Most front desks work a recall list the same way: pull everyone overdue for a hygiene visit or a follow-up, and start dialing from the top. It feels productive. Somebody is on the phone. But if you time it, you'll notice most of those calls go to voicemail, and the ones that connect are a coin flip on whether the patient actually books.
The problem isn't effort. It's order of operations. Dialing a cold recall list front-to-back spends your most expensive resource — a human at the front desk — on the contacts least likely to answer. Flip the sequence, and the same staff hour books more chairs.
Full disclosure: I work for ReadySMS, and we sell the SMS this post leans on. I'll show the math so you can check it against whatever you're using now.
Why dialing everyone burns the front desk
Here's the constraint nobody writes down. One staffer working manual dials handles somewhere around 100–120 records in a full day. Between ringing, voicemails, wrong numbers, and the patients who pick up and want to talk for four minutes, a recall list of 800 overdue patients is a full week of one person's time.
And most of those calls produce nothing. Recall lists skew toward people who've already ignored a postcard and an email. Cold-dial connect rates on that kind of list land somewhere around 15–25% — call it 20% as a round working number. So out of 800 dials:
- ~160 connect
- ~640 hit voicemail, no-answer, or a dead number
You just spent a week of front-desk labor to have 160 live conversations. Some of those book. But the 640 misses cost you exactly as much dialing time as the connects did — you don't find out it's a voicemail until the phone has already rung six times.
Text first, and the list sorts itself
Send a compliant recall text to the whole list before anyone touches the phone. Something plain: "Hi [name], it's [practice] — you're due for a cleaning. Reply YES and we'll grab you a slot, or STOP to opt out."
Opted-in patient lists respond to SMS far better than they answer cold calls — response rates for a well-timed recall text often land around 30–50%, framed loosely. Two things happen:
- A chunk of patients reply YES or call in on their own. Those book with near-zero front-desk dialing time.
- The non-responders self-identify. That's your dialer list — smaller, and pre-warmed by a message the patient has already seen.
On 800 patients, a 35% text response rate means ~280 patients raise their hand. Even if only half of those actually book, that's ~140 appointments your front desk barely lifted a finger for. The remaining ~520 non-responders become the dial pass — and because they've seen your name in a text an hour earlier, connect rates on that second pass tend to run higher than a truly cold dial.
One caution before you send: a recall reminder and a "we've got 20% off whitening this month" promo are not the same consent. Crossing that line is the mistake I see most often — here's the exact distinction, and a template pack annotated for the consent line if you want language you can paste.
The worked comparison
Let's put both approaches side by side on the same 800-patient list. Assume $22/hr fully-loaded front-desk cost, and that a manual dial pass runs ~15 records/hour when you account for connects and conversations.
| Dial-everyone | Text-then-dial | |
|---|---|---|
| Contacts texted | 0 | 800 |
| SMS cost (1 segment each) | $0 | 800 × ($0.02 + $0.0045) = $19.60 |
| Text responders (~35%) | — | ~280 |
| Books from text, near-zero dial time | 0 | ~140 |
| Records left to dial | 800 | ~520 |
| Front-desk hours to dial (at 15/hr) | ~53 hrs | ~35 hrs |
| Labor cost | ~$1,166 | ~$770 |
| Books from dial pass (warmer list) | ~40 | ~50 |
| Total appointments booked | ~40 | ~190 |
| Total cost (labor + SMS) | ~$1,166 | ~$790 |
| Cost per booked appointment | ~$29 | ~$4.20 |
The text costs almost nothing — 800 single-segment SMS at ReadySMS Standard pricing is under twenty bucks. What it buys you is 18 fewer hours of dialing and roughly 4x the appointments. The dial pass still does real work; it just works a shorter, warmer list.
Keep the message to one segment and this stays cheap. A recall text at 160 GSM-7 characters is one segment. Add an emoji and the limit drops to 70 characters, splitting a normal-length message into 2–3 unicode segments and tripling the send cost for no clinical benefit. Skip the emoji on transactional recall.
Why the phone still earns its seat on the non-responders
Text-first doesn't mean text-only. The ~520 non-responders include patients who never read texts, patients with a stale mobile on file, and older patients who'll happily book if a human calls. Those chairs don't fill themselves.
This is where a dial pass on the non-responder segment does real work — but on a shorter, warmer list than the full 800. Because those patients saw your name in a text an hour earlier, connect rates on the second pass tend to run higher than a truly cold dial, so each front-desk hour on the phone books more chairs.
The speed-to-lead version of the same idea
The text-then-dial order isn't only for a batch recall list. It's the exact pattern for new-patient inquiries, just compressed in time.
A prospective patient fills out a form or replies to an ad. Fire an instant text ("Thanks — we can see you this week, want a call?"), and if they don't reply within a couple of minutes, call them while intent is still hot. The 60-second window data is stark: close rates fall off a cliff once a lead sits 30–60 minutes. Same sequencing logic, different clock.
Compliance guardrails that don't slow you down
Two things to get right before the first send, because carrier filtering and TCPA exposure both bite hard if you skip them:
- 10DLC registration. Recall texts go out on a registered A2P route. Brand and campaign registration happens in-app, runs roughly ~$10/mo per brand plus ~$20/mo per campaign, and typically approves in 1–3 days. Unregistered traffic gets carrier-filtered — your recall texts quietly don't arrive. Here's the 10DLC explainer if it's new to you.
- Automatic STOP handling and quiet hours. Inbound STOP opts the patient out across campaigns, and sends are held outside permitted local hours. That's baseline TCPA hygiene — every text at $500–$1,500 of potential exposure is not a gamble worth taking. For dial passes on larger cold-adjacent lists, a litigator/DNC scrub at $0.005/contact suppresses the numbers most likely to file.
And keep PHI out of the message body — a recall text can be perfectly legal and still breach HIPAA in what it says. "You're due for a cleaning" is fine. Naming a diagnosis is not.
The takeaway
The recall list isn't the problem. The order you work it is. Texting the whole list first costs pennies, books the easy chairs with almost no labor, and leaves you a shorter, warmer list to dial — which is where your front-desk phone time earns its keep on the non-responders. On an 800-patient list, that's roughly the difference between ~40 appointments at ~$29 each and ~190 at ~$4 each.
If you want to plug your own list size, response rate, and staff cost into this, the cost calculator will do the arithmetic. Start with the 2,500 free SMS credits, run one recall batch text-first, and compare the booked-chairs count against your last dial-everyone week. The math tends to make the case on its own.