Dental Practice Net Promoter Score: What a Good NPS Is and How to Use It

Only one clinic out of 42 surveyed could name a single quality improvement that came out of its Net Promoter Score results. That finding, from a systematic review of 12 peer-reviewed studies in Health Expectations, three of them involving dental clinics, is the most useful thing published about NPS in healthcare, and it has nothing to do with what a good score is.

Here's our position, and plenty of owners will hate it: if your practice is reporting an NPS above 70 and you can't name the last three operational changes that number caused, the score is decorative and you should stop putting it on the morning huddle board. A high dental NPS is more often evidence of who you asked than evidence of how you performed. NPS doesn't measure whether patients like you. It measures whether they thought you were worth it.

The Benchmarks Disagree by Nearly 40 Points

Start with the number every owner asks for first, and watch it fall apart. CustomerGauge puts the healthcare industry average at 65, with a blended figure of 63 since 2017, and says dental practices generally land between 50 and 70 depending on care quality, scheduling and follow-up. Meanwhile Dentistry Today, reporting on a practice that hit 88 in 2022, cites a healthcare industry average of 27 and describes roughly 50 as a respectable benchmark.

Those two published averages are 38 points apart. They are not describing the same thing. One is pulling from vendors whose clients survey happy patients at the end of a good visit. The other is pulling from hospital and health-system data, where the survey arrives after a bill.

The peer-reviewed dental work makes the spread even stranger. A British Dental Journal study of 64 dental practices using the Denplan Excel Patient Survey defined its above-average group as practices scoring higher than 89 and its below-average group as practices under 80. In that dataset, an NPS of 82 was mediocre. Under the CustomerGauge framing, an 82 would be an outlier worth a press release.

So when a consultant tells you 60 is good and 40 needs work, ask which population produced that number, who administered the survey, and at what point in the patient journey. If they can't answer, the benchmark is a sales tool. Our view: cross-practice NPS comparison is close to worthless, and the only comparison that carries information is your own score against your own score, collected the same way, on the same schedule, for at least four quarters.

What Most Owners Get Wrong: They Survey After the Cleaning, Not After the Crown

The specific mistake is sampling bias, and the owners who make it are the ones who turned NPS on as a checkbox inside their practice management software or their recall-texting add-on. The default configuration fires the survey after a completed hygiene appointment, to patients who showed up, and usually only to patients with a valid mobile number who have already opted into marketing texts.

Think about who that excludes. The patient who got a crown that needed two adjustments. The patient quoted $4,200 for quadrant scaling who went silent. The patient who waited 35 minutes in your reception area and then rescheduled twice. The patient whose insurance estimate was off by $600. Those are the people carrying the information you would pay for, and the default survey never reaches them.

The second mistake follows from the first: turning the aggregate score into a team scoreboard or, worse, a bonus trigger. The moment a front desk bonus depends on NPS, you have taught your team to ask for tens from the patients most likely to give them, and to skip the asks that might sting. You have bought yourself a beautiful number and destroyed the only thing the number was for.

We'd kill NPS-linked bonuses outright. Pay on collections, on recall reactivation, on treatment acceptance, on unscheduled-treatment follow-through. Never pay on a satisfaction score you also control the sampling of.

Value for Money Is the Score You're Actually Measuring

This is where the dental research earns its keep. In that British Dental Journal analysis, practices with above-average NPS scored a mean Patient Perception Index of 80 against 73 for the below-average group, and the single quality factor most highly correlated with a high NPS was perceived value for money.

Not chairside manner. Not parking. Not how new the operatory chairs looked. Whether patients felt what they paid matched what they got.

That reframes the whole exercise for an owner. A detractor score is frequently a pricing and communication problem wearing a customer service costume. The patient who gave you a 5 didn't necessarily dislike the doctor. They probably didn't understand why the buildup was a separate fee, or they compared your $1,450 crown to a $900 quote from the practice two towns over without knowing what either included, or they were a PPO patient who felt processed rather than treated because your write-off math forces 40-minute appointments on 55-minute procedures.

That last one matters more than the industry admits. The original framing from Fred Reichheld and Bain, who built NPS from the single "how likely are you to recommend" question, rests on the finding that answers to that one question correlate with repurchase, referral and growth. In dentistry, repurchase is recall retention and case acceptance. So if your NPS is genuinely soft, the hypothesis to test first is not "we need to be friendlier." It's "our fee structure, our insurance mix, or our case presentation is making patients feel overcharged."

A 9 Percent Detractor Rate Costs This Practice $63,150

Numbers make this concrete. Take a general practice with 1,800 active patients and $1.35 million in annual production, which works out to $750 of production per active patient per year.

Run a census survey rather than the default sample, and say 9 percent come back as detractors, meaning scores of 0 through 6. That's 162 patients.

Now the cost. Assume baseline 12-month attrition of 10 percent across your active base, and assume detractors leave at 40 percent. The excess is 30 points on 162 patients, so roughly 49 patients walk who otherwise would have stayed. At $750 each, that's $36,750 of production gone in year one.

Referrals compound it. If each detractor costs you just 0.15 new patients a year, through silence plus the occasional warning to a coworker, that's 24 new patients you never see. At a conservative $1,100 of first-year production per new patient, that's another $26,400.

Total annual drag: $63,150. And the attrition piece is not a one-year event. Those 49 patients would have produced for years, so on a four-year forward view the retention loss alone is $147,000.

Here's the part that makes surveying worth doing. Suppose a same-week owner or office-manager callback recovers half the excess attrition, 25 patients. That's $18,750 of retained production in year one. The cost of making those calls: 162 conversations at roughly 12 minutes each is about 32 hours. At $28 an hour fully loaded, that's about $896. Call it a 20-to-1 return on the labor.

Those attrition and referral rates are illustrative. Pull your own from your practice management software before you trust any of it: 12-month active-patient attrition, production per active patient, and new patients by referral source. The structure of the math holds even when your inputs differ. The score costs you nothing and earns you nothing. The callback earns the money.

How We'd Run the Question in an 1,800-Patient Practice

If you're going to do this, do it in a way that produces decisions rather than a dashboard tile.

• Survey completed restorative and surgical cases, not just hygiene. Fire it 24 to 48 hours after crown seat, extraction, endo, denture delivery and quadrant perio. That's where value-for-money perception is formed and where the money is.
• Survey the cancellations and no-shows too. A patient who dropped off your schedule twice is data, not a lost cause. Send them the same two questions.
• Two questions, no more. The 0-to-10 recommend question, then one open box: "What's the one thing we could have done better?" The open box is the asset. The score is the index.
• Code the verbatims into five reason buckets. Cost and value, wait and scheduling, clinical result or discomfort, insurance and billing clarity, staff interaction. Count them weekly. The counts tell you where to spend.
• Route every detractor to a human call within 48 hours, made by the owner or the office manager. Not an email. Not an apology template. A call that ends with an appointment or a clear reason why not.
• Track one outcome metric alongside the score: the percentage of detractors who have a scheduled future appointment 30 days after the survey. That is the number that proves the program works.

That last bullet is the direct answer to the one-in-42 problem. The review found NPS has limited demonstrated value for driving healthcare quality improvement, and that's not a flaw in the question. It's a flaw in what practices do after they ask it.

When Your Own Score Is Telling You Nothing

Three situations where we'd ignore the movement entirely.

First, small numbers. With 40 responses, a single detractor moves your score by 2.5 points. A quarter-over-quarter shift under five points on a base that small is noise, and treating it as a trend leads to expensive reactions to nothing.

Second, mix changes. If you added a new associate, dropped a plan, opened Saturdays, or shifted from 30 percent PPO to 60 percent PPO, your NPS will move for reasons that have nothing to do with quality. Annotate the timeline so you don't misread the cause.

Third, a score that only ever goes up. Steady upward movement with no detractors is almost always a sampling artifact, not excellence. Somebody on your team has figured out who to ask. The published dental practice that reached 88 is genuinely rare; if your number is climbing toward that territory within two quarters of turning the survey on, audit the send list before you celebrate.

Four Moves Before Friday

Pull your production per active patient and your 12-month attrition rate out of your practice management software, and write both on one page. Change your survey trigger so it fires after completed restorative and surgical cases and after second cancellations, not only after hygiene. Set a standing 48-hour rule that every detractor gets a phone call from you or your office manager, and log the reason in one of five buckets. Then book a 20-minute review in 30 days with exactly one number on the agenda: what share of those detractors now has a future appointment on the schedule.

What we're tracking next is whether the value-for-money finding holds in fee-for-service-heavy practices as strongly as it does in plan-based ones, because if it does, then a soft NPS is a pricing signal and the right response is a contract review, not a customer service training day. What would change our mind on all of this: a practice that can show four consecutive quarters of score movement and tie it to a specific operational change and a production number. If you have that, we want to see it.

If your detractor comments cluster on cost, rushed appointments or insurance surprises, that's a contract problem rather than a culture problem, and the fix is a fee and plan-mix decision you can model. Our PPO-exit calculator will show you what dropping your worst-paying plan does to production, patient attrition and take-home, using your own write-off percentages. Go run your own numbers before you spend another dollar on a satisfaction survey.