gaash.ai

Measurement & Reporting · 9 min read · July 15, 2026

Measuring AI mentions: which insurance questions actually lead to mandates

SCORE

Not every AI mention is worth the same. Whoever shows up in ChatGPT for "best insurance broker Munich" might pick up a click or two, but mandates come from specific questions about specific lines and specific situations. This guide shows you, as a broker, how to measure which AI answers actually mention you, which of those mentions turn into paying clients, and which insurance questions deserve your attention first.

Why AI visibility works differently for brokers than it does for Google

On Google, a prospect clicks through ten blue links and compares them for themselves. In ChatGPT, Gemini or Perplexity, they get a single finished answer with one to three recommended contacts instead. That changes your business fundamentally: if the AI names you for "I'm self-employed — which occupational disability insurance fits me, and who advises on it independently?", that doesn't just replace a click, it replaces half of an initial consultation. The prospect arrives pre-qualified, already trusting you, with a concrete line of business already in mind.

That's exactly why simply being mentioned isn't enough. A mention for the generic question "What does an insurance broker do?" brings you almost nothing. A mention for "Which broker in Augsburg specializes in medical professions and switching to private health insurance?" brings you a mandate with a real chance of closing. Measuring AI visibility means keeping these two cases cleanly apart, and building only the second kind on purpose.

As a broker, you actually have an advantage over the big insurers here. Large insurers answer product questions, but AI models tend to recommend independent, specialized advisors for advisory and trust questions. If your website, your specialist articles, and your reviews clearly show what you focus on, you become the obvious recommendation for exactly these valuable questions — where an insurer can only point to a product.

The difference between reach questions and mandate questions

Split the questions you could be mentioned for into two buckets. Reach questions are broad and informational: "How does term life insurance work?" or "What does private health insurance cost?" They generate a lot of volume, but the person asking is rarely ready to buy. Mandate questions are narrow, situational, and local: "I'm 34, a doctor, employed, and want to switch to private health insurance — which independent broker in Cologne can help me with that?" Questions like that already carry buying intent, a line of business, and a region built in.

For your business, almost only the second bucket matters. A client who arrives through a mandate question already has a rough idea of what they need and is actively looking for an advisor. That's exactly where the conversations start that turn into ongoing mandates — occupational disability cover for the self-employed, commercial insurance for trades businesses, retirement planning for high earners. So don't measure your total visibility. Measure your visibility per mandate question.

In practice, that means writing down 20 to 30 typical mandate questions your ideal customers would actually ask — exactly as a real prospect would type them into ChatGPT. Use their words, not your jargon. "Protection against loss of income" becomes, for the customer, "What happens to my salary if I can't work anymore?" This list is the foundation for everything else in this guide.

How to measure, concretely, where the AI mentions you

The simplest method costs nothing but time. Take your 20 to 30 mandate questions and ask them, one after another, in ChatGPT, Gemini, and Perplexity. For each question, note three things: are you named directly? is a competitor named instead? and which source does the AI cite — your website, a comparison portal, an industry directory? Doing this by hand once a month shows you, in black and white, where you actually stand by line and region.

It matters that you test locality too. Ask once neutrally and once with your city added: "...broker in Regensburg" or "...near me, postal code 93047". AI answers are often locally sensitive, especially for brokers. You might surface for "occupational disability broker Regensburg" but not for the generic occupational disability question. You need to know that pattern to see where you're already strong and where the gap actually is.

If you want something more systematic, there are GEO tools that run these prompts automatically and repeatedly, giving you a mention rate over time. To start, though, a simple spreadsheet is enough: question, model, mentioned yes/no, cited source, date. After two or three months you'll start seeing trends — and whether what you're doing is actually working.

From mention to mandate: closing the attribution gap

The hardest question is whether an AI mention actually leads to a mandate. The tricky part: ChatGPT and similar tools rarely leave a clickable link with clean tracking attached. The prospect reads your name, then types it into Google separately or just calls you directly. In your analytics that shows up as "direct traffic" or "brand search" — the real source, AI, stays invisible. This attribution gap is why many brokers underrate how much their AI visibility is actually worth.

You can close that gap with a simple habit: ask every first contact how they found you, and actively offer "via ChatGPT / an AI / a chatbot" as one of the answer options. Log it in your CRM or contact list. After 30 to 50 inquiries you'll have a reliable read on what share of your new clients comes through AI recommendations — and which line of business dominates.

Now combine that customer feedback with your mention table. If you show up strongly for the commercial occupational disability question, and at the same time a noticeably high share of commercial occupational disability inquiries say "found through an AI", you have your proof. Linking the measurement on top (mentions) with the measurement underneath (mandates) is the real work here — and the point where GEO stops being a gut feeling and becomes something you can actually manage.

Which insurance lines are worth pursuing first for AI visibility

Not every line is equally worth the effort. Prioritize by three criteria: how advice-intensive it is, how good the commission or portfolio compensation is, and how often people actually ask an AI for an advisor on it. Occupational disability, private health insurance, commercial and business liability, and company pension schemes almost always come out on top. They need explaining, they stay on the books for years, and people are reluctant to decide on them alone.

Motor and household contents insurance, on the other hand, tend to be price- and portal-driven. Here the AI recommends a comparison calculator rather than a personal broker, and the margin rarely justifies the effort. That doesn't mean you drop these lines entirely — it means you spend your limited time on content and optimization where a single won mandate pays off for years.

A realistic example: a broker who focuses on medical professions deliberately targets questions like "As a junior doctor, should I switch to private health insurance or stay statutory?" and "What occupational disability cover do I need as a female dentist?" The niche is narrow enough that he regularly surfaces in AI answers as a specialist, and valuable enough that every mandate is worth it. Depth beats breadth here, almost every time.

What to actually optimize so the AI mentions you

AI models mention you when three signals line up: clear specialization, verifiable substance, and external confirmation. For specialization, state plainly on your website who you serve and what you stand for — not "all insurance for everyone", but "independent broker for occupational disability and private health insurance for doctors and the self-employed in eastern Bavaria". That clarity is what an AI can actually reproduce as a recommendation.

You build substance with specialist content that answers real mandate questions. A well-written guide titled "Occupational disability cover for doctors: the five most common mistakes when signing up" gives the model something quotable and positions you as a source. Focus on clear structure, concrete figures, and an honest assessment — AI systems tend to prefer content that answers a question completely and neutrally over pure sales copy.

External confirmation comes from Google reviews, professional directories, press mentions, and listings with broker pools or associations. When several independent sources connect you to the same specialization, the odds go up noticeably that an AI mentions you for the matching question. Make sure those sources describe the same positioning as your own site — consistently.

A simple monthly measurement routine for your firm

Turn measurement into a routine, or it fizzles out. Block off 60 minutes once a month. In the first half hour, ask your fixed list of mandate questions in ChatGPT, Gemini, and Perplexity, and log the mentions in your table. In the second half hour, check your CRM for how many new first contacts in the previous month said "via an AI", and which line they were asking about.

Pick two or three metrics you'll actually track: your mention rate across all mandate questions, your mention rate in your two most important lines, and the share of first contacts attributed to AI. You don't need more than that to start. What matters more than any single number is repetition — only the trend over several months tells you whether a new guide article or a batch of new reviews actually moved your visibility.

If a metric doesn't move over two months, change one thing deliberately — sharpen the specialization on your homepage, add another specialist article on the affected question, actively collect more reviews. That's how AI visibility stops being a gamble and becomes a channel you can manage as seriously as referrals or your portfolio work.

Honest limits: what measuring won't promise you

Be realistic: AI answers fluctuate. The same question can surface your name today and a competitor's tomorrow, depending on the model version, the wording, and plain chance. Your measurement is a sample, not an exact ranking. Draw conclusions from patterns across several weeks, not from a single answer that thrills or annoys you.

Equally honest: AI visibility replaces neither good advice nor your referral network. It's an additional channel that often brings especially pre-qualified prospects, but it rarely delivers volume fast. Someone who starts today, measures consistently, and works patiently on specialization and substance builds an advantage over six to twelve months that late-starting competitors will find hard to close.

And finally: the rules that govern insurance sales still apply. Your content has to be accurate and balanced, your reviews genuine, your advice documented. GEO isn't a trick for getting around the rules — it's the digital translation of what already sets you apart: clear specialization, real substance, and a reputation good enough that a machine can now read it too.

Mon–FriTue–Satdaily?

Questions we hear often

I'm a generalist broker without a clear niche. Is AI visibility even worth pursuing?

Yes — but you'll first need to pick one or two focus lines where you want to be visible, say occupational disability cover for the self-employed or commercial insurance for trades businesses. AI models tend to recommend recognizable specialists. You can stay a generalist in how you present yourself overall, while still building your AI visibility deliberately around your most profitable, advice-intensive lines. Depth beats breadth in AI recommendations almost every time.

How do I actually know whether a new client came to me through ChatGPT?

Don't rely on your website analytics — AI-driven traffic usually just shows up as direct traffic or brand search. Instead, ask every first contact how they found you, and explicitly mention the AI or chatbot option when you ask. Log the answer in your CRM. After 30 to 50 inquiries, you'll have a reliable read on what share of your mandates comes through AI recommendations.

Which insurance line should I optimize for AI visibility first?

Start with a line that has high advisory intensity, solid portfolio value, and genuine demand for an advisor — typically occupational disability, private health insurance, commercial or business liability, or company pension schemes. Motor and household contents insurance are portal- and price-driven; there, the AI recommends a comparison calculator instead of you. Spend your limited time first where a single won mandate pays off for years.

Share

 

More on Insurance Broker →All articles on Measurement & Reporting →