Marketing Measured Wrong

The Attribution Problem

A dental practice spends money on marketing. New patients arrive. The practice cannot connect either event to the other with any specificity. Someone asks how the new patients are doing, and the answer is a description of overall new patient volume, not a breakdown by channel, not a cost per acquisition by source, not a comparison of which channels are producing patients and which are producing clicks that become nothing.

Attribution is the term for knowing which channel produced which patient. It is the prerequisite for every other marketing decision a practice makes. Without it, the practice cannot evaluate whether its Google advertising is producing patients or activity. It cannot confirm that word-of-mouth referrals, which feel like the primary source, are actually accounting for the volume that the owner believes they are. It cannot make an informed case to stop, start, or adjust any spend. Every marketing decision is made on impression rather than evidence, and the impression is frequently wrong.

W50 in the marketing workflow source is labeled the prerequisite for every other marketing decision. That label is not rhetorical. A practice that cannot measure attribution cannot measure marketing ROI. A practice that cannot measure marketing ROI is, by definition, guessing about which channels to fund, which to cut, and which to grow. Some of those guesses will be right. The ones that are wrong are invisible, and they continue being funded because there is no evidence to correct them.

The “How Did You Hear About Us” Problem

Most dental practices ask new patients how they heard about the practice. The question is usually asked verbally at the front desk during check-in, or included on the new patient intake form as a text field. The answers that come back are: “Google,” “online,” “a friend,” “Instagram,” “I just drove by,” “my dentist retired,” “my coworker,” and dozens of other variations entered in dozens of different ways by different staff members across different days.

That data is unusable. “Google” and “online” might refer to the practice’s organic search presence, its Google Business Profile listing, its paid search ads, or a local directory that shows up in search results. “A friend” and “my coworker” are both patient referrals, but they are entered differently and will not aggregate correctly. The free-text field produces noise rather than signal, and nobody is turning that noise into a report that informs anything.

The structural fix is a controlled vocabulary: a defined list of source options that staff select from rather than type. The list should reflect the actual channels the practice uses and expects referrals from: paid search, organic search, Google Business Profile, Facebook, Instagram, patient referral, doctor referral, neighborhood mailing, a specific directory, signage, or other. The field should be required, meaning intake cannot be marked complete without a source selected. And the vocabulary should be stable enough that monthly comparisons are valid: if the option list changes every few months, month-over-month data cannot be trusted.

A controlled vocabulary applied consistently across all intake staff for 60 days produces attribution data that is usable. It produces a monthly table of new patients by source. It produces a comparison of this month’s distribution against last month’s. It produces the starting point for connecting source to production.

The Agency Report Gap

A practice working with a marketing agency receives reports on clicks, impressions, cost per click, and sometimes conversion rate to a landing page form submission or a phone call. These metrics do not connect to the practice management system. The agency is measuring what it can measure: activity on digital properties that it controls. It cannot measure whether those clicks became phone calls, and it cannot measure whether those phone calls became booked patients, unless the practice gives it the tools to do so.

The gap between “clicks” and “patients” is where most dental marketing measurement fails. An agency can report 500 clicks on a Google ad and a cost of a specific dollar amount per click. None of that tells the practice owner whether any of those clicks became a patient sitting in a chair. Without a connection between the ad and the intake record, the conversion is invisible.

The agency reports on what it can measure. The practice cannot measure what matters. Both parties leave the reporting meeting with information that is accurate and useless.

Call Tracking as the Bridge

Assigning a unique phone number to each marketing channel is the mechanism that connects click-level data to call-level data. One number for the Google paid search ad. A different number for the Google Business Profile. A different number for the Facebook campaign. A different number for the local directory listing. Each number forwards to the practice’s main line; callers experience no difference. But the source of each inbound call is now known.

When a patient calls the number associated with the Google paid search ad and books an appointment, the call is logged as originating from that channel. When that patient’s intake record is created and the source field is completed, the two records connect: this patient came from paid search. When that patient receives their first treatment, the production associated with their visit can be attributed to the channel that produced them.

Call tracking converts click-source attribution into call-source attribution. Combined with the controlled vocabulary at intake, it completes the chain: spend on a specific channel produces an identifiable inquiry, the inquiry becomes a booked patient, the patient’s source is recorded at intake, and the patient’s first-year production is attributed to the originating channel.

That chain, built and maintained, produces a monthly marketing performance report that answers the questions that actually matter: which channels are producing patients, what does each patient from each channel cost to acquire, and which channels are producing patients who go on to accept treatment.

The Dependency Chain

The dependency chain for marketing measurement runs in one direction, and no step can be skipped. W11, intake data quality, is the foundation: if the source field is not captured consistently at intake, attribution cannot be built. W49, attribution, is the middle layer: if call tracking is not assigned by channel and the source field is not controlled and required, new patients cannot be linked to their source. W54, marketing ROI measurement, is the output: if the first two layers are not in place, the monthly performance report cannot be produced.

A practice that tries to measure marketing ROI without fixing intake attribution will arrive at numbers that look specific and are not. It may be able to say that it spent a certain amount on marketing and acquired a certain number of new patients, and that the average cost per patient acquired was therefore a certain amount. But that average obscures the channel-level reality: some channels may be producing patients at a fraction of that average cost, and others may be producing almost no patients at all.

Channel-level attribution produces a different report: this channel produced patients at this cost, that channel produced patients at that cost, and this third channel produced clicks but almost no patients at any cost. That is actionable information. The average is not.

What Two Changes Produce

Adding a required source field with a controlled vocabulary at intake and assigning call tracking numbers by channel produces more actionable marketing intelligence than adding a new agency or increasing spend. These are administrative process changes, not marketing changes. They do not require a new budget line or a new vendor relationship. They require a workflow change at the front desk and a phone number assignment on the back end.

The source field change takes an afternoon to implement and requires training the front desk on the controlled vocabulary and the requirement that the field be completed for every new patient. The call tracking change requires selecting a call tracking provider, assigning numbers to channels, and updating the phone numbers listed in each channel’s advertising or profile.

Both changes begin producing useful data immediately. Within 60 days, the practice has enough data to produce a meaningful channel comparison. Within six months, the data is stable enough to inform real allocation decisions: increase spend where cost per patient is favorable, reduce or eliminate spend where it is not, identify the referral sources that are growing and invest in reinforcing them.

A practice that has neither of these changes in place is making marketing decisions with no measurement. It may be spending well, or it may be spending poorly, and it has no way to know which. That uncertainty is the marketing measurement problem, and it is one of the most consequential administrative gaps a practice can have, because every dollar spent on channels that do not produce patients is a dollar not spent on channels that do.