetch pages live, as Google describes in its account of how AI features in Search work and as OpenAI documents for its own crawlers, can pick up a rewritten about page within days. A buyer question that names three competitors will not move on that timescale, and the contrast quietly trains people to keep checking the easy one.
Check your name once. Write it down. Then spend the rest of your attention on questions where somebody who has never heard of you is doing the asking. What a realistic score looks like on those is covered in what a good AI visibility score looks like.
How to build a list worth measuring
Start from the last ten people who paid you and write down what each of them was worried about the week before they got in touch. Not the service they bought. The problem they had. Those sentences, lightly tidied, are your list.
Add the cost question and the comparison question for your category, because both sit close to a decision and both force an answer to name somebody. Then attach a place or a stage to anything that reads too broadly, which is usually what turns an unwinnable question into a winnable one. The full method for assembling and pruning a set is in how to pick the 25 questions that matter.
Run each survivor a few times before you commit to it, and record who gets named. The Princeton generative engine optimization study is a useful reminder that what gets pulled into an answer depends heavily on how quotable the source material is, which means a question with three strong published answers behind it is a harder fight than a question with none. Reading the resulting patterns is covered in what to do with your AI test results.
One honest limit worth carrying. This sorting is judgment, not arithmetic, and reasonable people will disagree about which box a question belongs in. The value is not in getting every call right. It is in noticing that half the list was never winnable, which is a conclusion nearly everybody reaches and nearly nobody reaches before spending a quarter on it.
If you would rather start from a set somebody else has already sorted, the free check asks ChatGPT 25 real buyer questions from your niche at $0 and emails a plain-English readout naming who got named in your place. It runs as soon as you ask for it, and the report lands within the hour between 7am and 5pm Central.
Questions we hear the most
Which AI questions are worth winning?
The ones a real buyer would type, close to the moment they spend money, where the answer already names specific businesses. Cost questions, comparison questions, and who-should-I-hire questions in a named place usually meet all three conditions.
What makes an AI question a vanity question?
Nobody outside your business types it. Your own name, your exact service wording, and your methodology all return flattering answers that reach only people who already know you, which means winning them changes nothing.
Should I track whether ChatGPT knows my business name?
Check it once, then stop. It tells you whether your site can be read at all, which is worth confirming. It tells you nothing about whether strangers looking for what you sell will ever find you.
How do I know if a question can be won at all?
Ask it a few times and see whether any business gets named. If every answer is general advice with no names in it, that question has no winner and your absence from it is not a problem to fix.
How many questions should I track?
Enough to cover your main services and buyer situations, and few enough that you will re-run them. For most solo businesses that lands between ten and thirty, checked on a fixed schedule rather than whenever you feel curious.
Are broad questions like best coach worth chasing?
Rarely. Superlative questions tend to return directories, magazines, and well-known brands, and the field is enormous. A narrower question with a place or a stage attached is far cheaper to win and far closer to a buyer.
How does AnswerHalo choose which questions to ask?
We build the set from how buyers in your niche actually phrase things, not from your service names, and we favor questions where answers already name businesses. The set is reviewed and approved before anything is run.
What does the AnswerHalo free check include?
It asks ChatGPT 25 real buyer questions from your niche and emails you a plain-English report: your score, who got named in your place, and the one fix to start with. It costs $0 and it runs as soon as you ask for it.
Your buyers are already asking. Find out what AI tells them.
The free check asks ChatGPT 25 real buyer questions about your niche and sends you the report within the hour, at $0. If it shows you are already getting named everywhere, we will say so in plain words.
Get my free check →