Using AI to Prep for Insurance Appointments

In the last video you learned how to run an appointment from start to finish. This video is about what happens before you walk in the door. Specifically, how to use AI to do in 5 minutes what used to take 30. We are talking about researching a prospect's area, understanding local plan options, and building talking points so you walk into every appointment already sounding like an expert.

Most new agents skip appointment prep entirely. They show up, open the quoting tool, and figure it out in real time. That works until the client asks a question you were not expecting. Why are there so few PPO options in my county? Which hospitals are in this network? Is there a plan that covers my specific medication at a lower tier? When you cannot answer those questions confidently, the client hesitates. And hesitation kills sales. Fifteen minutes of AI-assisted prep eliminates most of those surprises.

Here is how to do it. Before your appointment, open any AI chat tool. You do not need anything fancy. A free tool works fine. You are going to give it three prompts that build on each other.

Prompt one is the area research. You type something like — I am a Medicare insurance agent meeting with a client in [ZIP code]. Give me a summary of the Medicare market in this area. How many Medicare Advantage plans are available? Are there more HMO or PPO options? What are the major hospital systems and provider networks in this region? The AI will give you a snapshot of what the local market looks like. This is not plan-specific advice you are giving the client. This is background knowledge that helps you sound informed when the client mentions their doctor or their hospital.

Prompt two is the common concerns prompt. You type something like — what are the most common Medicare questions and concerns for people in [state]? What local factors affect Medicare coverage in this area, like rural access or provider availability? This gives you a feel for what your specific client might be dealing with before they even tell you. If you are meeting someone in a rural county, access to specialists is probably a concern. If you are in a metro area, network size and plan variety might be the bigger topic. Knowing this ahead of time means you are not caught off guard.

Prompt three is the talking points prompt. You type something like — based on what you just told me about this area, give me three talking points I can use to start a natural conversation with a Medicare-eligible client. Keep them conversational, not salesy. The AI will draft a few openers you can adapt. You are not reading these word for word. You are absorbing the ideas so that when you sit down with the client, you have something relevant to say beyond — so, tell me about your Medicare situation.

One important note. AI is a prep tool, not a quoting tool. Do not use AI to recommend specific plans or give coverage details to clients. That is what your licensed quoting and enrollment tools are for. AI helps you walk in informed. Your actual tools and your training are what help you give accurate, compliant recommendations.

Here is what this looks like in practice. Say you have a 10 AM appointment in a ZIP code you have never worked before. At 9:30 you open your AI tool, run the three prompts, and spend 5 minutes reading the results. You learn that the county has 12 Medicare Advantage plans, mostly HMOs, with one regional PPO. The dominant hospital system is a large network that most plans include. Rural clients 20 minutes outside town have fewer in-network options. Now when your client says they see a specialist at a smaller clinic outside the main hospital system, you already understand why that matters for plan selection. You did not memorize a plan. You built context that makes the rest of the conversation smarter.

Your action step. Before your next appointment, run all three prompts using the client's ZIP code. Read through the results and pick one thing you learned that you can bring into the conversation naturally. It does not have to be a big moment. Even something as simple as — I know there are a lot of plan options in your area, so let me help you narrow it down — shows the client that you did your homework.

Any general-purpose AI chat tool works for this. ChatGPT, Claude, Gemini, Perplexity — all of them handle this kind of market-level research fine on their free tiers. Pick the one you're most comfortable with and stick with it so the prompting becomes second nature. What matters is the quality of your prompts, not which platform you're on. Avoid Medicare-specific AI tools that promise plan recommendations — those cross into regulated territory and most aren't vetted for compliance.

Yes — as long as you're using it for background research, not client-facing advice. Researching the local Medicare market, understanding common concerns in a region, or drafting conversational openers are all educational activities. What's not compliant is copying AI output into a recommendation you give the client, feeding client PHI into an AI tool, or using AI to make coverage decisions. Keep AI on the prep side of the line. Your licensed quoting tool, your certifications, and your carrier training are what power the actual recommendation.

Good for context, not for specifics. AI is strong at telling you the general shape of a market — is this area HMO-heavy, who are the dominant hospital systems, how does Medicare work in rural vs. urban parts of this state. It's unreliable for specific plan counts, plan names, or current pricing because that data changes every year and AI training data lags. Use AI for orientation. Use your quoting tool for facts about specific plans.

Five to ten minutes if you're efficient. The first time you run this workflow it'll feel slow — you're figuring out how to phrase the prompts and reading carefully. By your fifth or sixth appointment, you'll have your three prompts saved somewhere, you'll paste them in with the ZIP code filled in, and you'll skim the results while your coffee is brewing. Aim for a 15-minute window before any appointment in an unfamiliar area. For ZIP codes you work repeatedly, you can skip prompts one and two and just run the talking points prompt.

No. Never paste the client's name, date of birth, address, Medicare number, medication list, or any other personally identifiable or protected health information into any AI tool — even if the tool claims to be HIPAA-compliant. Use ZIP codes, states, and general demographics only. Your prompts should look like "Medicare market in 30309" not "Medicare market where Jane Smith lives at 123 Main Street." Keep the AI prep tool completely separate from your quoting, enrollment, and CRM workflows where client information belongs.