# Randy Kowalski, Owner at Kowalski Insurance Group — read of Insurance Lapse Recovery Voice Agent, June 7 2026

> 14 years writing P&C and life in the Milwaukee suburbs, three staff, one office, and a renewal spreadsheet I've been promising to replace since 2021.

## How I got here

Googled "automate insurance renewal calls" on my lunch break because I had three policies cancel last week and I knew all three of them were preventable. Clicked through four or five results. Most were generic call center software or carrier-specific tools that don't work with independent agents. This one came up and the headline was specific enough that I kept reading.

## What I clicked first

"Stop cancellations before they happen" is fine, I've seen that. What actually made me stay was the specific number: "Most independent agents lose 15-30% of renewals to simple neglect." That's not "many agents struggle with churn." That's a number that made me go grab my spreadsheet and check. Mine's closer to 22 right now, which stings.

The scenario in the problem section: "You call at day 45 and get voicemail. Prospect never calls back. Day 60 arrives and the policy cancels. You just lost that premium forever." Yeah. That's Tuesdays for me.

## Where I paused

The sample call script line: "What's changed since we last talked?" That's a real question. That's what I'd ask. A good agent in 2012 would open with that. So when I read that, I thought, okay, whoever wrote this copy has been in the room with an insurance agent before. That earned them another minute of my time.

## What I distrusted

Okay, so I'm reading along, thinking this might be worth a trial, and then I hit this at the bottom:

> "Honest disclosure: we don't have live customers on this idea yet. We shipped the strategy package; you ship the customer conversations."

And then: "Unlock the dossier $5" and "Adopt the build $99."

Wait. So the pricing table with "$2.50 per call" and the "Start 30-day free trial" button -- that's for a product that doesn't exist. This isn't a product page. It's a pitch deck for someone to go build the product. The "financial upside: 3/10" and "$-35,320 Year-1 take-home" metric are for a *builder*, not a buyer.

I'm an independent agent. I don't want to build a voice AI startup. I came here to solve my Tuesday problem. The page presents itself as a product you can sign up for, and then tells you at the bottom it's a business idea you can purchase.

That is genuinely confusing, and I'd call it dishonest even if they didn't mean it to be.

Also: "No robocall feel. Real conversation." Every voice AI company says this. I'd need to hear an actual sample call before I believed it, especially with a policyholcer who's 60 days out and already skeptical.

## What would convince me

One thing: a real call recording. Not a polished demo. A messy one where the prospect is a little rude, says "I already switched" partway through, and the agent recovers or gracefully ends the call. That's what I deal with. Polish tells me nothing.

Second: a real independent agent, by name, with their agency on record, saying what their 90-day recovery numbers looked like before and after. Not a testimonial quote. A named person I can look up on the state DOI website. That would do it.

## What I'd ask in an email reply

1. When you say "Start 30-day free trial" on the pricing page, is there an actual working product I can test, or is this still being built? I need a straight answer on that.

2. Which CRMs does the call sync actually work with right now? I'm on EZLynx. If it just emails me a PDF transcript I'll manage it myself, but I want to know upfront.

3. Has this been tested with any real calls to real policyholders, or is the "18-22% recovery rate" a modeled number? If it's modeled, what's the basis?

## Verdict: on-the-fence

The problem statement is the most accurate I've read in this space, and the per-call pricing model makes sense for a shop my size. But I genuinely cannot tell if I'm reading a product page or a business plan, and that confusion alone might make me bounce before a founder who was actually shipping would have lost me.

---
*Memo by skeptic persona, generated 2026-06-07. Studio breaks own self-grading loop.*