Ask a practice owner which marketing works and you usually get an impression rather than a number. The impression is shaped by whichever patient most recently said something memorable, which is a poor basis for spending thousands of dollars a year.
Measurement in dentistry is genuinely harder than in most businesses: the decision to choose a dentist takes weeks, involves several touches, and ends in a phone call that leaves no digital trail. That is a reason to build tracking carefully, not a reason to give up on it.
What you will learn
- Why dental attribution is hard, and the two questions worth answering anyway.
- The minimum tracking stack and what each piece does.
- How to capture new patient source reliably at the front desk.
- The metrics that matter, with formulas, and the ones that mislead.
- How to calculate cost per new patient and payback period.
- A one-page monthly marketing report, and rules for when to cut a channel.
The measurement problem, stated honestly
A patient sees your sign for a year, hears your name from a coworker, searches, reads reviews, looks at your website twice, and calls. Ask them how they heard about you and they say "the internet," which is true and useless. Meanwhile the ad platform claims credit because it showed an ad, and the referring coworker is never counted.
Perfect attribution is not available. What is available is good enough to make decisions with, if you accept two rules: measure trends rather than individual patients, and never rely on a single source of truth.
The two questions worth answering
- Are new patients increasing, and at what cost per patient in total? This is the question that decides whether your marketing budget as a whole is working.
- Which channels are clearly producing and which are clearly not? Not a precise ranking. A clear separation between the working and the not-working is enough to reallocate money.
The minimum tracking stack
| Piece | What it tells you | Notes |
|---|---|---|
| Referral source field in your practice software | What the patient says, patient by patient | The backbone. Must be required and must have a short, usable list of options. |
| Call tracking numbers | Which channel produced a call | Use a distinct number per paid channel. Keep your main number consistent everywhere else. |
| Phone system reports | Calls received, answered, missed, and when | The denominator for everything. Answer rate is a marketing metric. |
| Website analytics | Visits, which pages, and which sources sent them | Configure with your vendor and counsel so no protected health information reaches a third party. |
| Google Business Profile performance report | Calls, direction requests, website clicks, and search queries | Trends and queries are the useful parts. Ignore raw views. |
| Ad platform reporting | Spend, clicks, and platform-claimed conversions | Treat claimed conversions as a hint, never as your patient count. |
| Your own monthly report | The numbers reconciled into one page | Nobody else will do this for you. It is the part that matters. |
Configure website analytics and any advertising pixels with your web vendor and your attorney. Keep advertising tracking off booking flows, intake forms, and portals unless the setup has been reviewed. Lesson 3 covers this, and the operations marketing chapter covers the current guidance.
Capturing source so the data is worth having
The referral source field fails in every practice for the same three reasons: it is optional, the list is too long, and the answers are too vague. Fix all three.
- Make the field required before a new patient record can be saved. In Open Dental, referral sources and the new patient workflow are covered in Module 2.
- Keep the list short. Eight to twelve options at most: patient referral, another dentist or physician, insurance directory, Google search or map, paid ad, website, social media, signage or drive by, mailer, event or sponsorship, returning patient, other.
- Train the ask. "How did you hear about us?" produces "the internet." Better: "Did someone recommend us, or did you find us online?" That single question splits the two largest categories cleanly. If they say online, follow with "do you remember whether it was a search, a map, or an ad?"
- Always capture the referring person's name when a patient was referred. It is what makes thank-you notes possible and it shows you who your advocates are.
- Audit weekly. Pull the new patients added and count blanks and "other." If either is more than a small share, it is a training problem and it belongs in the huddle.
Reconcile two sources every month rather than trusting one. Compare what patients said in the source field against calls by tracking number. When they disagree sharply, the truth is usually that several channels contributed, which is a reason to keep the cheap ones running rather than to cut them.
The metrics that matter
| Metric | Formula | What it tells you |
|---|---|---|
| New patients per month | Count, defined consistently | The headline number. Define "new patient" once and never change it. |
| Call answer rate | Calls answered divided by calls received | How much demand you are losing before it becomes a patient |
| Call to appointment rate | New patient appointments booked divided by new patient calls | How well the phone is handled |
| Website conversion rate | Calls and form submissions from the site divided by visits | Whether the site is a tool or a brochure |
| Cost per new patient by channel | Channel spend divided by new patients attributed to it | Which channels to expand and which to cut |
| Blended cost per new patient | Total marketing spend divided by all new patients | The honest overall number, including patients from free channels |
| New patient production, first visit and first year | Production from new patients over the period | Whether the patients you attract are the ones you wanted |
| New patient retention | Share of new patients who return within twelve to eighteen months | Whether you are buying patients or buying visits |
| Reviews received per month | Count | Whether the Lesson 4 system is actually running |
Two metrics that mislead. Impressions and clicks tell you the platform did something, not that a patient did. And first-visit revenue understates a new patient's value, which causes practices to cut channels that are actually profitable over a full year. Our KPI guide covers definition discipline, and building a practice dashboard covers assembling a weekly view.
The expensive mistake: judging marketing on new patient count alone. A discount campaign that brings thirty patients who never return is worse than a channel that brings twelve who stay for a decade. Track retention and first-year production alongside the count, or you will optimize toward the wrong patients.
Cost per new patient and payback
Hypothetical example. A practice spends $2,400 on search ads and $600 on agency management in a month, and attributes 14 new patients to the channel between the source field and the call tracking number. Cost per new patient is $3,000 divided by 14, or about $214. If a new patient contributes roughly $900 over two years, the channel returns several times its cost, and the practice should consider expanding it.
Payback period is the second question: how long until the patient has produced enough contribution to cover the acquisition cost? If a typical new patient contributes about $250 in the first visit and the acquisition cost is $214, payback is roughly one visit, which is comfortable. If acquisition cost were $700, payback would stretch across a year, which is survivable for an established practice with cash and dangerous for a startup without it.
Every number above is invented to show the arithmetic. Use your own. The method is what transfers.
The one-page monthly report
Build this once and fill it in on the same day each month. It should fit on a single page and take under an hour.
Monthly marketing report
- New patients this month, with the previous three months next to it
- New patients by source, ranked
- Calls received, answered, and answer rate
- New patient calls and the share that became appointments
- Website visits and conversions
- Profile calls, direction requests, and the top search queries
- Spend by channel
- Cost per new patient by paid channel, and blended across everything
- Reviews received and average rating
- New patient production this month and rolling twelve months
- One decision made and one test planned
The last line is what turns a report into management. Every month, one decision and one test. Reports without decisions become a filing habit.
When to cut and when to wait
The hardest part of measurement is knowing when you have enough data to act. Three rules.
| Situation | What to do | Why |
|---|---|---|
| A channel has produced fewer than about twenty clicks or calls | Wait | The sample is too small to distinguish a bad channel from a slow month |
| A channel has run under 60 to 90 days | Wait, unless it is obviously broken | Search campaigns need time to accumulate data, and dental decisions lag |
| Cost per new patient is well above your allowable figure across a full quarter | Cut or restructure | A quarter is enough to separate signal from noise |
| Calls are arriving but appointments are not | Do not cut the channel; fix the phone | The channel is working and the conversion is failing |
| Patients arrive but do not return | Look at who the campaign attracts and what it promised | You may be buying the wrong patients, often with a discount offer |
The fourth row is the most common misdiagnosis in dental marketing. Before cutting any channel, look at the call to appointment rate. If it is low, the marketing is doing its job and the front desk is not. Front Office Fundamentals Lesson 1 is the fix, and it costs nothing.
The annual review
Once a year, step back from the monthly view and ask four questions. Did total new patients grow, and did the patients we attracted match the Lesson 1 positioning? What did a new patient cost us on a blended basis, and how does that compare to what one is worth? Which channels have earned a larger share and which should be retired? And what changed in our market, our payer mix, or our capacity that should change the plan?
Then rebuild the budget from the same math as Lesson 5, with a year of real data instead of assumptions. That is the whole loop, and running it once a year puts a practice ahead of most of its competitors.
Finishing the course
You have gone from a positioning decision to a local search routine, a converting website, a compliant review system, disciplined paid advertising, and the measurement that keeps all of it honest. The durable habits are three: capture the source on every call, review the one-page report every month, and make one decision from it.
From here, the marketing and patient acquisition chapter is the reference to keep, our Front Office Fundamentals course covers the conversion step that most marketing dies at, and building a recall system covers the cheapest patient growth available, which is the patients you already have. If you are opening a practice, our Practice Startup 101 course covers the rest of the project.
Try it
- Audit the source field. Pull every new patient added in the last sixty days and count blanks, "other," and "internet." Write the percentage down. That number is the ceiling on how good your measurement can currently be.
- Rewrite the source question. Replace "how did you hear about us" with the two-step version in this lesson, put it on the call sheet, and practice it in a huddle. Re-audit in thirty days.
- Build the one-page report. Create the template with all eleven lines and fill it in for last month using whatever data you have. The gaps you cannot fill are your tracking to-do list.
- Calculate blended cost per new patient. Total every marketing cost for last month, including website, review software, call tracking, agency fees, and staff time on social media, and divide by all new patients. Compare it against what a new patient is worth.
- Check the call to appointment rate. Count new patient calls last month and how many became appointments. If it is low, stop reading reports and fix the phone before changing any marketing spend.
- Set the monthly meeting. Same day each month, thirty minutes, the one-page report, one decision, one test. Put it on the calendar for the next twelve months right now.
Check yourself
1. Why is perfect attribution impossible in a dental practice?
Because choosing a dentist involves several touches over weeks, ends in a phone call that leaves no digital trail, and relies on patient memory that flattens everything into "the internet." The practical approach is to measure trends, reconcile two or three imperfect sources, and look for clear separation rather than precise rankings.
2. What is wrong with the question "how did you hear about us?"
It produces vague answers that cannot be acted on. Asking "did someone recommend us, or did you find us online?" splits the two largest categories cleanly, and a follow-up about search, map, or ad narrows the online half further.
3. Why track new patient retention alongside new patient count?
Because a campaign that brings many patients who never return is worse than one that brings fewer who stay. Counting patients alone optimizes toward volume and discounts, which attracts exactly the patients least likely to build the practice.
4. A channel produces calls but few appointments. Should you cut it?
No. The channel is doing its job and the conversion is failing at the phone. Cutting it hides the real problem and reduces demand at the same time. Fix the call handling first, then re-evaluate the channel.
5. How long should a new channel run before you judge it?
Generally 60 to 90 days, and until it has produced enough calls or clicks to be more than noise, unless something is obviously broken. Dental decisions lag, campaigns need data to stabilize, and cutting early is how practices conclude that nothing works.
This guide is educational content and does not constitute legal, financial, tax, or clinical advice. Laws and regulations vary by state and change over time. Consult your own dental-specific attorney, CPA, and state dental board before acting.