# Ryan Cassidy, Founder at Cassidy Voice Works — read of Trucking Fleet Insurance Renewal Voice Campaign, June 15 2026

> 9 years in sales tech, 4 at a voice platform vendor, now running a 5-person shop that builds outbound voice automations for boring verticals that actually pay. Looking for the next niche.

## How I got here

Someone in a Slack I'm in (the Retell AI founders channel, about 200 people) dropped this link with zero context. I assumed it was a competitor's product demo page. Clicked it expecting to reverse-engineer their funnel. Took me about 45 seconds to realize this is a business-idea marketplace, not a live product. That realization colored everything after.

## What I clicked first

The hero pulled me in. "Outbound voice calls that book policy review meetings 60-90 days before renewal" is a clean, specific use case I can immediately picture. Trucking renewal is genuinely painful — I know because a client of mine tried to build something adjacent last year and the data problem ate them alive. So the concept passed the sniff test fast. Then I scrolled and hit the score block.

## Where I paused

The Fermi numbers. "$-11,360 Year-1 take-home." That's negative. I read it three times. Then I read the "1 in 8 meaningful-success odds." These numbers are presented like transparency badges, which I actually respect, but I'm sitting here trying to work out: is the -$11K loss because of build costs? Because CAC is brutal in this space? Because the market is smaller than it looks? The page never explains what's in the Fermi model. It just hands me the output and says "here's the math." I want to see the inputs. That's the whole point of a Fermi estimate.

## What I distrusted

A few things stacked up.

"Integrates with FMCSA safety score data." This is a very specific claim and it reads as credibility-building detail, the kind of thing you put in to make the feature list feel real. But the disclaimer right below says "we don't have live customers on this idea yet." So: is this integration built and working, or is it on the roadmap, or is it "possible to build and here's how"? Those are three very different things and the page does not distinguish them.

"Broker data feeds." Which ones? Applied Epic? AMS360? A CSV upload? This matters a lot in practice. Real agency management systems have terrible API surfaces and the brokers who run trucking books are not handing over their renewal data to a new vendor without a serious conversation.

"Calendars auto-sync." With what?

These are the kinds of claims that sound like features in a pitch deck before anyone has written a line of integration code.

## What would convince me

Show me the call recording. Not a scripted demo, an actual call where the voice agent navigates a trucking broker going "I don't know what FMCSA score has to do with my liability limits." That conversation is where this either holds up or falls apart. If they have a 3-minute real call with a real carrier and a real handoff to a calendar booking, I'm in. I'd also want to see what "starter code" at the $99-$199 tier actually means: a Twilio Studio flow, a Retell AI agent config, a Vapi prompt library? Those are not the same level of headstart.

## What I'd ask in an email reply

1. The Fermi shows a negative year-1 take-home. What's the biggest cost driver in that model, and does it assume I'm building this myself or hiring someone to build it?

2. When you say "broker data feeds" in the Intelligent Timing feature, do you have a working integration with any AMS today, or is that describing a build task I'd own after buying the dossier?

3. Has anyone taken the $99 adopt tier to a paying client? Even one? I'm not looking for a case study, I just want to know if the starter package has been stress-tested outside the team that wrote it.

## Verdict: on-the-fence

The vertical is real, the use case is specific enough to be credible, and the pricing is honest about what you're getting. But I came here expecting a product and found a blueprint, and I'm not sure the blueprint addresses the two hardest parts: data access and the E&O conversation that happens when a voice agent mischaracterizes coverage gaps on a recorded call.

---
*Memo by skeptic persona, generated 2026-06-15. Studio breaks own self-grading loop.*