The Arch Standard
Every number this industry publishes is undefined. Cost per lead in this niche is quoted anywhere from 15 dollars to 750 dollars, which is not a performance gap, it is four different definitions of the word lead. Here is ours, in writing, so you can hold us to it.
Section 01
Before media runs, we profile the census and demographic composition of your actual service radius: age distribution, household income bands, homeownership, insurance mix, and provider density. That profile is then matched against practices in our network with a comparable composition, and the campaign structure that produced arches in those markets becomes the starting point for yours.
This is a starting point, not a promise. A comparable market is a reason to expect a structure to work, not evidence that it did work in your market, and we do not report it as if it were. What the match buys you is the removal of the discovery phase, which is the phase most practices pay for twice.
It only works at volume. Comparables require enough markets to compare against, which is why the network figure is a mechanism in this method rather than a credential on a sales page. Stated plainly: that network is mixed specialty, spanning general dentistry, orthodontics, pediatric, and implant practices. The market level read generalizes across all of it. The full arch case level comparables are the full arch accounts only, and that subset is smaller than the network figure. We report it that way rather than letting the larger number stand in for the smaller one.
Practices in network
1,000+
Practices marketed for across our network, spanning general dentistry, orthodontics, pediatric, and implant practices. This is a reach figure. It is not a count of full arch clients and it is not a production claim. We are not going to inflate it into one.
Section 02
These eight terms carry the whole report. Every rate we publish is one of these divided by another one of these, so the definitions are the contract. If we change one, the change is dated and visible here, and prior reports keep the definition they were written under.
Term
Definition
Section 03
A conversion rate without the number it was calculated from is a decoration. Ninety percent show rate on ten booked consults and ninety percent on two hundred are different businesses. Both get written as ninety percent by everyone else in this category. On our reports the denominator sits next to the rate, every time, so the rate can be checked rather than admired.
Treatment planned production is the number most of this industry reports, because it is the biggest number available and it is available first. The gap between planned and collected is large: cases get downgraded, financing gets declined, patients disappear between the plan and the deposit. Reporting on planned roughly doubles a return on ad spend figure without anybody technically lying. We report collected, reconciled monthly against the practice management system.
A lead generated in January that closes in July belongs to January. Report by lead cohort, not by close month, or the numbers flatter whichever month you are standing in. Close month reporting makes a bad month look great the moment an old cohort lands, and it makes month one of any engagement look like a failure. Cohort reporting tells you what the media actually bought.
Disqualification reason is coded on every disqualified lead and appears in the monthly report: out of radius, not a clinical candidate, failed the financial screen, never reachable, wrong treatment entirely. Nobody in this category publishes a disqualification breakdown. It is the most useful page in the report, and it is the page that tells you where to fix the practice rather than the ads.
Section 04
Six standing refusals
Section 05
Attributing a seated arch back to the ad that produced it takes seven working parts. Missing any one of them and the report reverts to counting form fills.
If the pipeline says contacted and the definition says two way conversation, then an outbound attempt cannot sit in that stage. Every report we publish is only as honest as the stage the front desk clicks, so the stages get named after the definitions and nothing else.
The click identifier is written to the CRM record on submission and carried for six to twelve months. This is where most implementations fail, because the platform's own attribution window is far shorter than a full arch sales cycle, and a case that closes in month seven arrives with no origin attached.
Browser side pixels lose events to blockers, private modes, and iOS. Sending server side with hashed identifiers recovers the loss, and deduplication keeps one real event from being counted twice when both paths fire.
When a case moves to started or collected, that event goes back to the ad platforms carrying the real opportunity value, not a flat placeholder number. A placeholder tells delivery every case is worth the same, which is the fastest way to buy more of the cheapest case.
The majority of dental conversions happen on the phone, and an untracked call is an unattributed case. Dynamic number insertion keeps the source attached to the call so a phone lead is not quietly credited to nothing.
The number we report is collected, which means it has to be checked against the ledger rather than the CRM. This is the step almost nobody does, and it is the step that decides whether the rest of the stack is reporting reality.
Once real collected values are flowing back, delivery can optimize toward the dual arch case instead of the cheapest form fill. Until then, bidding is optimizing toward volume it cannot price.
Most practices cannot run this, and it is not a competence problem. Practice management systems are closed and hostile to integration, so the reconciliation step is manual work somebody has to own every month. The person who would own it is usually the same person answering the phone. That is why this is part of the engagement rather than a line item.
Section 06
Ad platforms categorize an account as health related automatically, based on what is on your site and in your copy. Nobody sends you a notice. What happens is that the bottom of funnel optimization options quietly stop being available, and the account starts behaving worse for reasons that look like a media problem and are not.
Conversion events named after a health condition get auto disabled. So events are named neutrally, by funnel stage rather than by diagnosis, and the naming convention is documented so nobody on the team helpfully renames an event back to something descriptive six months later.
The single highest value creative rule in this category is that copy stays third person and descriptive rather than second person and diagnostic. Copy that asserts knowledge of a viewer's medical status gets flagged. Describing a treatment and who it suits does not. This one rule decides whether an account runs.
Lead data from a full arch funnel should be handled as if it were protected health information, regardless of how the legal question ultimately resolves. That means hashed identifiers, server side transmission, restricted access, and a signed authorization from anybody appearing in creative. We treat all of this as operating cost we absorb, not as a reason for anybody to be nervous.
Score your practice against these definitions, or check whether your market is still open.
