Greenville dental practices: the patients already on your list
The inactive patient list is the cheapest growth channel a Greenville practice owns, and it usually sits untouched. The free fix, and where it stops.
Show contents
Pull the inactive list out of the practice management system and work it in three segments by last visit. Give each segment a message with a specific reason to come back, call the twenty most valuable records by hand, log what happens, and keep anyone who says stop off the list for good. Do that before buying anything.
The recall report is the document nobody at the practice reads on purpose. Run it on a Tuesday in October, filter for patients whose last visit was before January, and the export comes back longer than the week's schedule. For an office that has been open a decade in Greenville, South Carolina, the count is usually in the low thousands, and every name on it already said yes once.
Those patients are not prospects. They chose the practice, sat through a new patient exam, handed over a phone number and an insurance card, and paid for a visit. The list cost real money to build and nothing to keep.
What it costs to leave alone is the subject of this story, and so is the reason reactivation keeps failing. The work is awkward, the front desk is busy, and "we miss you" is a sentence about the practice. The fix starts with an export and a phone.
What a quiet list costs a Greenville practice
A hygiene chair is paid for whether or not someone sits in it. The hygienist is on the clock, the room is stocked, and a 9:40 appointment nobody fills turns into chart notes and a call to a patient who is not coming.
The visible loss is that hour. The larger loss is the patient who never comes back, because a lapsed recall is rarely a decision. It is a move, a plan change, a card that arrived in a bad week, and then silence.
Run the arithmetic on your own list.
- Count the patients last seen 12 to 18 months ago.
- Multiply by the share with a working mobile number, then by the share who book when asked with a specific time.
- Multiply by your average hygiene visit, then add the exam and the treatment that follows a postponed visit.
A practice with 900 reachable records does not need a heroic response rate to beat a month of new patient advertising. The ad buys a stranger. The list already holds someone who chose you, and the Greenville dental market is full of offices willing to take that patient's next call.
Export four columns: last visit, last contact, phone type, and where the number came from. A record with no consent source cannot be texted, and a landline is a wasted send. The campaign is only as big as its clean, permitted records.
Why it happens here
The Greenville market makes this problem specific, and none of it is a staffing failure that a lecture fixes.
- The lapse is administrative. A move from Simpsonville to Greer, a plan change at open enrollment, a recall card that landed during a school concert. Very few inactive patients decided to leave.
- The county kept moving. Transplant patients who settled around downtown and Falls Park picked a practice before they knew the roads. That loyalty is to convenience, not to a building.
- The front desk answers the person at the window. Reactivation happens in the gaps, and there are no gaps between nine and eleven on a Monday.
- The default message is about the practice. "We miss you" asks the patient to feel something. "Tuesday at 9:10 or Thursday at 2:40" asks them to pick one.
None of that is a reason to send more messages. It is a reason to send fewer, better ones.
The manual fix
Everything below runs on the practice management export, a phone and a shared spreadsheet. No software, no vendor call, and one person can start this week.
Time since the last visit decides what the message says, because a patient who missed one cleaning and a patient gone four years are not in the same conversation.
- 12 to 18 months. One missed recall, probably still insured. Two specific times, and nothing else.
- 18 to 36 months. The plan on file may be stale. One question about insurance before any offer.
- 36 months and older. Possibly moved or seeing another dentist. Small batches, one attempt, and the record cleanup is a real result even when nobody books.
-
Pull the fields you need and nothing more.
Last visit, last contact, phone type, consent source, insurance. Two people can finish this in an afternoon.
-
Cut the segments into separate tabs.
Remove anyone with an upcoming appointment first. A patient who already has a booking is the most common build error in this project.
-
Write one message per segment.
Under 160 characters, the practice name, one specific reason to come back, and either two times or one question. No link in the first message, because an unknown number with a link reads as a robocall.
-
Call the top twenty by hand.
Rank by last-visit value or by distance. The call is where you learn that the last visit ended in a billing dispute, which no message would have fixed.
-
Log four outcomes.
Booked, not now, moved or wrong number, stop. Four codes stay usable in a busy week, and a free text column holds the useful sentences.
-
Build the stop list before the first send.
One tab, one rule. Anyone on it never receives another marketing message from this practice, and no campaign gets an exception.
-
Send in batches of a hundred.
A hundred a week is a size one person can answer, and it keeps the campaign from becoming a blast.
The phone script is four sentences. Name the practice, say you are going through the schedule, and ask whether the practice lost track of the patient or something happened. The second answer is the valuable one, and it is usually insurance, a move, or a bill the patient never resolved.
Where the manual version breaks
The manual fix works, and everyone who has run it knows where it stops.
- Three thousand records is a quarter of work. At twenty calls a day, one person with no other duties needs months, and a front desk does not have twenty calls a day.
- The spreadsheet has no memory. A booked patient still gets the third email because nobody crossed the tab. A household with one phone number gets two messages, and one STOP removes a family of four.
- The work dies in the busy week. The week reactivation matters most is the week the schedule is full and the phone will not stop.
- Nobody can report on it. A reply rate computed by hand is two weeks old and does not survive its author.
- Consent gets handled late. A relationship covers appointment communication, not marketing, so the list from the practice management system is not automatically textable.
That last point is where most campaigns should have stopped.
What a system does differently
The difference is not a better sentence. It is that the list stops being a document and becomes a set of rules that run on a Tuesday when nobody remembers the campaign exists.
The build is short. Import the export with the fields intact. Save the three segments as filters that refresh themselves instead of exports that go stale. Build a four-touch sequence across three weeks: a text that asks for a reply, a text with two times, an email for the people who ignore both, and a final note that offers a clean close. Put two stop conditions on it, one for an appointment and one for an opt-out. Replies land in a single inbox on the patient record, and quiet hours run in Eastern Time.
That is what we use GoHighLevel for, and the reason it is worth naming is the stop condition, not the message. A patient who books on day two never sees day eighteen, and a patient who replies STOP is suppressed before the next step rather than after it.
A sequence that keeps sending after the patient books is spam with a schedule. The value is the rule that removes someone the moment they no longer need the message, and the second rule that removes them the moment they ask.
It still cannot do several things. It does not know the clinical record, so it cannot decide who is due for what. It is not the practice management system. It cannot manufacture consent, and it cannot make a vague offer specific.
Three questions decide whether this project is allowed.
- No protected health information in any message. No procedure names, no diagnosis, no member ID, no "your crown is due".
- A business associate agreement is required with any vendor handling protected health information. Get it before the import, and confirm in writing that it covers texting, not only email.
- Marketing texts need prior express written consent. A number collected for appointment reminders is not that consent, and a list inherited from a retiring dentist is never that consent. Read the HHS guidance on HIPAA privacy, the FCC page on unwanted calls and texts, and the A2P 10DLC documentation before launch.
The same system, assembled and running in a week.
Most of what is described in this story can be done by hand until it cannot. When that happens, this is the stack we use, what it costs, and where it stops.
What it costs
Software is the smallest line in this budget, and it is not the line that decides whether the project works. The person who owns the replies decides that.
| Item | Rough cost | Notes |
|---|---|---|
| Platform plan | $97 to $297 per month | The entry plan covers a single practice. Higher tiers are for agencies. Confirm current pricing. |
| HIPAA configuration | Priced separately, on top of the plan | Required if the vendor touches protected health information. Confirm current pricing. |
| SMS usage | About a cent per segment | A reply and a booking exchange is four to eight segments. |
| Carrier registration | One-time fee plus small monthly charges | A2P 10DLC registration. Required for U.S. business texting. |
| Phone number | $1 to $2 per month | A local 864 number, not a toll-free line. |
| The person who owns replies | Already on payroll | If nobody answers within a day, the sequence is a slower postcard. |
A single practice usually lands between $110 and $180 a month all in, before the HIPAA configuration, and the trial runs fourteen days. Confirm current pricing before you commit, because plan limits change more often than the workflow does.
What to measure
Five numbers, checked weekly, separate a campaign from a blast.
- Reply rate by segment. Replies divided by sends. If the 12 to 18 month group is quiet, the message is wrong, not the list.
- Booked from replies. Good replies with weak bookings mean nobody is offering a time.
- Opt-out rate per send. A spike means the list, the frequency or the consent record is wrong. Stop and find out which.
- Hygiene reappointment. The share of reactivated patients who return for the next recall unchased. That number says whether the campaign produced a patient or a transaction.
- Cost per booked appointment. Spend divided by appointments, next to what a new patient costs from advertising.
If the first number is fine and the second is not, the problem is the booking path. If both are fine and the fourth is not, the campaign is recruiting one-time visits, and the recall process needs the attention instead.
Why this matters in Greenville
Greenville, South Carolina keeps handing practices the same opportunity and the same excuse. The people who arrived for work around downtown, Falls Park and the BMW plant in Spartanburg picked a dentist from a search result, and they will pick again the moment the relationship goes quiet.
The calendar helps. January resets deductibles, open enrollment in the fall changes plans, and both moments make a lapse feel like a decision to a patient who is only confused about coverage.
The Greenville edition collects the rest of the work in this market, including the retention stories about keeping patients a practice already has. Start with the export this week. If two hundred records and a phone produce bookings without software, you have learned the most important thing about this project. A practice that pairs the campaign with appointment reminders at a Boise dental office keeps recall and reactivation on one record.
How far back should a reactivation list go?
Start with 12 to 18 months and stop at 36. Older records are worth one attempt in a small batch, mostly to clean the file. A patient not seen in five years may have moved out of Greenville County or changed dentists, and the practice needs to know which.
Can we text a patient who never opted in to marketing?
Not for marketing. A patient relationship generally covers appointment communication, confirmations and reschedules, and it does not automatically cover a promotional message. If the consent record is thin, use email, which follows different rules, or pick up the phone. A purchased list, or one inherited from another practice, is never usable.
What should the first message actually say?
The practice name, one specific reason to come back, and one easy next step. "This is [practice] in Greenville. It has been about a year since your last visit. Reply YES and we will send two times this week." That earns a reply. A paragraph about how much the team misses the patient earns nothing.
How many touches before giving up?
Four across three weeks, then stop. A fifth message in the same quarter is where a busy patient turns into an opt-out. If the practice wants another attempt, wait six months and change the offer and the timing, not the volume.
Do we need a business associate agreement to text patients?
Yes, if protected health information can reach the vendor, and a practice should assume it can. Request the agreement before the import, confirm that it covers texting as well as email, and keep the signed copy. A vendor that cannot produce one is telling you something about how it handles patient data.