At multi-location groups we have worked with, the first hint of a drifting location comes from a site visit, not a report. At one group, the owner wasn’t there about money.
Office managers feel an insurer slowing down weeks before any group report shows it. Nothing the owner read each month said anything about one insurer at one location.
What the group report showed
Her consolidated report was calm. Days-to-pay for the group had held steady all year, moving about one day. And the number wasn’t wrong. Every location’s data was in there, added up honestly.
It was answering the question it was built for, how is the group doing, and that question has no bad quarters in it. Blend the locations and the strong ones absorb the weak one automatically. That’s what blending is. Medical practice reporting built on one blended number cannot show a bad location.
What the location’s own history showed
Then we asked a different question. We took that one location and put that one insurer against its own past twelve months. The insurer was running weeks off its own pace, and had been for a full quarter. Real money, arriving late, at one address, for three months.
At group altitude it rounded away. At location altitude it was the difference between a comfortable quarter and a tight one, and the office manager had felt every week of it. She knew. The report didn’t.
Sit with that for a second, because it’s the real finding. Every location has an office manager who knows something the consolidated report doesn’t, and none of them had anywhere to send it. The group’s early-warning system was whichever location the owner happened to visit that month, for whatever reason brought her there. A site visit had just outperformed the entire reporting stack.
Why do consolidated reports miss location problems?
Because blending is the job: every location averages into one number, and one location’s bad quarter rounds away. The wider the group, the duller the blend, and the location-level answer never surfaces.
Why averages do this
Nobody built the report wrong. Averaging is what averages do, and the effect gets stronger as you grow. Every location you add widens the range the blended number calls normal, so the threshold for what looks worth asking about rises with each acquisition or opening.
A two-location group notices a bad quarter at one address. A large group notices a bad year, maybe. The instrument gets duller exactly as the thing it watches gets bigger, which is the opposite of what an owner assumes scale buys.
A different question needs a different instrument
The group question and the location question can’t share a report. The location question is: for each insurer at each location, how does this month compare to that insurer’s own last twelve months at that address? Three ways it can break.
Slower, the days drift past their own range. Lighter, the payments arrive on schedule with fewer cents on the dollar in them. Quieter, less money arrives than that insurer’s history says should, even when every payment that does arrive looks fine. Any of the three flags the week it starts, with a location, an insurer, a date, and a dollar figure attached.
At group scale the output stays small on purpose. One weekly list. A normal week reads empty, or one line. When a line appears, it routes to that location’s manager the same day, and the group watches one thing about it afterward: how long it stays open. The owner never went looking for problems again. The problems learned to introduce themselves, properly dressed, with their paperwork attached.
What turned up, what we built, what holds
Found: one insurer at one location, weeks off its own pace for a quarter, sitting invisibly inside a consolidated number that barely moved.
Built: the location-by-insurer panel. Every insurer at every address against its own history, refreshed weekly, with breaks surfacing as a short list instead of a feeling somebody has to mention during a site visit.
Holds: the accident became a system. The office manager who knew now has somewhere to send it, and the owner reads a list that’s usually empty and occasionally the most important thing on the desk. The next drift got flagged in its second week, at a different location, from an insurer nobody was worried about.
Your version
If you run more than one location, try this with your biggest insurer. Chart its days-to-pay at each location separately, twelve months back. If the lines move together, your blend is honest.
If one line has wandered off on its own, your consolidated report has been averaging away exactly the thing you needed to see, and the people at that location already know which line it is. While you’re in there, check whether your location ranking is grading geography and whether your newest location is coasting on a grace period.
We’ll run every insurer at every location against its own twelve-month history and hand you the list your average has been eating. Grab 30 minutes with us. Prep nothing.