Skip to content
AI for Insurance Agencies insurance agencies cross-selling account rounding AI for insurance

The Insurance Agencies Problem Nobody Talks About: You Have 1,000 Clients and Only Sell Them One Policy Each

Mike Giannulis | | 12 min read
Share:
The Insurance Agencies Problem Nobody Talks About: You Have 1,000 Clients and Only Sell Them One Policy Each

Here is a number worth sitting with: if you have 1,000 clients and they average 1.8 policies each, you are leaving somewhere between 900 and 1,300 additional policies on the table compared to what leading agencies carry in the same sized book. That is not a production problem. That is a system problem.

Most agency principals know cross-selling matters. Very few have built anything that makes it happen consistently. This post covers what the data shows, what agents in the field are actually doing, and how to build a process that produces results without pressuring your clients or burning out your team.

The Gap Is Bigger Than Most Principals Realize

A 2026 benchmark compilation reports that the average independent agency sits at 1.8 policies per client, while top performers operate between 2.7 and 3.1. The same source attributes a minimum target of 2.3 policies per client to IIABA guidance, though that figure warrants independent verification before you use it in a formal business plan.

What those numbers mean in practical terms: if your agency has 1,000 client households and you are at 1.8 policies per household, reaching 2.3 would mean 500 additional policies. At a modest average commission of $150 annually per policy, that is $75,000 in recurring revenue from your existing book, with no new clients required.

The retention data makes this even more compelling. Single-policy clients reportedly retain at 77.1 percent, while clients holding five or more policies retain at 84.7 percent. That is a 7.6-point retention gap. Agencies that treat cross-selling only as a revenue tactic are missing the more durable argument: multi-policy clients stay longer, complain less, and refer more.

These figures come from a secondary benchmark compilation rather than a directly published IIABA or Liberty Mutual primary study, so treat them as directional. The pattern they describe, however, matches what agents report from actual practice.

What Industry Professionals Are Actually Saying

Agent forums tell a consistent story. The agencies winning on policies per client are not doing anything exotic. They are doing ordinary things with unusual discipline.

One Reddit agent described a straightforward conversation framework: “I noticed that you have your cars and home with us here at SF. But I don’t see any life insurance on file here. Do you have that with another company?” That question, asked consistently, is the entire strategy. The same thread cited a practical funnel worth tracking: 25 offers lead to 5 to 7 quotes, which produce 1 to 2 issued policies. Five conversations per day across your team could generate 4 to 8 additional policies per month.

The IIAT personal lines best practices guide puts it plainly: write clients, not policies. The best-performing agencies report more than two policies per client and credit that to making account rounding a non-optional expectation with training, monitoring, and rewards attached.

Agents on Reddit consistently identify the same failure points when cross-selling is not working. Producers fail to ask. Cross-selling is treated as a personal habit rather than a workflow. Staff present too many products at once. There is no clean list of mono-line households to work from. Follow-up evaporates after the first conversation.

None of those are talent problems. They are system problems. And system problems have system solutions.

By the Numbers: Where Your Book Probably Stands

MetricAverage AgencyTop-Performing Agency
Policies per client1.82.7 to 3.1
Single-policy client retention77.1%N/A
Five-plus policy client retentionN/A84.7%
Manual cross-sell conversion4% to 7%N/A
Automated outreach conversionN/A14% to 22%
Agents using AI individually65%N/A
Agencies with AI formally deployed6% (2024) to 46% (2026)N/A

Sources: Secondary benchmark compilation and Liberty Mutual Agency Universe Study data. Conversion figures are directional; primary-source methodology has not been independently confirmed.

The AI adoption numbers deserve a closer look. Liberty Mutual’s research found that 65 percent of agents had used AI for work in the prior year, with average reported time savings of four hours. But only 6 percent of principals had formally implemented an AI solution at the agency level as of 2024. A later 2026 Agency Universe Study reportedly shows agency-level AI adoption at 46 percent, up from 15 percent in 2024, suggesting the gap is closing fast.

Liberty Mutual also specifically identified identifying cross-sell opportunities as one of the leading intended AI applications among agency principals. The interest is there. The deployment has lagged.

Strategy 1: Build a Mono-Line Report You Actually Use

The first problem with a 1.3 to 1.8 policies-per-client average is not agent behavior. It is visibility. Agents cannot cross-sell gaps they cannot see.

The fix starts with a weekly report that segments your book by coverage status. Auto-only households. Home-only households. Renters-only clients. Commercial accounts without workers compensation. Property clients without an umbrella. Life clients without a review in 24-plus months.

This is not complicated to generate. Any agency management system worth using can produce this list. The problem is that most agencies pull it once, feel briefly motivated, and then set it aside because there is no process attached to it.

The process needs to be this specific: every Monday, the report runs. Every producer gets assigned a segment. Every assigned account gets a documented outreach attempt within five business days. The metric tracked is not how many policies were issued that week. It is how many coverage-gap conversations were completed.

The Hanover agent resource on top independent agency practices reinforces this structure. Agencies that consistently outperform on policies per client have made account rounding a monitored expectation, not a voluntary activity.

For a 1,000-client book, your first mono-line report will probably surface 400 to 600 households with an obvious gap. That is your pipeline. You did not buy it. You did not market for it. It was already there.

Strategy 2: Make Account Rounding a Workflow, Not a Personality Trait

The second problem is that cross-selling in most agencies depends entirely on which producer is handling the interaction. Some agents ask. Most do not. The difference between a 1.8 and a 2.7 policies-per-client ratio often comes down to whether asking is built into the process or left to individual initiative.

Building it into the process means a few specific things.

First, every service interaction needs a scripted one-question add. Payment calls, renewal reviews, claims follow-ups, and policy-change requests are all natural opportunities. The agent who takes the call already has the client’s file open. The question costs 20 seconds. The Reddit funnel suggests that if five agents ask five clients per day, the agency produces 4 to 8 additional policies monthly from that habit alone.

Second, the two-step workflow consistently outperforms a single combined call. Step one is a discovery conversation: identify the gap, acknowledge the life event, and confirm the client is open to a review. Step two is a scheduled proposal appointment. This approach reduces the pressure the client feels and gives the agent time to prepare a recommendation that actually fits the household’s situation.

Third, agencies that bundle effectively report up to 80 percent of business flowing through package placements. But the same agents caution that bundling should not override suitability. If a separate carrier genuinely provides better coverage or pricing, forcing a bundle for the sake of a metric damages trust. The goal is to make package placement the default offer while preserving the judgment to recommend otherwise when it genuinely serves the client.

You can review how RunFrame structures this kind of workflow at /how-it-works/ and see how it applies specifically to insurance agencies at /industries/insurance-agencies/.

Strategy 3: Connect Outreach to Life Events, Not the Calendar

The reason cross-selling feels pushy is usually that it is disconnected from anything happening in the client’s life. A call timed to your production quota is a sales call. A call timed to a client’s teenager getting a learner’s permit is a service call.

Life-stage triggers that agents cite as the most effective cross-sell entry points include a teenager beginning driver education, a new home purchase, marriage, divorce, the birth of a child, a job change with income increase, a business acquisition, a major asset purchase, and a renewal with a significant rate increase.

Each of these events creates a natural reason to revisit coverage. A new driver changes liability exposure. A higher income changes life insurance needs. A business acquisition opens the door to commercial coverage conversations. These are not pretexts. They are genuine coverage gaps that the client probably has not thought about.

The operational challenge is that most agencies have no system for detecting these triggers. They learn about them accidentally during service calls, or they never learn about them at all. A client buys a vacation home through a different agent because nobody from your agency asked.

Automating trigger detection is where technology actually earns its place in this process. This is exactly what RunFrame is built to do. The system analyzes your book of business, identifies coverage gaps, flags life events based on policy data and renewal timing, and delivers a prioritized action list to your producers every Monday. Agents do not have to remember to ask. The ask is already scheduled and contextualized.

If you want to see where your agency currently stands on AI readiness for this kind of deployment, the AI Readiness Scorecard at /scorecard/ takes about five minutes and gives you a specific starting point.

Implementation Roadmap: What to Build First

If you are starting from scratch, here is a sequence that works without requiring a technology overhaul on day one.

Week 1 to 2: Pull the data you already have.

Generate a mono-line report from your AMS. Segment by product type. Flag households with auto but no home, home but no auto, property without umbrella, and any client without a documented life insurance conversation in the past two years. This is your initial cross-sell pipeline.

Week 3 to 4: Set activity targets, not production targets.

Assign each producer a weekly number of coverage-gap conversations to complete, not policies to issue. The Reddit funnel benchmark of five conversations per producer per day is a reasonable ceiling. Start with two or three and build. Track conversations in your CRM or a simple shared log.

Month 2: Add the two-step workflow.

Train producers on separating discovery from proposal. The first call identifies the gap. The second call, scheduled before the first call ends, is the coverage review. This reduces client resistance and improves close rates.

Month 3: Layer in trigger-based outreach.

Begin connecting outreach to life events. Use renewal data, claims records, and policy change history to identify clients who have experienced triggering events recently. Build a simple calendar of proactive calls timed to those events.

Month 4 and beyond: Automate what is repeatable.

Once the manual process is working, identify which steps can be systematized. Trigger detection, outreach scheduling, and follow-up reminders are good starting candidates. Agencies that reach this stage are typically the ones moving from 1.8 toward 2.5-plus policies per client.

For agencies that want to skip the manual phase and deploy automation from the start, RunFrame’s AI operating system covers the full stack. For agencies that want ongoing management of the process rather than a one-time build, fractional AI ops provides that layer.

How RunFrame Approaches This

RunFrame’s approach to insurance agency cross-selling is not a chatbot that answers policy questions. It is a book-of-business analysis engine that identifies which of your existing clients have coverage gaps, maps those gaps to life events and renewal timing, and gives your producers a ranked action list each week.

The output is specific. Not “here are your mono-line clients.” Instead: “These 12 households added a driver in the last 90 days and do not have an umbrella policy. These 8 clients renewed at a rate increase above 15 percent and have never been quoted for a competing home carrier. These 5 clients have a life policy with no beneficiary update in four or more years.”

Producers get context, not just names. The Monday list replaces the manual sorting that most agencies either do inconsistently or skip entirely. And because the recommendations connect to real events in the client’s situation, the conversations that follow feel like service, not sales.

If you want to understand exactly how this would work in your agency, booking a discovery call at /book/ is the fastest path to a specific answer.

The Bottom Line

The gap between 1.8 and 2.7 policies per client is not a gap in effort. Most agency principals and their producers are already working hard. The gap is in systematization. Leading agencies have made cross-selling visible through regular reporting, repeatable through documented workflows, timely through life-event triggers, and accountable through tracked metrics.

You have the clients. You have the carrier relationships. You probably have more cross-sell opportunity in your existing book right now than you will generate from new business this year. The question is whether you have a process to find it and a workflow to act on it before those clients buy that second or third policy somewhere else.

Start with your AI Readiness Scorecard to see where your agency stands, or book a call to walk through what a cross-sell system would look like for your specific book.

Ready to Deploy AI? Book a Free Assessment

30 minutes. No pitch. No pressure. Just a conversation about what is possible for your company.

Book Your Free Call
Mike Giannulis

Mike Giannulis

Founder of RunFrame and Anthropic Partner Program member. 20+ years in direct response marketing. Building AI operating systems for companies with 5 to 50 employees.

Ready to See What AI Can Do for Your Company?

30 minutes. No pitch. No pressure. Just a conversation about what is possible.

Book Your Free Assessment