The patient recall system most practices don't actually run

Every dental practice management system in the country can print the same three reports: patients due for hygiene, patients with treatment diagnosed and never scheduled, patients who have not walked in for two years. A patient recall system is what turns those reports into booked chair time. In most practices, what exists instead is a printout somebody works for ninety minutes on a slow Tuesday and abandons the moment the phones pick up.

That list is the cheapest demand a practice will ever have. Those patients are already in the chart, already know where the parking is, and already cost marketing money once. Nobody has to be convinced the office exists. And yet recall is the only front-desk task where no patient is waiting on the line, which means it loses every time something rings.

The rest of this is the operating detail: the four segments the list should be split into, a cadence you can copy, what the outreach conversation should actually ask for, where the "yes" gets dropped, and the four numbers leadership should see every month, broken out per segment and per location.

What a patient recall system actually is

Three things get confused with each other, and the difference matters because they have different economics.

Recall covers patients whose due date is approaching or has just passed. The relationship is intact and the ask is small.

Patient reactivation covers patients who lapsed 12 to 24 months ago with no future appointment on the books. The relationship has cooled, so the ask is bigger and the conversation is longer.

Appointment reminders confirm a booking that already exists. Reminders protect the schedule you have. Recall and reactivation build the schedule you do not have yet. A practice can have excellent reminder automation and a completely unworked recall list, and many do.

Underneath all of this sits a measurement problem. Most offices cannot state their true recall number, and the reason is incomplete tracking rather than bad performance. Practice Analytics makes this point directly in its own benchmarking across client bases. If you cannot say what last quarter's recall effort produced, you do not have a system. You have an activity.

How big the unworked list really is

Being on the chart is not the same as being on the schedule, and the national numbers make that gap hard to argue with.

Among U.S. adults aged 19–64 with private dental benefits, only 53% had at least one dental visit in 2023, according to the ADA Health Policy Institute. The figure drops to 24% for adults with public coverage and 16% for the uninsured. These are people who, in the insured case, have already paid for the visit they are not taking.

A few more figures worth having in front of you when you size the list:

• Roughly 35% of U.S. adults aged 18 and over had no dental visit in 2019, with utilization falling further in 2020 (CDC/NCHS Data Brief 435).
• 56% of adults with an oral health problem did not seek care from a provider or an emergency department, per a 2025 CareQuest Institute report.
• 21.3% of U.S. adults have untreated caries (2017–2020), also from ADA HPI.

A word on the benchmarks you will see quoted

Almost every hygiene recall benchmark circulating in this category is vendor-published. Ranges of roughly 55–74% for average practices, with top performers meaningfully higher, appear in industry compilations such as this recall-rate benchmark roundup. Treat that as directional. The honest version of the claim is that somewhere around a third of patients do not come back on schedule, and that your own number is the only one that matters.

You will also see a "25–35% of patients are lapsed" figure repeated across the category. We could not trace it to a primary source, so it is unlinked here and we would not build a business case on it. Count your own four segments instead. It takes an afternoon, and the number will be real.

Why recall fails operationally, every time, for the same six reasons

None of this is about how hard anyone at the front desk works. It is a problem with how the work is designed.

1. No owner. Recall belongs to whoever has a spare afternoon, which means it belongs to nobody during a busy week.
2. No cadence. Outreach happens in bursts: two hundred calls in March, none in April, a panic push in June when the hygiene schedule looks thin.
3. No escalation path. One voicemail goes out, nobody calls back, and the patient gets written off without anyone deciding to write them off. There is no second attempt because no rule says there should be.
4. It competes with inbound. A ringing phone is a person who will hang up. A list never complains. The phone wins.
5. Single channel. Voice-only outreach gets screened, text-only outreach gets ignored, and picking one halves your reach for no reason.
6. No measurement. Ask most offices what last quarter's recall effort produced and you get an anecdote about a crown case that came back.

There is a mirror image of this problem on the inbound side. Leakage that happens before a patient is ever in the system (the new-patient call that goes to voicemail, the referral that never gets logged) is a different failure with a different fix, and we covered it in The Hidden Cost of Intake Leakage in Healthcare Practices. Recall leakage is the same money leaving through a different door, except these patients are already on the chart.

Segment the list before you touch the phone

Most reactivation advice treats "overdue patients" as one undifferentiated pile. That is why campaigns produce noise. Four segments, each with its own ask and its own expected outcome.

Segment 4 is the one to run first. It is small, the patients cancelled recently enough to remember why, and it gives you a clean read on whether your cadence and your ask work before you spend the big list.

Each segment needs its own script, its own attempt count, and its own line in the report. If all four are blended into one number, you cannot tell whether the campaign worked or whether one segment carried it.

For groups, one more rule: the segment definitions have to be written down and identical across every location. If the Naperville office counts "overdue" at seven months and the Aurora office counts it at nine, your per-site comparison is fiction. Standardizing how each site defines and works the list is what makes the reporting usable at the group level.

Cadence, attempts and channel

Here is a cadence you can copy and adjust. The point is that it is written down, so a busy week does not silently cancel attempts three through five.

Two adjustments by segment. An unscheduled treatment plan worth four figures justifies more attempts than a 24-month lapse. And overdue hygiene often converts on the first or second touch, so burning five attempts on it wastes capacity you should be spending on segment 3.

Use voice and text together because they fail differently. A text gets ignored. A call gets screened. Running both raises the ceiling on how many people you actually reach, which is the metric everything else depends on.

What a good recall call actually asks for

The most common defect in recall outreach is a vague close. "Would you like to come in sometime?" is not an ask. It is an invitation to think about it.

The ask is a specific date and time. Two options, both real, both in the schedule right now. If the patient says neither works, offer two more. The conversation ends with a booking or with a defined next action, never with "we'll follow up."

Three objection paths are worth designing before launch, because they cover most of what comes back:

• Cost or insurance. Give a next action rather than a quote. Route to the person who handles benefits, or book a time when that person is available to talk.
• Moved or changed providers. Confirm it, mark the record, and stop contacting them. A clean list is worth more than a big one.
• Not right now. Get a month out of them. Put them back in the cycle with a date attached instead of a shrug.

For unscheduled treatment, the outreach references what is on file and offers a conversation with the clinical team. It does not explain, justify or advise on the treatment itself. That line is not negotiable, and it is a good reason to run these conversations off an approved script with defined escalation points rather than improvising.

The "yes" has to land in the schedule, not in a callback queue

Most reactivation campaigns fall apart right here, after the outreach has already done its job.

An outbound campaign manufactures inbound volume. You send two thousand texts, and people reply. They call back. They ask about insurance. They want a different time. All of that lands on the same front desk the campaign just outcompeted for attention, and now the desk is handling its normal Monday plus the campaign's tail. WEO Media names the outcome in its own guidance on reactivation campaigns: outreach that generates more replies than the front desk can handle produces leads that never convert.

So the design rule is simple. When a patient says yes, the appointment gets written into the practice management system during that conversation, not onto a sticky note, not into a callback queue, and not left to "someone from our office will reach out to confirm."

Every handoff you introduce is a chance for the patient to cool off and for phone tag to start. It is the same failure mode that eats inbound new-patient calls, which we broke down in How to Capture Every Patient Inquiry Without Adding Staff. Campaign overflow should route exactly the way any other inbound call routes, through the workflows you already trust, instead of into an improvised process invented the week of the push.

Why hiring a recall coordinator is the wrong fix

The default answer to an unworked list is to give it to a person. It is a reasonable instinct and it fails for structural reasons.

A single owner is a single point of failure. The backlog rebuilds during vacation, during sick days, and completely during the six weeks between one coordinator leaving and the next one getting up to speed.

Adding a campaign adds callers, not answerers. Ten patients call back at once. One person can talk to one of them. The other nine wait, and some of them hang up. The campaign added demand to the phones, not capacity.

One coordinator cannot produce consistent execution across twelve offices. They will work the sites they know best, and per-site results will diverge for reasons nobody can trace.

None of this is a comment on the people at the desk. Front-desk teams are already fully allocated. Every hour of the day is spoken for by patients standing in front of them, ringing phones, insurance calls, and the schedule. Recall is additional work that arrives with no additional hours. That is a capacity design problem, and you solve it by changing how the work gets done rather than by asking the same team to absorb more of it.

The four numbers leadership should see every month

This is the biggest hole in how the category talks about recall. Everyone says to track your results. Almost nobody says what to track. Four metrics, in order, because each one diagnoses a different failure.

Contact rate = attempts delivered ÷ list size. Did the outreach actually go out? This is the number that exposes the "we ran a campaign" that was really 60 calls out of 900. A low contact rate is a capacity problem.

Reach rate = conversations or replies ÷ attempts delivered. Did a human engage? A low reach rate points at channel, timing or data quality: wrong numbers, wrong hours, wrong channel mix.

Booking rate = appointments scheduled ÷ patients reached, reported per segment. Did the conversation work? A low booking rate with a healthy reach rate is a script problem or a schedule-availability problem.

Production per campaign = scheduled production attributable to the campaign. Track show rate separately. A booking that no-shows is not a recovered patient, and blending the two hides the difference between a campaign that filled chairs and one that filled slots.

What the monthly report should look like

Illustrative figures. The point is the shape of the report, not the numbers in it.

Run the production math off your own labeled assumptions. Take the overdue hygiene row: 166 booked × 82% show rate = 136 kept visits. Multiply by your average hygiene visit production (substitute your real number) and you have the segment's contribution. Do the same for unscheduled treatment using your actual accepted-plan values. Numbers built this way survive a conversation with an owner. Numbers borrowed from a vendor blog do not.

For groups, every one of these breaks out by location. Industry commentary on DSO retention makes the same point. Henry Schein One's analysis frames the operational difference between top performers and everyone else this way: the highest-performing groups have real-time, organization-wide visibility into unscheduled and overdue patients, while average groups find out in a quarterly report.

Running recall as infrastructure: how Greetmate builds the program

Everything above is a program. Programs need something to run them on a schedule that a busy week cannot cancel.

Greetmate builds and operates that outbound layer for dental groups, DSOs and multi-site medical groups. The engagement starts with scoping: we sit down with your team, define the four segments against your actual data, and agree the definitions in writing so every location counts the same way. Then we build the workflow, test it against real list samples, and guide the rollout. You are not handed a builder and wished luck.

What that looks like in practice:

• Segments pulled from the practice management system, so the list is current instead of a printout from three weeks ago. Greetmate integrates with dozens of EHRs covering scheduling, notes, medications and lab results, so your team keeps working in the systems they already use.
• Workflow-triggered outbound voice and SMS on the cadence you approved, with the attempt counts you set per segment, and escalation to a staff member on the conditions you define.
• Bookings written back into the schedule during the conversation. No callback queue, no sticky notes, no second chance to lose the patient.
• Campaign overflow routed through the same inbound workflows as everything else, so the replies your campaign generates do not land as a surprise on Monday.
• Analytics that show what each segment produced, broken out by location, so leadership sees contact rate, reach rate, booking rate and production without anybody assembling a spreadsheet.

The call flows are structured, with guardrails and defined escalation paths, so outreach follows the script the practice approved. It is HIPAA-ready with a BAA available, and basic deployments are often live within hours. Engagements start at $999–$2,500 per month plus usage above an included allowance set per customer, with a two-month minimum initial engagement and a 60-day money-back guarantee on that paid engagement. Greetmate was recognized in the 2025 Globee® Awards for AI and the 2025 Titan Awards IT — AI & Automation.

If you want the detail specific to your setting, the dental and DSO workflows and the multi-location standardization model go deeper.

A 30-day plan to get the first campaign live

This works whether or not you ever talk to us.

Week 1: count the list. Pull all four segments. Write down the exact definitions and get every location using them. You now know, probably for the first time, how much demand you are sitting on.

Week 2: write the ask. Two segments only: overdue hygiene and unscheduled treatment. Write the specific-time ask and the three objection paths. One page each.

Week 3: test on the smallest segment. Run cancelled-never-rebooked. It is small, it closes fast, and it will expose whether your cadence and handoff work. Measure all four metrics.

Week 4: fix one thing, then scale. Look at where the funnel narrows. Low contact rate means capacity. Low reach rate means channel or timing. Low booking rate means the ask. Fix the weakest step, then point it at the hygiene list.

FAQ: patient recall and reactivation

What is a patient recall system?

A patient recall system is the standing process that identifies patients due or overdue for care, contacts them on a defined cadence across voice and text, and books them into open schedule capacity. A report listing overdue patients is an input. A system adds an owner, a cadence, an escalation path and per-segment measurement.

What is the difference between patient recall and patient reactivation?

Recall targets patients whose due date is approaching or recently passed, so the relationship is intact and the ask is small. Patient reactivation targets patients who lapsed roughly 12 to 24 months ago with no future appointment. Reactivation needs a re-introduction, a concrete hook and more attempts, and it converts at a lower rate.

How many times should you contact an overdue patient?

Five touches across three weeks is a reasonable default: text, call, text, call, final text, then return the patient to the list next quarter. Vary it by segment. A high-value unscheduled treatment plan justifies more attempts than a 24-month lapse. Honor opt-outs immediately and respect state calling windows.

Should recall outreach use text or phone calls?

Both, in sequence. Text first for reachability and to warm up the number, then a call within 48 hours to have the actual booking conversation. The two channels fail differently, since texts get ignored and calls get screened, so running them together raises your reach rate more than adding attempts on a single channel.

How do you measure whether a dental recall system is working?

Track four numbers per segment: contact rate (attempts delivered ÷ list size), reach rate (conversations ÷ attempts delivered), booking rate (appointments ÷ patients reached), and production per campaign, with show rate tracked separately. Each one diagnoses a different failure: capacity, channel, script, or schedule availability.

The list is already paid for

The overdue and unscheduled list is the cheapest demand a dental group will ever have access to, and the only reason it sits unworked is that nobody designed it as a program. Give it an owner. Split it into four segments with different asks. Set a cadence that a busy week cannot cancel. Make sure the "yes" lands in the schedule during the conversation. Then put contact rate, reach rate, booking rate and production in front of leadership every month, per segment and per site.

Greetmate runs that program as operational infrastructure: segments pulled from your practice management system, outbound voice and SMS on your approved cadence, bookings written back into the schedule, and reporting that shows exactly what each segment produced across every location. We scope it with you, build and test it before go-live, and stay hands-on through rollout.

Book a demo and bring your four segment counts. We will walk through what a first campaign would look like against your actual list.