How Do You Prove a No-Show System Earned the Money (and It Wasn’t Just Luck)? “We recovered $50,000 for your practice this year!” You’ve seen that line on a vendor dashboard, and maybe you even believed it for a minute. Then you start asking questions. How many of those patients would have called back anyway? How many slots got filled because your receptionist picked up the phone and rang someone? This is where most recovery claims fall apart. Some patients rebook on their own. Some slots get filled by a walk-in, or by a front desk that’s genuinely good at its job. If a system counts all of that as its own work, a sharp practice owner notices eventually, and once they notice, they stop trusting every other number on the screen. We’d rather show a smaller number that’s true, and part eight of our Clinic Scheduling & Recovery series is about how. Why do most recovery numbers overcount? Because they measure everything that happened after the system switched on, and give the system credit for all of it. But clinics were rebooking cancelled patients long before any software showed up. A good receptionist with a phone and a waitlist on a sticky note recovers plenty. The only fair question is how much more gets recovered with the system than without it, and that “how much more” is the only part we should take credit for. What are we trying to find out? We want to know how much of the revenue recovered after switching on automated backfill and follow-ups can fairly be credited to the system, rather than to patients and staff who would have acted anyway. To do that, we’ll first set a baseline that shows what happens to cancellations when the system does nothing at all. We’ll then measure the lift from backfill offers and from follow-up messages separately, so each feature has to earn its own keep. From all of that, we’ll produce one honest “recovered by the system” figure each month, and design the revenue dashboard so it shows not just that number, but how we got it. How does a control group work here? It starts with a baseline. For four to eight weeks, the system logs everything but does nothing, which shows us what “normal” looks like at that particular clinic. Once the features go live, we hold back a random 10 to 20% of cancellations as a control group. Those patients don’t get any automated follow-up, and the staff simply do whatever they’d normally do. We then compare rebooking and refill rates between the group that got automated help and the group that didn’t. On top of that, every rebooking is tagged with its source, whether that’s our link, a phone call, a walk-in or a staff action, so that even within the treated group, we can see what actually did the work. What’s the attribution formula? The credit the system gets is the treated rebooking rate minus the control rebooking rate, multiplied by the number of treated cancellations, multiplied by the average slot value. So if 40% of treated cancellations rebook and 28% of the control group rebook anyway, the system gets credit for the 12-point difference, not the full 40%. The average slot value comes from the procedure-level work in part seven. It’s a less exciting number, but it’s one that survives an accountant. What data does this need? Every cancellation in the demo carries a flag showing whether it’s in the treated group or the control group, which is the heart of the whole comparison. Every rebooking or refill records its source, so we know who actually did the work. Each slot has a value attached, which is what turns rates into dollars. And the demo keeps the data from the baseline period, so we always know what “normal” looked like before the system switched on. Isn’t holding back follow-ups a bit harsh? This is the hard conversation, and we’d rather have it up front. Holding back automated follow-ups from a slice of cancellations means leaving some revenue on the table on purpose. That’s a real cost to the clinic, and it deserves to be said plainly. But it’s also the only way to prove the rest of the number is real. And those patients aren’t abandoned. The front desk still calls them the way it always has; they just don’t get the automated message. We’d explain it to a clinic like this: “This small holdback is what lets us prove the rest of the number is real. Without it, we’re just guessing, and so are you.” After a few months, once the lift is well established, the holdback can shrink. Previously, we looked at how much a cancelled appointment actually costs. Next, we ask whether deposits and cancellation fees help, or just drive patients away.