There is a growing adoption of artificial intelligence technologies among insurance agents. Research from Liberty Mutual Insurance found that the use of AI in independent agencies is growing, with more than one-third of today’s agency employees already exploring AI for work.
Generative AI – tools that create text, images, audio or video – are among the most commonly used tools. However, getting the most out of these tools requires experimenting with AI prompting – the art of asking your AI solution the right questions and giving the right instructions to achieve the results you want. Our previous article, “Crafting effective AI prompts: A quick-start guide,” laid out the key ingredients for generative AI prompts and provided tips for tweaking prompts to get effective outputs.
Generative AI is not just for writers and marketers. No matter your role in your agency, generative AI tools can help you communicate more clearly, reduce the time you spend on routine tasks and streamline your work.
Here are more focused recommendations for use cases and effective prompting techniques across sales, service, marketing and leadership functions.
Agents, producers and sales professionals
Personalize sales scripts and outreach. Use prompts that include your target audience (“Texas small business owners”) and desired tone (“friendly yet authoritative”). “Draft a voicemail script introducing our auto/home bundle options to first-time homeowners in Pittsburgh.”
Overcome objections with prepped responses. Ask AI to generate lists of common objections plus sample responses. Request reasoning or examples tied specifically to your market or product lines.
Summarize complex coverages. Prompt AI tools to explain nuanced policies at various levels—from basics for clients new to coverage through deeper dives for business-savvy prospects.
Account managers, customer service representatives (CSRs) and service team members
Create client communications. Leverage AI for policy renewals/proposals: “Draft three text message templates reminding personal lines clients about upcoming annual reviews.” “Create an email explaining cyber liability basics tailored toward local small law firms.”
Draft clear policy explanations and reminders. Specify policy type/context so outputs fit the client’s needs: “Write an email explaining recent renewal changes on personal auto policies in simple terms.”
Streamline customer follow-ups. Build reusable templates via prompts for things like claim updates or review requests—then refine as regulations/products change.
Create step-by-step guides/checklists. Ask the AI assistant to produce process outlines (e.g., certificate issuance), then request improvements based on real-life bottlenecks seen by your team.
Marketing team members
Generate hyper-relevant content quickly. Include target region/audience/specialties when prompting; e.g.: “Write a blog post about flood coverage concerns specific to Montgomery County.”
Optimize across channels with platform-aware prompts. Instruct AI which platform style/tone you want: LinkedIn vs. Facebook vs. email newsletter. “Summarize this article into five LinkedIn posts designed to create engagement.”
Brainstorm campaign ideas efficiently. Task generative models with producing campaign frameworks (“Suggest three cross-selling campaigns targeting existing renters”) then iterate until aligned with brand voice/goals.
Leaders and owners/principals
Enhance strategic decision-making. Have AI analyze internal reports (e.g., retention data/loss ratios). Ask it to flag trends/opportunities—and back up findings with clear explanations.
Improve standard operating procedures and training documentation. Prompt guides aimed at onboarding new employees or rolling out tech/process updates agency wide. Draft best-in-class standard operating procedures for quoting commercial insurance.
Foster agency-wide consistency. Standardize proven prompts/templates across departments, so communication remains sharp even as staffing evolves or markets shift.
No matter your specialty, treat every interaction with generative AI as a conversation, not one-and-done instructions. Give feedback, supply additional context and most importantly, always verify AI-generated content and outputs before sharing externally or making final decisions.
