There is a number on your aging report you trust. It says a certain amount is collectible. Some of it is not, and has not been for a while, and you have been making decisions on it the whole time.
Every practice inherits a set of categories and codes when its system is first set up, and almost nobody opens them again. They came preset, they look official, and the report adds up, so you trust the buckets. The “other” line. The adjustment codes. The write-off categories. They sit there doing their work quietly, and quietly is the problem, because a category you never question is a category that can drift away from the truth without anyone noticing.
What we found underneath the categories
At one practice, we found $104K in AR that showed as collectible but had already been written off. Silent reclassification codes had moved those balances into a status that still landed on the collectible line. The money looked like it was coming. It was not. It had been gone for months, and the report never said so.
We only caught it by going underneath the categories, into the individual transactions, all 94,714 of them, and following where each balance actually went versus where the report said it was. That is the only way to catch a number that is lying quietly. The summary looks right. The detail tells the truth.
A write-off that does not look like one
Here is how a balance disappears without anyone deciding to let it go. A code gets applied that reclassifies the balance into a status the report still counts as open. To the person who applied it, it was a routine adjustment, one of hundreds that week. To the report, the balance still reads as collectible. Nobody made a decision to write off $104K. It happened one quiet code at a time, each one defensible on its own, and the total only showed itself when someone added them all up against what was actually collected.
This is why it is so hard to catch from the top. No single entry looks wrong. Each one is a small, ordinary adjustment that any reviewer would wave through. The damage is only visible in the aggregate, and nothing in the standard reporting ever shows the aggregate, because the report is built to total the categories, not to question whether the categories still mean what they say.
Why this is the one that costs a finance leader sleep
The danger goes beyond the $104K itself. You have been making decisions on a collectible number that included money that was never coming. Cash forecasts, staffing, distributions, all built partly on AR that had already evaporated. The forecast was confident and wrong, and it was wrong in the most dangerous direction, too optimistic, because the phantom balances made your collectible position look stronger than it was.
And the gap does not announce itself. It surfaces the day you go to collect and find there is nothing there, or the day someone with sharper eyes than your last reviewer goes looking and asks why the collectible line does not match what is underneath it. If that someone is a lender during diligence, or a buyer’s quality-of-earnings team, the question arrives at the worst possible moment, and now it is not just a cleanup, it is a credibility problem, because they will wonder what else the categories are hiding.
A real write-off versus a phantom one
There is nothing wrong with a write-off. Every practice writes off balances it genuinely cannot collect, and doing so keeps the books honest. The difference here is who decided. A real write-off is a decision someone made on purpose, with eyes open, and recorded as what it is. A phantom write-off is the same outcome reached by accident, through a reclassification code that moved the balance out of collectible without anyone choosing to give it up, and without the report admitting it happened. The money is just as gone either way. What is missing in the phantom case is the decision and the record of it, which is exactly what you need when someone later asks why the collectible number was off.
How finance solved this a long time ago
Finance has a standard answer for exactly this: the suspense account. Anything that cannot be cleanly classified goes into a holding account that gets reconciled every single close. The rule is that an unexplained balance never gets to grow. You clear it or you explain it, every period, no exceptions. The discipline is not in being clever about the categories. It is in refusing to let any bucket run unexamined.
Healthcare lets the “other” bucket and the adjustment codes run for years without anyone reconciling what is actually inside them. The categories were set up once and trusted ever since. The fix is not complicated, it is the habit finance already has, applied to the buckets a practice never thought to question.
Why a one-time scrub does not hold
You could clean this up once. Pull the transactions, find the phantom balances, correct the collectible line, and the number is honest again for a quarter. Then it drifts back, because the same reclassification codes are still in use every day, and nothing changed about how they get applied. A one-time scrub fixes today’s total. It does not fix the process that produced it. The only thing that holds is reconciling the categories on a schedule, the way a close reconciles a suspense account, so a bucket cannot quietly drift away from what it claims to be without someone catching it that period.
Found, fixed, and held
Found: $104K sitting on the collectible line that had already been written off.
Fixed: the categories get reconciled against the underlying transactions, so collectible means collectible.
Held: the buckets get checked on a schedule, so a number cannot quietly drift away from what it claims to be.
What this means for you
You can spot-check this without us. Take your largest adjustment or “other” category and pull the underlying transactions for one month. Follow ten of them all the way through. If the summary line and the detail disagree, you have got a number reporting something other than the truth, and it is worth knowing which direction it is off and by how much.
Book 30 minutes. We will reconcile your collectible AR against what is actually underneath it, on your own data, and you will see whether the number you forecast on is the number you will collect. Your numbers, not ours.