What makes a dental practice unattractive to DSO buyers?

You have walked away from practices that looked good on paper. Collections were fine. The doctor was likable. The chair count made sense and the real estate worked. Something in the file made you hesitate, and six months later you still cannot name it cleanly enough to put in a memo. That instinct is usually right, and it usually has one source. Unattractive practices are rarely bad practices. They are fragile ones. Revenue that arrives through a single person, a shrinking new patient line, or a name that nobody outside a three-block radius searches for is revenue that stops arriving the day the seller stops caring. Here is what that fragility looks like before it shows up in your year two numbers.

Why does a practice with clean collections still fail your diligence?

Clean collections tell you what happened. They tell you almost nothing about what happens next. A practice producing $1.6M on a stable overhead line looks durable right up until you ask where the production actually originated, and the answer is one operatory, one clinician, and eighteen years of accumulated goodwill that walks out the door on your closing date. You are not buying last year. You are buying the mechanism that produced last year. When that mechanism is a person rather than a system, the trailing numbers describe a relationship, not an asset. Across 201,000+ US practices the pattern holds with uncomfortable consistency: strong historical performance and weak forward signal show up together far more often than acquisition committees expect. Your diligence team feels this before it can prove it. The financial file passes. The gut says no. What the gut is reading is the distance between production and reproducibility, and that distance is the most reliable predictor of a disappointing second year you will ever have access to. Durable practices produce revenue without anyone in particular making it happen.

What does owner-dependent revenue actually look like in the numbers?

It rarely announces itself. It hides inside metrics that look healthy in isolation and only turn when you read them against each other. The tell is concentration: not just of production, but of the reasons patients said yes.

• Case acceptance collapse by provider. The owner converts high-value treatment at two or three times the rate of the associate, and nobody has asked why. The answer is trust that took fifteen years to build and cannot be assigned in a purchase agreement.
• Referral sources with a first name attached. When the specialist pipeline runs through a personal friendship rather than a documented relationship, it terminates at retirement.
• Hygiene reappointment that depends on a single coordinator. With 33.9% of practices actively recruiting hygienists, one departure in a fragile department is a production hole you cannot fill on your timeline.
• New patients who arrive by name, not by search. If people find the practice because they already knew the doctor, you are buying a mailing list.

Each of those is survivable alone. Stacked, they describe a practice whose revenue is a person. You know how that story ends because you have underwritten it before.

Why do new patient trends tell you more than trailing EBITDA?

EBITDA is a rearview measurement dressed as a forecast. New patient volume is the closest thing in this industry to a leading indicator, and it is the number sellers are least prepared to defend. A flat new patient line in a practice with growing collections is not stability. It is a practice extracting more from a shrinking base, and that extraction has a ceiling you will hit roughly fourteen months after close. Ask for the trailing 36 months, not 12. The 12 month view flatters almost everyone. The 36 month view shows you whether the practice is replacing attrition or slowly consuming itself. A practice losing 15% of active patients annually and adding 12% is dying quietly with excellent collections. Your model does not care about the collections. It cares about whether the top of the funnel exists at all. When new patient counts decline while production rises, you are looking at a practice that has already spent its future, and the seller may honestly not know it. That is the version of this you should worry about most, because there is nothing to catch in the disclosures.

How does poor visibility turn into a valuation discount?

Because it converts an assumption into a line item. Your model prices a practice as an ongoing demand engine you are acquiring. If the practice generates no independent demand, you are not acquiring an engine. You are acquiring a location, a chart list, and a rebuild project you did not budget for. This is where the discount comes from, and it is why your integration team keeps arriving at numbers your deal team did not. Around 432,000 dental searches now run through AI systems every month, and 70% of practices are invisible in those answer sets entirely. Your target is statistically likely to be in the invisible group. That means the demand you assumed transfers with the practice does not exist yet, and the cost of creating it belongs to you. Practices that already own their demand capture system arrive with something you can multiply. Practices that do not arrive with something you have to build. The gap between those two states is the discount, whether or not anyone writes it down as one.

What does an AI readiness score below 40 mean for your model?

It means the practice has no standing in the systems that increasingly decide who gets considered. Average AI readiness across the audit sits below 40 out of 100, and only 8% of practices clear 65. Those are not technology scores. They are legibility scores: how clearly the practice communicates what it does, who it does it for, and why it should be named when someone asks an AI system for a recommendation in that zip code. A score below 40 tells you the practice has never made that case to anyone. Which means every patient it currently has arrived through a channel that ends at the closing table: the doctor's name, a neighbor's recommendation, a sign on a road. You are underwriting revenue with no discoverable source. When you build the pro forma, the growth assumption you plug in for years two and three quietly depends on visibility the practice does not have. That is not a modeling error. That is a diligence gap, and it is the one that most often turns a fair multiple into an expensive one.

Why does a thin Google Business Profile change your integration timeline?

Because it is the fastest thing to fix and the last thing anyone budgets for, which means it sits untouched for the first two quarters while production drifts. Complete profiles generate seven times the clicks of incomplete ones, and 82% of local dental searches end in a Maps interaction rather than a website visit. Your target's Maps presence is therefore not a cosmetic detail. It is the front door, and on most of the practices crossing your desk that door is half painted. What this does to your timeline is subtle. You planned to spend months one through six on clinical integration, systems, and staff retention. Instead you discover in month four that new patient volume has fallen further than the transition attrition you modeled, because the practice was never actually being found, and the owner's personal referrals stopped the week he left. Now you are rebuilding demand during the exact period you needed stability. Practices that arrive with a complete profile give you a quarter back. That quarter is worth more than the negotiation you had over the multiple.

Why are referral-dependent practices the riskiest ones on your list?

Referral dependence looks like strength during diligence. High case acceptance, low acquisition cost, strong margins, a doctor who says with genuine pride that he has never had to advertise. Every one of those is true, and every one of those is a warning. A referral engine is a trust network, and trust networks are attached to individuals. When the individual leaves, the network does not transfer to the buyer. It disperses. What makes this dangerous for you specifically is the timing. Referral revenue does not fall off a cliff at close. It decays over twelve to twenty four months, slowly enough that your monthly reporting reads it as noise, until the trailing twelve suddenly looks nothing like the trailing twelve you bought. By then the diagnosis is expensive. The practices that survive this transition are the ones where patients were already arriving through discovery rather than introduction, because discovery is a property of the practice and introduction is a property of the person. When you read a referral-heavy file, you are not reading a low-cost growth engine. You are reading a countdown.

What makes a high-value procedure mix fragile instead of premium?

The mix itself is genuinely attractive. Implants are growing 8.5% annually at roughly $4,500 per case, cosmetic 6.8% at $3,800, orthodontics 5.1% at $5,500. Those are the categories your model leans on for margin expansion. The fragility is in who is currently saying yes to them and why. High-value case acceptance is disproportionately a function of clinician trust, and trust built over years does not survive a name change on the door. So the practice that shows a beautiful implant percentage may be showing you one doctor's personal conversion ability, not a repeatable case flow. The distinction matters enormously. Patients who arrive through AI-driven discovery book high-value treatment at two to three times the rate of other channels, and they do it before meeting anyone, because the decision was made during the research phase. That is a transferable asset. A practice where high-value cases close only in the operatory, only with one clinician, is a practice whose most profitable revenue line is also its least durable. You should price those two situations very differently, and most models do not.

What are you really buying when you buy a patient base?

You are buying the probability that those patients return and that new ones replace them. Nothing more. Charts are not assets. Behavior is. The audit puts the average unrealised revenue sitting inside a single practice at roughly $147,000 a year, which sounds like upside and often is. But read it carefully before you underwrite it. Unrealised revenue is only capturable if the practice can reach the people it is failing to reach, and if the practice is invisible in AI and Maps, that $147,000 stays theoretical no matter how good your operations team is. This is the moment where acquisition math goes wrong most often. You see the gap, you assume your platform closes it, and you pay for the closing. Then you discover the gap exists precisely because there is no discovery path into the practice, and your platform's playbook assumes one already exists. The upside was real. Access to it was not included. That is the difference between a practice with unrealised revenue and a practice with unreachable revenue, and only one of them belongs in the model.

Why do the practices you passed on end up in someone else's portfolio?

Because a rejected practice is not a bad practice, it is an unpriced one. DSO ownership now accounts for 32% of a $179.4B market, and competition for defensible targets has compressed the pool enough that groups are increasingly willing to underwrite the rebuild you declined. Sometimes they are wrong. Sometimes they are right, and the practice you passed on shows up two years later inside a competitor's same-store growth story, which is an uncomfortable thing to read in a trade publication. The difference is almost never the diligence. It is whether the acquiring group has a repeatable way to install visibility into a practice that had none. If your platform has that capability, fragility is a discount you can capture. If it does not, fragility is a risk you can only avoid. Knowing which of those two you are is more useful than any single deal decision, because it tells you what kind of pipeline you should be building and which rejections were actually strategy rather than caution. Most groups have never made that distinction explicitly.

What separates a durable acquisition from an expensive one?

One question, asked early: if the seller vanished tomorrow, what would still work? Not what would survive for a quarter on momentum. What would still actively produce. In a durable practice, the answer includes a discoverable name, a complete profile that patients find without knowing who owns it, an associate whose case acceptance is not half the owner's, and a new patient line that has grown for three consecutive years without anyone's personal network doing the work. In a fragile practice, the honest answer is the building and the equipment. Everything else is a person. That question is worth more to your committee than another week of quality of earnings, because it reframes the entire file from performance to portability. Portability is what you are actually purchasing. The practices worth your multiple are the ones where the reason patients choose it has been made explicit, made public, and made independent of anyone's tenure. Positioning clarity is what makes that independence possible, and its absence is what quietly makes a practice unattractive long before anyone runs the numbers.

Source: The Dental Index national practice audit · 2026

Dana eventually changed one thing about how her group screens. Before the financial review, someone searches for the practice the way a patient would, in AI and in Maps, and writes down what comes back. If the practice does not appear, that goes in the file next to the multiple. Not as a disqualifier. As a price. Her rejection rate fell and her year two variance fell with it. What she had been sensing all along was legibility: whether a practice had ever made its own case to the people it wanted, or whether it had simply been carried by one person's reputation. That is the same signal that decides whether a practice ranks in Maps and gets named by AI systems. For your targets, and for the locations already inside your portfolio, positioning clarity is not a growth initiative. It is the difference between revenue you own and revenue you rented from whoever is about to retire.