Read this first
What this walk-through is
On the Oct 14 ops call Paul Smith showed how he has read every quarter for years: open the Go Grow month summary in Dental Intel, sort by division, ROD and practice, look at three signals per office, and only open the ROD Dashboard for the practices that fail two of the three. It is not a marketing tool. It is the read that decides whether a practice has a marketing problem at all. Nine in ten of the outliers do not.
Use this before every ROD conversation about new patients. Bring the three signals, days worked, lead time and the limited-exam share for each practice on your list, agree the cause with the ROD, and only then talk about levers. Week 2 of the Power BI SOP covers the 18-month window in depth; this page is the field procedure.
Paul's rule: the quarter, not the month
Compare a quarter to the same quarter last year. February is always down against January; a single month tells you about working days, not the practice.
Paul's rule: one day is five percent
A month has about 20 working days. Days to schedule up by one day is five percent of the month in the wrong direction. Down one day is five percent in the right direction.
Paul's rule: two of three
Net patient gain, the 12-month new-patient trend, and same quarter last year. Negative on two of the three and you open the practice. One alone is noise.
1What to set
Period: the closed quarter, this year against the same quarter last year. Right now that is Q3 2026 against Q3 2025. Paul switched the company from a monthly to a quarterly read because a monthly read punishes short months and rewards long ones.
Scope: your ROD's practices only. Paul deselects Specialty and the other division so the view loads faster and the eye stays on one book of business.
Columns you need visible: net patient gain, new-patient trend last 12 months, total new patients same period last year, days worked year over year, and the percent of new patients seen as a limited exam.
2The three signals
Net patient gain
Window: rolling 18 months. Patients seen in the last 18 months, rolling on minus rolling off.
Why it mattersThis is the size of the active patient base. It moves slowly, so a negative here is structural: retention, not a bad month.
Good looks likePositive and steady. Paul's note on the Oct 14 call: East has usually run about flat on this number, so a large positive deserves a second look too.
Bad looks likeNegative for several months running. More patients are falling out of the 18-month window than coming in.
New-patient trend, last 12 months
Window: trailing 12 months, month by month. Right now that is Oct 2025 through Sep 2026.
Why it mattersDirection, not level. It answers whether the monthly new-patient count is drifting up or down over a year.
Good looks likeFlat to rising across the year, with the seasonal dip in the short months.
Bad looks likeEach quarter lower than the last. A single soft month is not a trend.
Total new patients, same period last year
Window: the closed quarter against the same quarter last year.
Why it mattersThe cleanest apples-to-apples count. Same season, roughly the same working days, one number up or down.
Good looks likeUp, or flat with days worked flat.
Bad looks likeDown by more than a handful with days worked flat. Down with days worked down is a different problem, which is Step 3.
3What to look at
Doctor days worked and hygiene days worked, this quarter against last year. If the practice lost a provider, moved a doctor off Fridays, or closed two days a month, the new-patient count was always going to fall. That is an operations conversation, and the ROD is already having it.
October 2026 is the live example. Through Oct 12 the network was about 456 new patients behind last year. Scott's analysis split it roughly in half: half explained by 175 fewer doctor days, including the doctor summit on Thursday and Friday and the hurricane closures, and half a steeper decline in new patients than the decline in days. The second half is the part this walk-through is for.
4What to pull per outlier
Net production, year to date and the last quarter
Window: Paul reads year to date, then the trailing three months and the last month, to see whether the trend is turning.
Why it mattersConfirms whether the new-patient decline is showing up in dollars yet, or whether hygiene and existing patients are carrying the office.
New-patient lead time, doctor and hygiene
Window: current. Median days from the call to the first available appointment.
Why it mattersThis is capacity. The Q4 plan treats more than 14 days as tight. Paul's shorthand: every extra day is five percent of the month.
Good looks likeUnder 10 days for hygiene and the doctor. The practice can take the patients it is asking for.
Bad looks likeOver 14 days. Demand is not the problem. Any marketing spend here buys calls the office cannot seat.
Provider days and hygiene utilization
Window: current quarter.
Why it mattersUtilization under about 80 percent with short lead times means the chairs are there. Utilization over 85 percent means they are not, whatever the lead time says.
5The four causes
Cause A
Structural
Closed, divested, transitioning, or a doctor leaving. Anderson and the Mississippi divestitures in Q3. Jesup in Q4, where the doctor is relocating. Nothing to market. Mark it and move on.
Cause B
Days worked
Doctor or hygiene days down against last year. Holcomb is the example: a doctor moved off Fridays and the office lost two days a month. The ROD owns the schedule fix. Marketing can still do foundation work while it is fixed.
Cause C
Capacity
Lead time over 14 days or hygiene over 85 percent. Patients are calling and cannot get in. Lever is a hygienist or a template change, not ads. This is the Q4 "capacity first" bucket.
Cause D
Demand or conversion
Days fine, lead time fine, still down. These are the only outliers where a marketing lever applies, and the first check is the phones: are calls being answered, and is the new-patient call converting. Then the website and Google profile. Paid media last.
6The operational play Paul runs on cause D
- Meet with the office manager and the doctor. Are we saying yes to the patients who call?
- Are we answering the phones? What does the missed-call rate look like, and who owns it?
- If both of those are fine, then we talk about where the next patient comes from.
7How to read it
The company average is about 55 percent limited. A practice at 70 percent is seeing most of its new patients as one-and-done emergencies: a toothache, an extraction, no comprehensive exam and no hygiene relationship. Those patients often have no insurance and do not feel they have permission to become an ongoing patient.
Paul's lever here is the membership plan, not advertising. The question for the office is how many of last quarter's limited-exam patients came back for a comprehensive exam. That conversion metric does not exist in Power BI today; Paul asked Scott to build it on the Oct 14 call. Until it does, the limited share itself is the signal.
8The sentence
Practice, then the two or three signals that fired, then the cause letter from Step 5, then the lever and who owns it. Example: "Holcomb: NPG negative, Q3 new patients down 87, days worked down two a month. Cause B with a limited-exam share of 71 percent. Levers: schedule fix with the ROD, membership plan, website and Google profile with marketing ops, call analytics on. No paid media."
9Next step
Take the sentence into the ROD conversation this month. Agree the cause. If the ROD's read differs, the ROD's read wins on causes A through C and the data gets corrected on the Q4 sheet. On cause D, the Growth Manager owns the lever and reports back at the Nov 30 PeerLogic checkpoint.
10Guardrails
- Never read a single month. Quarter against the same quarter last year, or the 12-month trend.
- Never call a closed, divested or transitioning practice a marketing problem. Check status first.
- Never recommend paid media where lead time is over 14 days or hygiene utilization is over 85 percent.
- Never argue the data in the room. If a number looks wrong, file the ticket and come back with a clean one.
- The ROD owns the operational diagnosis. The Growth Manager brings the data and the marketing read, not the verdict.
1The three signals
| Signal | Value | Read |
|---|
| Net patient gain, trailing 12 months | −93 (Aug −51, Sep −106, Oct to date +72) | Negative |
| New-patient trend, 12 months | 2025 YTD 1,354 → 2026 YTD 1,114 | Down |
| Q3 new patients vs Q3 2025 | −87 | Down |
Three of three negative. The practice is one of the twenty that drive half of the network's new-patient decline.
2Days worked and the ROD Dashboard
| Check | Value | Read |
|---|
| Days worked | Downshifted schedule, lost two days a month (ops note). Paul: a doctor moved off Fridays. | Cause B |
| New-patient hygiene lead time | 5 days | Room |
| Hygiene utilization | 78% | Room |
| Limited-exam share of new patients | 71% (company about 55%) | Membership lever |
| Website score / Google profile score | 59 / 72 | Foundation work |
| Neurality voice / online scheduler | Live / Live | In place |
| Paid media | None | Correct for now |
3The one-line read
Holcomb Family Dentistry Parks Pace · Kelly Caldwell
SignalsNPG negative, 12-month trend down, Q3 down 87. Days worked down two a month. Lead time 5 days, hygiene 78 percent utilized.
ReadCause B with room in the schedule. Ops owns the schedule. Marketing keeps the foundation work: website and Google profile with marketing ops, PeerLogic analytics on so the phones can be read, membership plan for the 71 percent limited-exam share. No paid media until the schedule and the phones are read. Ops has it red-flagged; Amy kept it on the marketing list on Oct 14 for exactly this scope.
1Sources
| Number | Where Paul reads it | Where a Growth Manager's Claude reads it |
|---|
| Net patient gain, new-patient trend, same period last year, days worked, limited-exam share | Dental Intel performance board, Go Grow month summary | sga-data MCP: net-patient-gain, new-patients, dentist-days-worked, hygienist-days-worked, visits-d0140-pct-of-d0140-d0150 |
| Net production, lead times, provider days, hygiene utilization | Power BI ROD Dashboard (year to date, then 3 and 1) | sga-data MCP: net-production, new-patient-dr-median-lead-days, new-patient-hyg-median-lead-days, return-patient-hyg-median-lead-days |
| Practice P&L | East: Power BI. West: the Excel P&L issued about six weeks after close, moving to Power BI. | Not yet in the MCP |
| Missed calls and new-patient call conversion | Daybreak, for practices on PeerLogic analytics only | PeerLogic daily data, for the same practices; the 42 Q4 practices without analytics are being turned on now |
Lead-day and rate metrics are never summed or averaged across months. Use the latest closed month and show the prior month next to it.
1Cause to bucket
| Cause | Q4 designation | Buckets that fire | Who owns it |
|---|
| A. Structural | RED FLAG - OPS | None | ROD |
| B. Days worked | RED FLAG - OPS, or FIX if kept on the marketing list | MO-1 website, MO-2 Google profile, A-1 PeerLogic analytics | ROD for the schedule; Wyatt and Korinne for the foundation |
| C. Capacity | FIX or OPP, work order 3 to 5 | GM-6 capacity, GM-3 scheduler. Paid on hold. | ROD and Growth Manager |
| D. Demand or conversion | FIX or OPP, work order 1 or 2 | GM-5 phones, then MO-1 and MO-2, then GM-8 paid once GoHighLevel and the landing page are live | Growth Manager, with Jessica for Neurality and Adam for paid |
2Karen's lever tracker
Karen's tracker carries one ops focus lever per office: hygiene production, patient reactivation, days-open capacity, doctor productivity gap, phone conversion and the rest. The marketing columns being added next to it are the Q4 designation, the work order, the Growth Manager, the buckets firing, the paid-media tier and the next marketing action. Phone conversion is the one lever that appears on both sides; when the ROD picks it, the Growth Manager's PeerLogic read is the evidence.