How DSOs Actually Evaluate Your Dental Practice: The 11 Underwriting Criteria They Score But Never Publish

Before you receive a Letter of Intent, a room you have never been in has already made decisions about your practice.

Inside that room, a Director of Quality of Earnings and Clinical Integration is presenting a memorandum to an investment committee. The memo does not describe your practice in the language your broker used — “highly profitable,” “exceptional team,” “strong growth trajectory.” It describes your practice in the language of risk-adjusted return. It identifies your coding anomalies by CDT code. It quantifies your provider dependency. It projects the integration cost of your fragmented PMS architecture. It calculates the compounding interest on your aged patient credits.

The memo concludes with a recommendation: acquire at the stated multiple, then deploy forensic audit protocols to re-trade the asset to its actual sustainable earnings base. The projected purchase price reduction: 28 to 40 percent.

You have never seen this memo. Your broker has never seen it. Your CPA does not know it exists.

This is the memo I spent years helping to execute — inside institutional consolidation at NADG and Aspen Dental. And understanding what goes into it is the foundational difference between a practice that exits at its stated enterprise value and one that loses seven figures at the closing table. Phantom EBITDA provides the definitive framework for building the defense.

What follows are the 11 criteria DSO underwriting teams actually score when they evaluate your practice — and the specific thresholds that separate premium multiples from discount ones.

Why the pitch deck is irrelevant

Your broker’s Confidential Information Memorandum is a marketing document. It is designed to attract interest and secure an LOI at a favorable multiple. It serves this purpose well. DSO buyers read it, note the headline EBITDA and collection trends, and issue a letter with a compelling number attached.

Then the actual evaluation begins.

Institutional buyers bypass your broker’s summary the moment exclusivity is signed. They extract data directly from your Practice Management Software — every procedure code, every provider, every date of service, every claim — and they benchmark it against national compliance standards. They run your refund history against your own five-year baseline to identify behavioral patterns. They model the cost of integrating your operational infrastructure. They calculate your government payer concentration risk.

The 11 criteria below represent what that evaluation actually measures. Not what your broker told you it measures. What the internal memo recommends.

The 11 criteria DSO underwriting teams score

1. Normalized EBITDA — and who controls the methodology

The first thing to understand about EBITDA in institutional dental M&A is that it is not a fact. It is a methodology. And whoever controls the methodology controls the valuation.

A multi-location dental practice with over $12 million in annual collections recently entered a sale process. Eighteen qualified institutional bidders received identical financial data and produced 18 different EBITDA calculations. The results ranged from $1.5 million to $2.6 million — a 73% variance on the same practice, the same period, the same raw numbers.

The seller’s broker selected the highest figure. The eventual buyer selected a number closer to the lowest. The final price was a negotiation between those two poles.

This is criterion one: does your EBITDA hold up when the buyer applies their normalization methodology? The three primary normalization levers institutional buyers deploy are owner compensation (normalized to 30th percentile MGMA benchmarks for your clinical role and geography), related-party transactions (below-market rent paid to founder-controlled real estate entities adjusted to third-party FMV), and pro-forma corporate overhead allocation (hypothetical post-acquisition management infrastructure costs applied against historical earnings). Each lever, applied at a standard multiple, erases millions in enterprise value before the clinical audit begins.

The threshold: A practice with a defensible, pre-emptively normalized EBITDA — calculated using institutional methodology and documented with third-party support — enters negotiations from a position of strength. A practice that allows the buyer to calculate its EBITDA for the first time during diligence has already conceded the most important number in the deal. See the full EBITDA normalization framework →

2. EBITDA margin sustainability — the trend, not the number

DSO underwriting teams do not look at a single-year EBITDA figure. They look at a 36-month EBITDA trajectory. A practice generating $800,000 in EBITDA after three consecutive years of 15% margin growth is a fundamentally different asset from a practice that generated $800,000 last year after two flat years and a spike driven by a one-time event.

Non-recurring income is removed from the trailing period with no discussion. This includes PPP loan forgiveness, Employee Retention Credits, one-time insurance settlements, and any revenue that cannot be demonstrated to recur under normal operating conditions. Two of the seven practices in a recent Q3 dental roll-up portfolio carried $310,000 combined in non-recurring income still embedded in their trailing twelve months by the time the buyer’s team examined them.

The threshold: Three years of consistent, documentable EBITDA margin growth — not revenue growth, margin growth — is what institutional buyers underwrite. Revenue growth with flat or declining margins signals operational problems that will manifest post-close.

3. Clinical code compliance — the CDT audit

This is where most practices lose the most money, and where almost no seller-side advisor is equipped to help.

Institutional buyers do not trust your P&L summary. They extract raw PMS data at the claim and CDT code level and benchmark your utilization rates against national compliance standards. Specifically, they look for procedures billed at rates that statistically deviate from what Cotiviti and similar third-party audit organizations define as clinically compliant.

The highest-yield audit target is D2950 — core buildup. A compliant practice bills this code on 30 to 40 percent of crown procedures. In a Cotiviti review of $52.8 million in actual D2950 billings, $8.4 million — 15.9% — were disallowed upon expert review. If your practice bills a core buildup on 92% of crowns, the buyer’s algorithm identifies the variance and removes the statistically non-compliant portion from your baseline EBITDA. At a 9x multiple, $108,000 in identified non-compliant D2950 revenue equals $972,000 erased from your enterprise value.

Other high-yield audit targets: D4341 and D4342 scaling and root planing (alternating with D1110 prophylaxis on the same patients — a pattern explicitly defined as fraudulent by major carriers), D7210 surgical extraction (billed for procedures meeting only simple extraction criteria), and any internal or unspecified codes used for out-of-pocket “material upgrades” on covered PPO services.

This last category is particularly lethal for practices that have negotiated creative billing structures with their PPO networks. The moment an acquiring DSO enforces master PPO contract compliance, revenue generated through balance-billing patients for materials upgrades on covered services vaporizes entirely. What appeared to be fee-for-service cash flow was Phantom EBITDA — revenue that existed on your P&L but evaporates under institutional compliance.

The threshold: Every high-revenue CDT code billed at rates outside published Cotiviti benchmark utilization ranges is an EBITDA reduction waiting to be discovered. Practices that have run their own pre-LOI clinical audit and corrected their utilization patterns before going to market present clean clinical data. Practices that have not give the buyer’s algorithm an uncontested mandate. Explore the QoE defense framework →

4. Platform vs. add-on classification — the structural gate

This is the most financially consequential underwriting criterion, and the one sellers most frequently misunderstand.

There are two distinct asset classes in dental institutional M&A, priced at radically different multiples:

Platform assets command 9x to 13x EBITDA multiples. They are defined by two non-negotiable criteria. First, scale — a hard floor of $3 million to $5 million in normalized EBITDA. A four-location group generating $1.5 million in clean EBITDA is not a platform. It is a premium add-on, regardless of what the broker’s CIM calls it. Second, enterprise infrastructure — a single interoperable Practice Management System across all locations, centralized revenue cycle management, standardized clinical compliance protocols, and a management structure that is entirely decoupled from the founding clinician’s personal production.

Add-on assets command 5x to 8x multiples. They are characterized by fragmented software across locations, owner-dependent patient flow, decentralized billing, and clinical protocols that vary by provider rather than by documented system.

The financial violence of reclassification is not incremental. The spread between platform and add-on is a chasm of 3x to 5x. A $3 million EBITDA practice downgraded from a 10x platform multiple to a 6x add-on multiple loses $12 million in enterprise value in a single boardroom conversation.

What triggers the reclassification is not the number of locations. It is the data architecture. A buyer who discovers that your four locations operate on three different, unintegrated PMS versions — that there is no centralized billing function, no unified compliance program, and that your data exists in silos requiring manual extraction — will argue correctly that this is not a platform. It is a loose confederation of independent clinical operations attached to a shared tax ID. See the full valuation framework →

The threshold: Platform classification requires documented enterprise infrastructure — not the aspiration of it. A single PMS across all locations, centralized RCM, standardized SOPs, and a management layer that functions without the founding clinician’s daily presence. Learn about PDA’s Data Room Build →

5. Provider dependency — the key-person discount

DSO underwriting teams model what happens to revenue when the founding clinician exits. If the answer is “significant decline,” the buyer prices that risk as a discount to enterprise value before the conversation begins.

Provider dependency manifests in three forensically visible ways. First, founder production as a percentage of total group production. If the founding dentist generates more than 40% of total collected revenue, the buyer models a post-close revenue deterioration scenario. Second, associate provider structure. Associates classified as 1099 independent contractors — particularly those whose patient relationships, equipment access, and clinical protocols are entirely dependent on the practice — create both successor liability and key-person concentration risk simultaneously. The IRS multi-factor test for employee versus contractor status is not ambiguous. When the buyer’s team identifies misclassification, they do not lower your multiple. They calculate unpaid FICA taxes and penalties, deduct the total dollar-for-dollar from cash at close, and lock it in escrow. Third, visa-dependent revenue. Associate clinicians on J-1 waivers or H-1B visas create a category of operational risk most sellers have never modeled. When post-acquisition billing compliance enforcement reduces production-based compensation, providers may resign — and the departure of a visa-sponsored clinician is not just a revenue event. It is a regulatory event.

The threshold: A practice where no single provider — including the founder — represents more than 35% of total collected production, where all associates are properly classified and compensated on compliant net collections models, and where patient relationships are documented as practice assets rather than individual provider assets. See production benchmarks →

6. New patient flow — system or founder?

DSO buyers pay a premium for predictable, scalable new patient acquisition. They discount practices where growth is attributable to the founder’s reputation, a single marketing vendor relationship, or any external factor that will not survive ownership transfer.

The question institutional underwriters ask is not “how many new patients does this practice generate?” It is “what produces those new patients, and will that mechanism continue without this owner?” These are not the same question.

A practice generating 80 new patients per month because the founding dentist is a local public figure with 20 years of community relationships is not generating 80 new patients per month as a system. It is generating them as a personal brand extension — one that begins depreciating the moment the sale closes.

The forensic evaluation of new patient flow runs through marketing analytics (Cost Per Acquisition by channel, campaign ROI measured against scheduled production, not just lead volume), telephony data (inbound call volume, call conversion rates, front desk performance metrics showing the team’s ability to convert organic interest into scheduled revenue), and patient retention data (the ratio of new patients to active patient count over 36 months, showing whether the practice retains patients or simply cycles through acquisition). See marketing acquisition benchmarks →

The threshold: A practice whose new patient flow can be traced to documented, channel-specific acquisition systems — systems that a new management team can operate and scale — is a growth asset. A practice whose growth depends on a person, a single Google Ads agency relationship, or word-of-mouth from a retiring founder is a liability priced accordingly.

7. Hygiene department — the recurring revenue engine

The hygiene department is the single most reliable predictor of practice financial stability that institutional buyers assess. It represents recurring, relationship-based revenue that continues irrespective of who owns the practice. A well-structured hygiene department with high pre-booking rates, strong recall compliance, and clean documentation is the closest approximation to a subscription revenue model that exists in private dental practice.

DSO underwriting benchmarks hygiene production as a percentage of total practice production. A healthy range is 28 to 35%. Below 25% suggests either hygiene capacity constraints, high attrition in the active patient base, or a practice that is over-indexed to new patient acquisition at the expense of retention. The D1110/D4910 alternating billing pattern — billing prophylaxis and periodontal maintenance codes on the same patient in alternating visits to maximize insurance reimbursement across benefit periods — is one of the first patterns the clinical audit flags. See the full hygiene department framework →

The threshold: Hygiene pre-booking rate above 85%, active patient recall compliance above 70%, and hygiene production between 28 and 35% of total practice production. Any practice below these benchmarks is either leaving recurring revenue on the table or is carrying documentation practices that will not survive clinical audit.

8. Accounts receivable and credit balance architecture

The AR aging report is a behavioral fingerprint. It tells a buyer’s analyst not just what is owed, but how the practice manages its financial obligations over time. And specifically in the 12 to 24 months before a transaction, it tells them whether administrative discipline has been maintained or whether the practice has been managed to optimize its appearance.

There is one behavioral signal in AR data that institutional buyers treat as non-negotiable: the refund rate. In a well-run practice, patient credit resolution is consistent — insurance overpayments are cleared, billing errors corrected, credits processed within a normal administrative cycle. When that pattern declines sharply in the period leading up to a transaction — more than 40% below the practice’s own historical baseline — buyers treat it as evidence that credit accumulation has been allowed (consciously or not) to inflate the collections picture ahead of a sale.

The legal consequence of aged patient credits is more severe than most founders understand. Patient credits sitting unresolved beyond the applicable state dormancy period — typically three years — become subject to state unclaimed property (escheatment) laws. The practice is legally required to report and remit these funds to the state comptroller. A practice carrying $84,000 in aged credits with five years of unfiled unclaimed property reports is not carrying an $84,000 liability. It is carrying a $113,000 liability after state interest and administrative penalties — and that liability does not transfer to the buyer in an asset sale. It stays with the seller’s legacy entity. See the full AR and collections framework →

The threshold: Monthly credit balance reconciliation with documented patient outreach for credits approaching the dormancy threshold. Current unclaimed property filings. AR aging with 90%+ of balances under 60 days. Any deviation from this profile is either a purchase price reduction, an escrow holdback, or a post-close personal obligation.

9. Payer mix concentration — the risk that changes the conversation

Heavy government payer concentration is not simply a lower-multiple risk. It is a different category of buyer risk — one that in extreme cases makes a practice functionally unsellable to an institutional acquirer regardless of EBITDA.

A practice generating 35 to 40% of collections from Medicaid, Denti-Cal, or similar state-funded programs is not a lower-quality commercial insurance practice. It is a different asset class with different risk characteristics. Government reimbursement rates are set by legislative appropriation. They can be reduced, frozen, or eliminated through policy changes the practice owner cannot anticipate or control. The buyer acquiring this practice is not acquiring a durable cash flow. They are acquiring a policy risk.

In practice, heavy government payer concentration suppresses the applicable multiple by 10 to 30% relative to a comparable commercial insurance or fee-for-service practice in the same market. It restricts the buyer pool — PE-backed DSO platforms built on commercial insurance economics frequently pass on government payer-heavy assets regardless of EBITDA. And in practices where a majority of revenue derives from a single government program facing legislative risk, the asset may be functionally unable to attract institutional capital at any multiple.

The threshold: Government payer concentration below 25% of total collections for institutional attractiveness. Practices above this threshold should begin systematic payer diversification — not because a transaction is imminent, but because the 5-year diversification track record is what buyers underwrite.

10. Operational documentation — the SOP audit

Institutional buyers acquire scalable systems. They do not acquire talented individuals. A practice that functions because of who is there — rather than because of what is documented — is not a scalable system. It is a founder-dependent organism with a staff dependency substituted for an owner dependency.

The operational audit runs through four dimensions. First, workflow documentation — are the practice’s top operational processes documented in written SOPs with version control and training records that a new employee could use to reach full competency? Second, software integration — does a single PMS serve as the data source of truth across all locations, or does the team reconcile across fragmented, incompatible systems? Third, management bandwidth — does the practice have an operational management layer that functions without the founding clinician’s daily decisions, or is the owner the single point of failure for scheduling, staffing, vendor management, and compliance? Fourth, SOP alignment with actual data — when the schedule, the ledger, and the clinical notes are cross-referenced against the documented SOPs, does the data show that the systems are actually being followed, or are the SOPs aspirational documents unconnected to operational reality?

The forensic framework for this alignment — what we call the Three-Way Clinical Match — cross-references the schedule (who was in the chair, for how long, and with which provider), the ledger (what was billed and collected), and the clinical notes (whether the documentation contains the insurance-mandated keywords required to justify high-value CDT codes). When all three tell the same story, the practice has a compliance fortress. When they diverge, the divergence is the liability. See PDA’s forensic methodology →

The threshold: Written SOPs for the top 20 operational workflows, a fully integrated PMS across all locations, a management structure that functions without the founder’s daily presence, and verifiable SOP alignment when clinical data is cross-referenced against documentation. Learn about PDA’s Data Room Build →

11. The five-year PMS record — the history you didn’t prepare

This is the criterion that ambushes sellers who have done everything else right.

Institutional buyers run a three-year P&L and a five-year PMS audit. These are not the same window. Financial statements cover the three years most sellers clean up. PMS data covers the five years that precede the transaction — including the two years before preparation began.

In that gap between years three and five, the PMS holds a record of every procedure, every billing pattern, every credit balance, and every anomaly that occurred before the seller started getting ready. That history does not disappear because the last three years are clean. It is archived in the servers, and the buyer’s analyst will find it.

The specific mechanisms they look for in the five-year window: refund history as a behavioral signal (a statistically significant decline in refund rate in the 12 to 24 months preceding the LOI is treated as evidence of credit accumulation), escheatment exposure (aged patient credits with no documented outreach or state reporting), and undocumented clinical pattern changes (a shift in procedure utilization rates that cannot be explained by an existing documentation record — a CE course, a protocol change, a staffing transition).

A practice that experienced a legitimate, clinically-supported 72% increase in periodontal diagnosis rates 30 months ago — but never documented the clinical rationale — is carrying a dual liability: a go-forward EBITDA reduction (the revenue is treated as potentially non-recurring without documentation proving it is clinically stable) and a historical indemnification requirement (the historical billings under that undocumented pattern carry retrospective audit risk from state Medicaid boards and commercial payer compliance teams). The buyer demands a specific indemnification escrow sized at 1.5x the annualized anomalous revenue, held for 36 months. That escrow is separate from, and in addition to, the standard holdback.

The threshold: The only defense against the five-year window is a clean history — built deliberately and maintained consistently. Monthly credit reconciliation, current unclaimed property filings, and documentation of every significant clinical protocol change at the time it occurs. Not assembled for a transaction. Built as a practice management standard. See PDA’s forensic analysis methodology →

What buyers score, what sellers assume, and where the gap lives

Here is the practical summary of where the 11 criteria separate a premium transaction from a discounted one:

Criterion	Premium Multiple Signal	Discount/Escrow Signal
Normalized EBITDA	Seller-calculated, third-party supported	Buyer calculates for the first time during diligence
EBITDA margin trend	36-month consistent growth	Single-year spike or non-recurring income embedded
CDT code compliance	All codes within Cotiviti benchmark ranges	D2950, D4341, D7210 utilization outliers
Platform classification	Single PMS, centralized RCM, decoupled mgmt	Fragmented PMS, owner-dependent operations
Provider dependency	No provider >35% of production, W-2 associates	Founder >40% of production, 1099 misclassification
New patient flow	Documented, system-driven acquisition	Founder reputation, single vendor dependency
Hygiene department	Pre-booking >85%, recall >70%, 28-35% of production	Below benchmarks, D1110/D4910 alternation
AR and credit balances	Monthly reconciliation, current filings	Aged credits, declining refund rate, unfiled reports
Payer mix	Government payer <25% of collections	Government payer >35%, restricted buyer pool
Operational documentation	Written SOPs, verified alignment with PMS data	Tribal knowledge, single-point-of-failure workflows
Five-year PMS record	Consistent history, documented protocol changes	Undocumented pattern changes, behavioral anomalies

The memo ends before you enter the room

Return to the memo from the beginning of this post. The Director of Quality of Earnings and Clinical Integration recommended authorizing all seven letters of intent at the sellers’ stated multiples — 8 to 10 times adjusted EBITDA — to secure exclusivity and prevent competitive bidding. The re-trade would begin after LOI execution.

The memo estimated aggregate purchase price reductions of 28 to 40% across the portfolio.

The sellers involved had brokers. They had CPAs. They had attorneys. They had negotiated multiples. And the buyer’s internal projection called for a 28 to 40% extraction from those multiples before closing.

The practices that survive this process intact are not the ones with the highest EBITDA. They are the ones whose EBITDA is defensible — fully corroborated by clinical data, operationally sustainable without the founder, compliant across every CDT code category a QoE team will examine, and documented across five years of PMS history that tells the same story the financial statements do.

Those practices exist. They are built, not discovered. They are built across years, not months. And they are built by founders who understand what the memo says before it is ever written about them.

The 11 criteria above are not a checklist for transaction preparation. They are the architectural specifications for building a practice that institutional buyers cannot re-trade. Every criterion has a threshold. Every threshold is measurable today, years before a buyer’s analyst extracts your first data file.

The question is not whether your practice will be evaluated against these criteria. It will be. Every practice that enters an institutional transaction is.

The question is whether you run the evaluation first.