Which ads actually filled the chair?

Two campaigns produce forty calls each. One fills hygiene appointments; the other produces three implant consults, two of which accept. On call volume they look identical. On revenue they are not remotely comparable, and the practice that cannot tell them apart usually keeps funding the wrong one.

Dentistry makes this worse than most industries because the spread between a cleaning and a full-arch case is enormous. Averaging across them hides exactly the cases that pay for the marketing.

CloseRev ranks channels by the treatment value that was actually produced, taken from your own practice-management export. Not booked value, not estimated value — what was invoiced.

A consult in March that accepts a plan in June is the case a click-attribution tool loses. The cookie is gone, the session is long over, and the ad platform has closed the books on that month.

Because matching is on the patient's phone number or email rather than on a browser session, the June revenue is still credited to the March ad. It also means you can answer the question retroactively: the quarter that already closed can be analyzed on your first afternoon, without having had anything installed at the time.

You are uploading patient contact details and treatment revenue, so this is a fair question and it deserves a direct answer. Data is encrypted in transit and at rest, isolated to your practice's workspace, and deletable — records and the original file — in one click. We act as a data processor and provide a DPA.

The match needs a phone number or an email, an amount and a date. Clinical notes, diagnoses and treatment codes are not needed and are not read, so the smallest export that works is also the least you ever hand over.

A crown posted at $1,480 in a PPO practice is not $1,480 of anything. The contracted adjustment comes off before the money moves, the patient portion arrives on a different day from the insurer portion, and a practice participating in six plans has six different real prices for one procedure. Rank channels on production and you have partly ranked them on which insurers their patients happen to carry, which is not a marketing decision anybody made and not a thing any campaign can be adjusted to fix. The number looks precise, it comes out of the practice-management system, and it is answering a question nobody asked.

This matters more than it sounds, because plan participation and channel are correlated rather than independent. Insurance-directory traffic and searches phrased around what a plan covers deliver patients on the worst fee schedules in the practice. Searches for a named procedure, and referrals from existing patients, skew toward people paying their own way at the full fee. A production-based ranking flatters the first group and understates the second, and a practice acting on it funds the channel that fills chair time with the least profitable work available. That is the opposite of what the exercise was meant to achieve.

The correction is to export the collections column instead, or the net figure after adjustments where the system keeps them apart. The match itself has no opinion — it reads a contact detail, an amount and a date, and treats whatever number you send as the money. So the choice is yours, and the only real rule is to keep it the same from one month to the next, because a switch halfway through the year produces a trend line about your export habits. Run it both ways once: if the two rankings disagree about a channel, that disagreement is itself the finding.

The first visit is rarely where a general practice makes its money. The value of a new patient is the recall cycle: two hygiene appointments a year, the restorative work found during them, and eventually the crown, the root canal or the implant that pays for the whole relationship. A channel judged on the revenue from a patient's first appointment is being judged on the smallest number that patient will ever produce. The emergency-toothache channel wins that comparison every time, and contributes almost nothing to the schedule twelve months later, by which point the budget has already been moved toward it.

Run the same reconciliation across a full year and the ranking rearranges itself. Because the join is on the patient rather than on a browser session, every subsequent visit from that person traces back to the advert that produced the first one, whether it happened in March or the following February. Sources delivering patients who attend recall separate visibly from sources delivering a single painful evening and a patient nobody sees again. Neither is worthless, but they are different purchases, and a practice that funds them from one budget on one number cannot tell which of them it is buying.

This is also the honest argument against reading much into a single month. Thirty days of data ranks channels on urgency, because urgent problems are the only ones that start and finish inside thirty days, and a practice acting on that ranking will slowly convert itself into an emergency clinic. Comparing twelve months against the twelve before it asks about the thing a general practice actually sells, which is a patient list rather than a series of appointments. That longer comparison costs nothing extra to run, because the records were kept for billing regardless of whether anybody was measuring marketing at the time.