What to standardize across dental offices, and what to leave local

Across dental offices, standardize the definitions and the measurements, and leave the execution and the relationships local. The group decides what a recall interval is, what counts as a worked call, how a write-off is coded, and which number each office answers for every morning. The office decides who makes the calls, how the huddle sounds, and what happens with the patient standing at the desk. If the group cannot measure something the same way in every office, it should not be a standard. If it can, it should not be optional.

The rule gets broken in both directions. Owners opening a second location standardize nothing, because they can still see both offices. Groups at eight or ten offices standardize everything, because somebody got burned once. Neither produces comparable numbers, and comparable numbers are the whole reason to run offices as a group instead of a collection, whether the group is dentist-owned or a DSO supporting affiliated practices.

The governance matrix

Five columns. Enterprise standard is what the group decides and every office runs identically. Local judgment is what the office decides, inside a boundary the group draws. Measured centrally is what the group watches without saying how to get there. Owner is the role that answers for the row, and the last column is the one number that says the row is working. The rows are the thirteen workflows that actually get argued about.

Fee schedules are the clearest row. The group owns the fee schedule and the adjustment types, because a write-off coded three ways is three different numbers. But carriers pay on their own UCR fee tables regardless of what you charge, so the office manager needs room for a courtesy adjustment when a plan paid less than the estimate. Standard codes, local mercy, one limit.

How do you decide which column a workflow goes in?

Three questions, in order.

• Does the group need to compare this number across offices? Then the definition is an enterprise standard, even if nothing else about the workflow is.
• Does doing the work well depend on knowing a specific patient, employee, or today's schedule? Then the execution is local judgment, inside the boundary the definition sets.
• Can the group verify it happened without phoning the office? If not, it cannot be measured centrally, and a standard the group cannot see is a suggestion.

The daily outreach list shows all three. The list source, the ranking rule, and the definition of worked are enterprise, or "we worked the list" means something different in every office. Who calls, when in the day, and how the conversation goes are local. The group needs one number: was the list worked by close. That is the idea behind the daily standard, extended past one list.

One row breaks the rule on purpose. PMS configuration and user permissions go entirely to the enterprise column, including things that feel local, like appointment types and adjustment reasons, because every other row reads from that configuration. If two offices name the same block type differently, the schedule template row cannot be measured. If one office has a custom adjustment type, the write-off row cannot be compared. Standardize the PMS first and most other rows get easier. Leave it local and none of them are measurable.

Standardize the definition and the measurement. Localize the execution and the relationship. Nearly every argument between a group and an office is one of those four things wearing a different name.

What goes wrong on each side?

Over-standardization is the failure groups do not notice, because the offices stop complaining. The signs:

• Office managers escalate decisions they used to make, because being wrong inside the rules is safer than being right outside them
• Scripts get read word for word, and acceptance slips in the offices that used to present treatment best
• The group starts standardizing things it cannot measure, like tone on a call, and the only enforcement left is an audit

The cost is the office manager's judgment, the most expensive thing in the building to replace. A good manager knows which hygienist runs long on Tuesdays and which balance is a misunderstanding. Take that away and the manager who had it leaves, and the replacement never develops it.

Under-standardization is louder and more common. The signs:

• Every report needs a footnote per office explaining why its numbers are not comparable
• "Presented" means a printed estimate in one office and a conversation in another, so acceptance rates cannot be ranked
• The group total looks fine while one office has not touched its list in weeks, and nothing in the total is defined tightly enough to show it

The cost is that the group cannot learn from itself. If office three books more unscheduled treatment than office five and the two define presentation differently, nobody can say why, so whatever office three does never spreads. More reporting does not fix that. The numbers are not missing; they do not mean the same thing.

The tell for which side you are on: count the rows in your matrix with an empty number column. Mostly standards with no number, and you are over-standardized, enforcing by inspection. Mostly numbers with no definition behind them, and you are under-standardized, comparing rumors.

How does the matrix change from two offices to twenty?

Two offices: decide the standards before the second one opens

The common mistake at two is deferring the matrix until there is a problem, because the owner is in both offices and can see everything. By then office two has its own PMS conventions, its own recall setup, and a front desk that learned the job from whoever was there. Retrofitting a standard onto an office with habits costs more than writing it before the lease was signed.

If you own one office and are planning a second, your standards already exist, in your head and in your PMS. Write the enterprise column before the second location opens. Copy the PMS configuration exactly. Set the huddle agenda, the list definition, the adjustment types, and the no-show definition, and hand the new office manager the local boundaries in writing. You will change some of it later. What you cannot do later is make office two's first months comparable to office one's.

Five to ten: the measured column becomes a job

Around office five the owner stops seeing everything and an operations lead appears, sometimes by title and often by accident. The measured column is that person's whole job. The owner column moves from names to roles, because a matrix that says "Maria" breaks the week Maria leaves. And the local boundaries get written down instead of understood, because the ops lead was not in the room when they were agreed.

This is also where the no-show row gets tested. A no-show rate is only comparable if every office classifies a missed appointment the same way, and a schedule gap response is only auditable if every office logs the outcome the same way. The response itself stays local. What gets logged does not.

Twenty: the matrix gets a review date

At twenty, some rows move left. Review responses and financing scripts tend to become enterprise standards, because a regulator, a lender, or a lawyer asked a question that needed one answer. Hiring stays local at the decision and becomes enterprise at the onboarding checklist. A regional layer appears, and the honest matrix admits the regional manager now holds some of the judgment the office manager used to hold.

The matrix also needs a date on it. A boundary that fit eight offices is too tight at twenty, and a number that was easy to collect by hand needs a system. Review it yearly, whenever an office is added, and whenever an office manager argues with a row; the argument is data.

Where the software fits

The matrix does not need software to decide, only to hold; enforced by memory, it drifts back to whatever each office did before. We built MyDentalForce inside a 19-office group; we were the customer first, and the product is roughly the enterprise and measured columns of this table. Morning Huddle gives every office the same start: the agenda and today's numbers on one screen. The daily list is one ranked list per office, rebuilt every morning from that office's own OpenDental, with outcomes logged and written back to the chart, so worked or not is a fact rather than a report. Production Command holds production, collections, goals, and tiered bonuses per office and rolled up, on one definition. Roles and permissions are scoped per role and per office, the last row of the matrix applied to the software. The local column stays local: the office manager still decides who calls and what gets said.

If you are deciding this for a second office, write the enterprise column this week, before the habits form. If you are holding it across ten or twenty, the question is whether the measured column is a spreadsheet somebody rebuilds or a screen every office opens. See how Morning Huddle starts every office the same way, and we will show you the matrix on your own offices.