Retention across the books I've bought since 2014 runs around 80 to 90 percent at twelve months. That's the number diligence measures and the number most earnouts pay against. Then another 10 to 15 percent walks between month eighteen and month twenty four. By then the earnout's paid out, the seller's gone, and nobody's really watching. If anyone notices, it gets written off as normal churn. I don't think it is. The people who leave in that window are the ones who never had a reason to go and never had a reason to stay. The service didn't get worse. But the guy they actually hired isn't around anymore, or the price went up once eight months back, or the local number rings somewhere else now. None of that makes anybody quit. It just means when a competitor calls, they take the call. So if you're pricing bolt ons off trailing revenue and measuring to twelve months, you're buying a number that hasn't finished moving. Ten of my eleven deals came from letters I sent owners myself. Service businesses, recurring revenue, all rolled into a company I already owned. No brokers, no bankers, no listing sites. A few closed off letters I'd sent two or three years earlier, to people who didn't remember getting them. If you already own a platform and you're adding bolt ons, brokered deal flow is the expensive way to do it. You're bidding against everybody working off the same list. I can help you build the other channel. Target list, the letters, the follow up, and the first conversation with an owner who's never really thought about selling. I've had a lot of those. If any of that's useful, message me.