Denise has been at the front desk for eleven years and she is very good. She knows which payer needs the claim submitted a particular way, which lab calls on Thursdays, how the statements run, the password to the portal that has no password recovery, and the reason the Tuesday hygiene column is built the way it is. None of that is written anywhere. Then Denise has surgery scheduled, or gets an offer somewhere closer to home, and the practice discovers that eleven years of institutional knowledge is leaving in a cardboard box.
Most offices respond to this by buying a binder, spending a weekend filling it, putting it on a shelf, and never opening it again. That is not documentation, it is an artifact of anxiety. This lesson is about the other kind: what genuinely deserves to be written down, what does not, how to make a document somebody reaches for on a busy Thursday, how a system stays true when reality changes, and the single test that settles every argument about whether your documentation is real.
What you will learn
- Three tests for how fragile your practice is, and what each one reveals.
- Four criteria for deciding what deserves documentation, and the things that are a waste of paper.
- Why one-page checklists at the point of use beat comprehensive manuals every time.
- How a written system gets updated when reality changes, and who is allowed to change it.
- How to run the temp test, which tells you honestly whether your documentation works.
How Fragile Is This Practice?
A practice that runs on one person's memory works beautifully right up until it does not, and the failure is sudden rather than gradual. Three questions measure it, and you can answer all three this afternoon.
The vacation test. When your longest-tenured person takes two weeks off, what stops? Not what gets harder. What actually stops, waits, or gets done wrong. If the answer includes anything that touches money or patients, that is not a vacation problem, it is a documentation gap that happens to be visible this month.
The resignation test. If that person gave notice today, how much of what they know could be transferred in two weeks? Be honest. Most offices find the answer is a fraction of it, and the rest gets rediscovered expensively over the following year.
The temp test. Could a competent temporary employee, who has done this job elsewhere but has never been in your building, run this position tomorrow from your documentation alone? This is the hardest of the three and the most useful, and the rest of this lesson is largely about passing it.
None of this is a criticism of the person. Long-tenured staff are the best thing a practice has. The risk is not that they know things. It is that the practice has quietly decided their memory is a business system, and memory is the one system you cannot back up, audit or hand to somebody else.
Do the three tests as a written list rather than a conversation, because a conversation produces reassurance and a list produces a queue. Rank the items by what they touch: anything involving money, patient records, payroll or a legal obligation goes first, regardless of how rarely it comes up.
What Deserves to Be Written Down
The instinct after failing one of those tests is to document everything, which produces the binder and guarantees the failure. Documentation has a cost: writing it, maintaining it, and the credibility it burns when it is wrong. Spend that cost where it pays.
The Four Criteria
Something deserves a written system when it meets at least two of these.
- It repeats. Daily, weekly, monthly. One-time tasks belong on a project list, not in a manual.
- It has consequences when done wrong. Money, patient safety, a legal obligation, or a mess that takes hours to unwind.
- The order matters. Sequence-dependent work is where people improvise and where improvisation costs something.
- It crosses a handoff. Anything that passes between two people or two roles, because that is where information is lost.
The systems most practices need are broadly the same, and rather than repeat the inventory, the list lives in Practice Startup 101, Lesson 5. It was written for a practice about to open, and it works just as well as an audit of a practice that has been running for a decade. Several of the documents on it exist as starting points in our templates library.
What Not to Write
Just as important, and rarely said out loud.
- Judgment. You cannot write a procedure for how to talk to an upset patient. You can write what the practice will and will not authorize, and then train the conversation. Front Office Fundamentals covers those conversations directly.
- Anything the software already enforces. If the system will not let you skip a field, do not write a step about not skipping it.
- Aspirations. A document describing how you wish the office worked is worse than nothing, because the first person who follows it and gets a different result will stop trusting every other document you wrote.
- Things that change constantly. Specific payer quirks belong in a living reference that is easy to edit, not in a manual with a version number.
Checklists That Get Used, and Binders That Do Not
The difference is not quality of writing. It is format, location and length, in that order.
What a Usable Document Looks Like
One page. If it does not fit, it is two systems pretending to be one, and splitting it will improve both. Steps in the order performed, each starting with a verb. Written for the person doing the job, not for a reader, which means no introductory paragraph explaining the philosophy of the task. Decision points stated as if-then. A name and a date in the corner: who owns this document, and when it was last known to be true.
That last detail carries more weight than it looks. A document with an owner is a document someone will fix. A document with a date lets a reader judge how much to trust it. A document with neither is anonymous and therefore nobody's problem.
Two more things earn their space. A short screen recording is often better than a written page for anything that happens inside the software, because the hard part of a software task is usually where to click rather than what to do. And a photograph is worth a paragraph for anything physical: how the tray is laid out, which valve, which cable goes where. Documentation does not have to be prose to count.
Where to Keep It
At the point of use, which is almost never in a binder in the manager's office. The opening checklist is taped inside the cabinet door where opening happens. The claim submission sequence lives next to the workstation where claims are submitted. The lab case procedure is at the lab bench.
For anything longer than a laminated card, pick one digital home and be strict about it: a shared drive folder, the practice management system's documentation area, or whatever tool the office already opens every day. The rule that makes it work is one source of truth. The moment there are two copies, one of them is wrong and nobody knows which. If you print a copy for a workstation, print the date on it and treat the digital version as authoritative.
If you have an old manual nobody has opened in two years, do not update it. Take it off the shelf entirely and start with the five systems that failed the vacation test. A stale manual on a shelf teaches the team that documentation is theater, and that lesson is expensive to unteach.
Keeping Systems True
Every written system starts decaying the day it is written, because the software updates, a payer changes a rule, a piece of equipment gets replaced, someone finds a better way. A system that does not have a maintenance path is a system that will be wrong within a year and ignored within two.
Three mechanisms handle almost all of it.
- The change trigger. Certain events automatically require a documentation review: a software update, a new payer contract, new equipment, a new role, or a mistake that cost real money. Name these events explicitly so the review is not a matter of someone remembering.
- The correction path. Anyone who follows a document and finds it wrong must have a fast, low-friction way to say so, and must see it fixed. One shared place to note it, reviewed weekly, with the fix made within a few days. The first time someone reports an error and nothing happens, reporting stops permanently.
- The scheduled review. Quarterly, alongside the template audit from Lesson 1. Not every document every quarter. Rotate through them, a handful at a time, and record the review date even when nothing changed.
Who is allowed to change a document matters as much as how. The owner can edit their own, but anything that affects another role changes only with that role's input, and anything touching money, compliance or clinical process goes through the office manager and the doctor. This is not bureaucracy. It is what prevents two people quietly running two different versions of the same process. The systems and workflows chapter covers the wider structure.
Cross-Training as Insurance
Documentation alone does not make a practice resilient. A perfect written system for a job nobody else has ever done is still a single point of failure, because reading a document under pressure for the first time is not the same as having done the work.
Practical cross-training looks like this. Every critical function has a named backup, and the backup has actually performed the job, not shadowed it. Rotate deliberately: the backup runs the task while the primary is present, then runs it while the primary is available but not watching. A half day a month is enough for most roles. And the backup is the person who reviews the documentation, because the person who wrote it cannot see what is missing from it.
One office's example. A practice that cross-trained its assistants onto the front desk for one half day a month found that the most valuable output was not coverage. It was the list of questions the assistants asked, most of which were about steps the front desk team had stopped noticing they performed. The documentation that came out of those questions was better than anything the front desk had written on its own, because it was written in response to somebody actually being stuck.
There is a fair objection, which is that cross-training costs productive time and people resist it, sometimes because job security feels tied to being the only one who can do something. That is worth addressing directly rather than pretending it is not happening. The honest framing is that nobody is allowed to be a single point of failure, including the owner, and that the practice will not punish anyone for being replaceable for two weeks. Our lesson on keeping good people deals with the trust side of that conversation.
The Temp Test
Here is the standard, and it is deliberately demanding. Could a competent temp, experienced in this role but new to your building, run this position tomorrow from the documentation alone?
Not perfectly. Not without questions. But could they open, work through a normal day, and close without anyone having to reconstruct a process from memory? That is the bar, and it is a realistic one, because temps are a normal part of dental staffing and the situations that require one arrive without notice. Our guide to working with temps in a dental office covers the logistics from both sides.
Run the test for real rather than imagining it. Pick a position. Hand its documentation to someone in the office who does not do that job and ask them to work through a normal day on paper, marking every point where they would have to ask a question. Each mark is a gap. The exercise takes about an hour per position and it produces a punch list you can actually work through, which is considerably more useful than a vague feeling that the office should be better documented.
Expect the first run to go badly. It goes badly for everyone, and the gaps are always in the same places: logins and access, what to do when the normal path fails, who to call for what, and the small sequencing details that the experienced person stopped noticing years ago.
Try this in your own office
- Run the vacation test on paper. Write down what actually stops when your longest-tenured person is out for two weeks. That list, in order of consequence, is your documentation queue.
- Write one system this week. One page, verb-first steps, an owner and a date in the corner. Pick the item at the top of the queue, not the easiest one.
- Move a document to the point of use. Take one procedure that lives in a binder and put it where the work happens, then ask the person who does that job whether they used it.
- Start the correction log. One shared place where anyone can note a document that is wrong, reviewed every week. Announce it, then fix the first entry fast enough that people notice.
- Name a backup for every critical function. Write the list, find the roles with no backup, and put one half-day rotation on the calendar this month.
- Run the temp test on one position. Hand the documentation to someone who does not do that job, have them walk a normal day on paper, and count the questions. Work the resulting list.
THE CHAIRSIDE TAKE
Do not write an office manual. Write five one-page systems for the five things that failed the vacation test, put each one where the work happens, and give each an owner and a date. That is a weekend of work and it removes most of the fragility that a three-inch binder never touched. The part I would push back on is the instinct to document everything, because a practice with sixty documents has zero maintained documents and the team knows it. What this lesson cannot decide for you is which five, and the vacation test is the only honest way to find out. Run it before you write a word.
Lesson 5 of 6 in Practice Management: Running the Day
This guide is educational content and does not constitute legal, financial, tax, or clinical advice. Laws and regulations vary by state and change over time. Consult your own dental-specific attorney, CPA, and state dental board before acting.