The question everyone asks is the wrong one
Clinic owners ask what their recovery rate should be. It is a reasonable question and it has no honest answer, because a recovery rate is a fraction and nobody agrees on the bottom of it.
If the denominator is every cancelled hour, the rate will always look bad, because a share of cancelled hours were never recoverable at all. If the denominator is only the hours that could have come back, the rate suddenly means something: it measures how well the clinic acts on the chances it actually had. Getting from the first denominator to the second requires sorting every lost hour by why it was lost, and that sorting is the real work.
Three ways an hour is lost, and only one is about speed
The first kind is lost to a rule. The authorization carries a daily limit on units, so a makeup cannot be stacked onto a day that already reaches it. The authorization period has an end, and a makeup after the end is spending a different unit from a different pool, if the unit exists at all. Some plans cap the week rather than the period. Whatever your letters say, an hour that has no legal place to land is not recoverable, and no amount of speed changes that. The specific terms differ by payer and plan, so the authorization on your own desk outranks anything written here.
The second kind is lost to fit. The hour was allowed and somebody could have taken it, but the somebody was wrong: unfamiliar to the child, unacceptable to the family, or clinically inappropriate for a learner whose plan needs a particular person. These hours look recoverable on a spreadsheet and are not. An available RBT is not automatically appropriate coverage is the longer treatment of that problem.
The third kind is lost to speed. The hour was allowed, an appropriate person was free, the family would have said yes, and the clinic did not get to it before the week reorganised around the hole. This is the only kind a faster process recovers. It is also, in most clinics, the largest kind, which is why the whole subject feels like a speed problem even though it is not only one.
Why lumping them together hides the truth twice
A clinic that does not split its losses makes two mistakes at once.
It blames its process for hours the process could never have saved. A scheduler who is told the recovery rate is poor, when a third of the misses were rule losses, learns that the target is impossible and stops trying for the hours that were possible.
And it fails to see the rule losses as a separate problem with a separate fix. If a family cancels the same weekday over and over and the makeup can never fit inside the daily limit, the answer is not a faster callback. It is a conversation with the family about the standing schedule, or with the treatment team about the plan. That conversation only happens if someone can see that the losses are rule-shaped.
What the split lets you decide
Once each lost hour carries a reason, three decisions become possible that were not before.
Speed losses tell you how much a faster first response is worth, which is the argument for fixing the workflow before buying anything. The first callback within the hour is where that starts.
Rule losses tell you which children are consistently unable to receive their authorized hours, which is a treatment-planning signal and sometimes a reauthorization risk. What happens to the units themselves is covered in what happens to unused ABA authorization units.
Fit losses tell you where the roster is thin around a particular child, which is a staffing and continuity signal rather than a scheduling one.
None of those decisions is available from a single percentage.
How to build the split from what you already have
You do not need software to start. Take last month’s cancellations and, for each one that was not made up, write down one of three words: rule, fit, or speed. Be strict about rule, because it is the one people reach for to excuse a speed loss. An hour is a rule loss only if there was genuinely no allowed slot in the period. If there was an allowed slot and nobody offered it, that is speed.
Most clinics that do this find the speed column is bigger than they expected and the rule column is smaller, which is uncomfortable and useful. If you want to put a rough value on the speed column, the Lost Hours Calculator will do it from clinic-level counts.
What the software does, exactly
In Infinite Suite OS, a makeup offer is checked against the authorization cap and the daily unit limit before it books, so a rule loss is refused before a family is asked to say yes to something that cannot happen. Each offer records what became of it: claimed, declined, or never had an allowed place to go. That is the split above, kept by the workflow instead of by hand.
The software does not decide whether a makeup should happen, does not judge clinical fit, and does not check credentials or state rules on the recovery path. It also does not make a rule loss recoverable. If the period has closed, it has closed. If you would like to see how the reasons show up on the board, book a walkthrough and bring a month of cancellations you were never able to explain.