We traced every appointment at a multi-provider practice across five months and built a table of where each one stopped. Twelve stages, 42,615 appointments.
One row came back zero. Cancellations.
Not a low number. Zero. Across five months and forty-two thousand appointments, the system had no record of a single patient cancelling.
What does it mean when a practice records no cancellations?
That cancellations are not being captured as a status. They are still happening, because they happen everywhere. The appointment is being deleted from the schedule, or recorded under a different status, or the field is never populated. Whichever it is, the practice has no record those appointments ever existed.
Why the empty row was the most interesting one on the table
Every other row said something about where money was sitting. This one said something about what the practice could not see at all.
A cancellation and a no-show produce the same visible outcome. An empty slot, a provider with a gap, revenue that did not happen. Treated as one number, they look like a single problem with a single fix.
They are opposite problems.
A cancellation comes with notice. That notice is the whole difference, because a slot with lead time can be refilled from a waiting list, from patients wanting to come in sooner, or from the referral queue. The revenue is recoverable and whether you recover it depends entirely on how fast the opening reaches somebody who can fill it. That is a scheduling density question.
A no-show comes with nothing. The slot is gone before anybody knows it is available, and nothing on the day recovers it. The work that would have prevented it sits weeks earlier in confirmation practice, reminder timing, and understanding which appointment types carry the most risk.
Fast refill fixes one. Reminder discipline fixes the other. Doing the wrong one produces effort with no result, and usually convinces the practice the problem is unfixable.
The number you cannot trust
There is a second consequence and it is worse than the first.
This practice recorded 3,765 no-shows. That figure is now unreliable in both directions.
Late cancellations are probably landing inside it, because a patient who calls an hour before arrives in the data looking identical to one who never called. That inflates the count.
Early cancellations that were deleted from the schedule are missing entirely. That deflates it, and it removes the exact population that was most recoverable, because a slot cancelled twelve days out is a nearly free slot.
So the true count of lost slots is higher than 3,765 and nobody can say by how much. Any effort to reduce no-shows at this practice was being measured against a number that does not describe what it claims to.
Why this happens
Usually convenience. Deleting an appointment clears the schedule view immediately and cancelling it may leave a marker somebody has to look at. Under pressure, the faster action wins, and nobody ever decided that cancellations should stop being recorded.
Sometimes it is configuration. The status exists and nothing requires it, so it goes unused and the practice assumes the absence of data means the absence of the event.
Either way, the deletion is not a data hygiene issue. It is a slot nobody knew was open.
Two cancellations, twelve days apart
Take the same appointment cancelled at two different moments and watch what changes.
A patient cancels a Thursday slot twelve days out. Somebody sees the opening on Monday, works down a list of patients who asked to come in sooner, and the slot is filled by Wednesday. The practice loses nothing. The patient who took it was waiting anyway.
The same patient cancels the same slot at eight in the morning on the day. Nobody is going to fill a Thursday afternoon with six hours notice. The provider sits, the hour is gone, and the outcome is identical to a no-show in every respect except that somebody made a phone call.
Both events are cancellations. Only one of them was recoverable, and the difference is entirely lead time.
Which is why the useful cancellation metric is not a count. It is the distribution. How many arrive with a week or more, how many with a day, how many inside a few hours. That distribution tells you what share of your cancellation volume was genuinely refillable, and it is the number that says whether faster refill is worth building.
A practice that records no cancellations at all cannot produce that distribution, cannot know how much of its lost capacity was recoverable, and has no way to tell whether its problem is refill speed or patient behaviour.
Found, fixed, and held
The find is one query. Cancellations recorded in a past month, next to no-shows for the same month. If the first number is zero or implausibly small, the record is incomplete.
The fix is not technical. Decide what happens to a cancelled appointment, tell the people doing it, and stop deleting. Most systems already have the status. What is missing is a rule requiring its use, because deleting clears the schedule view faster and under pressure the faster action wins.
What holds it is a check on the ratio. Cancellations against no-shows, monthly. Any practice will have both, and a month where cancellations disappear again is a month where somebody went back to deleting.
Nobody ever decided that cancellations should stop being recorded. That is what makes it worth checking rather than announcing.
What this means for you
Pull cancellations for a single past month. If the number is zero, or implausibly small next to your no-show count, cancellations are not being captured and everything you believe about your lost slots is built on a partial record.
The fix is a policy about what happens to a cancelled appointment, followed by a check that the policy is being followed. Neither costs anything.
Grab 30 minutes with us. Prep nothing. You will see which rows are empty for the wrong reason.
Questions people ask
What does it mean when a practice records no cancellations?
That cancellations are not being captured as a status. They are still happening. The appointment is being deleted from the schedule, recorded under a different status, or the field is never populated. The practice has no record those appointments existed.
What is the difference between a cancellation and a no-show?
Notice. A cancellation gives lead time, so the slot can be refilled and the revenue is recoverable. A no-show gives none, so the slot is gone before anybody knows it is available and nothing on the day recovers it.
Why should cancellations and no-shows be tracked separately?
Because they need opposite fixes. Fast refill addresses cancellations. Confirmation and reminder discipline addresses no-shows. Doing the wrong one produces effort with no result and usually convinces the practice the problem is unfixable.
How do I check if my practice records cancellations correctly?
Pull cancellations for a single past month. If the number is zero, or implausibly small next to your no-show count, cancellations are not being captured and everything you believe about lost slots rests on a partial record.