Which channels produced fitted, paying patients?

This is one of the few categories where print and mail genuinely outperform, and it is measured by counting redemptions at best.

A response list from a mail campaign belongs in the lead file, tagged as mail. Included there, mail ranks on fitted revenue against paid search, which is the comparison the budget has always needed and never had.

Hearing loss is accepted slowly. Patients test, decline, and return a year or two later when the problem is undeniable.

Because there is no attribution window, that eventual fitting still credits the campaign that produced the first test — and in this category that is a large share of all revenue.

A patient fitted with two devices at the upper technology level and a multi-year care plan pays several times what the headline in the advert suggested. A patient fitted with one entry-level device pays a fraction of it. Both arrived from the same campaign, attended the same free test, and are counted identically by every metric a practice has access to. The spread inside a single advertising source is wider in this category than in almost any other local health business, and the average conceals it completely.

Unbundling makes the arithmetic harder still. Where the device, the fitting, the follow-up appointments and the care plan are priced separately, a single patient generates several transactions over eighteen months, each of them appearing as its own row somewhere. A report reading only the device sale understates the relationship; a report counting each row as a sale overstates the patient count. Neither is obviously wrong until somebody tries to divide revenue by patients and gets a figure nobody recognizes. The practice ends up with two internally consistent reports that disagree with each other, and no obvious reason to prefer one.

The straightforward approach is to export every collected payment with the patient's contact detail. Every payment then credits the inquiry that produced the patient, so revenue by source comes out right; each payment is its own sale in the count, and a source whose patients keep paying for care after the fitting accumulates more revenue over a long range. A column marking devices and services apart is not one to add: what somebody bought from an audiologist is health information, and CloseRev signs no business associate agreements. That matters commercially as well, because the care plan is what turns a fitting into a relationship and it is the part a price-led campaign never mentions.

Trial periods mean a proportion of fittings come back, and the return rate is not the same across sources. Somebody who responded to an advert emphasizing a limited-time price behaves differently eight weeks later from somebody who came in on a clinician's recommendation and spent three appointments being talked through the options. A report built on fittings performed will rank the first source highly and will still be ranking it highly after the devices have been returned and refunded. The refund goes through the accounts weeks later and never reaches the spreadsheet where the campaigns are compared, so the two documents drift apart quietly and permanently.

This is one of the few cases in this catalog where a sales figure can go backwards, and it needs to be handled in the export rather than hoped away. The workable method is to post the refund as a negative row against the same patient on the date it was issued. Revenue for that source then nets down exactly as the accounts do, and a channel with a high return rate stops looking like a channel with high volume.

A completed year exported net of returns may show that a source which looked respectable on gross fittings is close to worthless, which changes the budget conversation immediately. It also gives the clinical team a figure for something they have been describing anecdotally for years, and a figure is considerably harder to argue with than an impression formed by whoever handled the difficult appointments. The clinical team has often known which campaign produces the returns for some time, in the way a dispenser knows which appointments are going to be difficult before they start. Turning that into a figure in a report is what makes it a budget decision rather than a grievance.

The people arriving at a clinic today are not the people who arrived five years ago. A meaningful share have already bought a self-fitting device online, used it for some months, and come in because it did not do what they hoped. They arrive better informed, more price-aware and further along in the decision, and they are reached by different adverts than the people who have never tried anything. A practice still judging campaigns against benchmarks from before that shift is comparing itself with a market that no longer exists.

In practice this means test bookings have become a less reliable indicator than they were. Two campaigns producing the same number of appointments can produce entirely different conversion to a fitting, because one is reaching people considering their first device and the other is reaching people who already own one and are now evaluating a professional. Booking counts cannot distinguish them, and nothing in an advertising account can either. The difference shows up weeks later in whether a fitting happened, which is exactly the point at which the advertising record stopped being updated.

Collected revenue does distinguish them, and doing it across the last two years shows how the mix has been moving. The data for that comparison already exists in the practice's own records, which is the useful part: this is not a change a clinic has to start measuring today and understand in eighteen months. It can be measured backwards this week, against the period when the shift was actually happening. It may show one long-standing campaign declining steadily for two years underneath a stable booking count, which is the sort of thing that is obvious once seen and invisible until then.