- Published
- Byline
- GoHighLevel Partners Editorial
- Filed under
- GoHighLevel Automation
How to wire a missed call text back automation in a client sub-account so it fires on the right calls, sends from the right number and lands in a thread somebody owns.
Missed Call Text Back: What It Does and Where It Breaks
Missed call text back is a small automation with an outsized effect on a client’s phone line: when an inbound call ends without anyone on your side answering it, the system sends that caller an SMS from the number they just dialled. The caller gets an acknowledgement while they still remember why they rang, and the agency gets the conversation into a thread that can be assigned, answered and measured instead of evaporating into a call log.
Most agencies build it in an afternoon. What separates a version that works from one that quietly annoys people is a handful of conditions, a sending number that is genuinely reachable, and a plan for what happens after the caller writes back.
How the automation is actually wired
Every version of this automation has the same three parts. A trigger listens for a call event on the sub-account’s phone number. A filter decides whether that particular call counts as missed. An action sends an SMS, and usually creates or updates a record so the caller is not a stranger the next time they appear.
The part people get wrong is the filter. A call event fires for answered calls too, and it fires for outbound calls placed by the client’s own team. If the workflow only checks that a call happened, the client’s staff will start receiving texts every time they dial a supplier, and callers who were helped on the phone will get a message telling them nobody was available. Both mistakes look identical from the outside: the automation is sending, so it appears to work.
The two conditions that do the real filtering
- Direction. Restrict the workflow to inbound calls. Outbound call events share the same trigger and are the more common source of accidental sends.
- Call status. Use the status the telephony layer reports — no answer, busy, failed — rather than inferring from duration alone. Duration is a useful second check for the case where a call is technically answered and then dropped after a couple of seconds, but on its own it will misread a long ring as a long conversation.
Which call outcomes should trigger a text
Write this table down before you build anything, and agree it with the client. It is the part of the setup they have opinions about, and the part that generates complaints if you guess.
| Call outcome | What the caller experienced | Send a text back? |
|---|---|---|
| No answer | The line rang out and nobody picked up | Yes — this is the core case |
| Busy | The line was engaged | Yes, with wording that acknowledges it |
| Failed or carrier error | The call never connected properly | Yes, and flag it for review |
| Voicemail left | The caller already explained the request | Yes, but a different message |
| Answered and handled | Somebody spoke to them | No — the common false positive |
| Answered then dropped under 10 seconds | Picked up and cut off | Treat as missed |
| Outbound call, no answer | Your client rang someone else | No, unless designed on purpose |
The number the reply comes from
The text has to arrive from the number the caller dialled. If it arrives from a different number, the caller sees an unknown sender referring to a call they made somewhere else, which reads as spam even when it is not — and their reply lands on a number nobody is watching.
Two configuration details make this true rather than assumed. First, the SMS action must be set to send from the number that received the call, not from a default sending number picked up from the sub-account settings. Second, that number must be able to receive SMS at all: some numbers are provisioned for voice only, and a text sent from a voice-only number either fails or delivers with no return path. Check the number’s capabilities in the phone settings before you test, because the workflow log will show the action as executed either way.
Registration and consent, before the first send
In the United States, application-to-person messaging on standard local numbers runs through carrier registration, and unregistered traffic is filtered rather than rejected with a clear error. That is the failure mode to plan for: the workflow says sent, the client says nobody is getting anything. Get the brand and campaign registration done for each sub-account that will send, and treat the approved use case as part of the build rather than paperwork that follows it. The A2P 10DLC registration steps that actually get approved are worth reading before you submit, because a rejected campaign is measured in days.
Consent matters for a different reason. A caller who dialled the business has given you a reasonable basis to reply to that call, but the message should stay on topic and be easy to stop. Include an opt-out instruction, honour the stop keyword at the sub-account level rather than only in your copy, and do not follow the missed call text with unrelated promotion later in the week.
Timing and wording
Send the first text almost immediately — the point is to reach the caller while they still have the phone in their hand. A delay of a minute or two is defensible if the client’s staff often call back straight away, but anything longer defeats the mechanism.
The message itself should do three things: identify the business by name, acknowledge the specific call, and ask one answerable question. A message that only apologises gives the caller nothing to reply to, and a thread with no reply is the same dead end as the missed call.
What to avoid in the first message
- A link in the very first text. Links in an opening message from an unrecognised sender attract carrier filtering and reader suspicion in equal measure.
- Multiple questions. One question, one reply, one thread to work with.
- Anything that reads as a broadcast. The caller knows they just rang; the message should sound like a continuation of that.
What happens after they reply
The automation’s job ends when the caller answers, and this is where most builds stop too early. A reply arrives in the conversations area of the sub-account and then sits there unless something routes it. Decide in advance who owns inbound texts, assign the conversation to that user, and set a visible expectation for how quickly it gets answered — a conversation SLA with a real escalation path turns “someone will see it” into a rule with a timer on it.
The workflow also has to stop talking once a human takes over. A queued follow-up that fires on top of a live conversation undoes the impression the first message created. Use a wait step that ends on a reply event, or a condition that checks for an inbound message before the next send.
Repeat callers and re-entry
The same person will call twice in an hour. If re-entry is off, the second missed call produces nothing and the caller meets silence on the attempt where they were most motivated. If re-entry is unrestricted, someone who rings four times gets four identical texts.
The workable middle is to allow re-entry with a cooldown, so a genuine second attempt on a different day is answered while four calls in five minutes are treated as one. The behaviour of that setting is not obvious and the re-entry rules that decide whether a workflow refires are worth understanding before you rely on them in a client account.
Reading the results honestly
Two numbers describe whether this is working, and neither of them is the count of texts sent. The first is the reply rate: of the callers who received a message, how many wrote back. The second is what happened to those threads — booked, answered, or abandoned in the inbox. A high send count with a low reply rate usually points at the sending number or the wording; a good reply rate with abandoned threads points at ownership, not at the automation.
Watch the false-positive count as well. Ask the client’s team, once a week for the first month, whether anyone received a missed-call text after a call they actually took. That question surfaces a broken status filter faster than any report, and it costs nothing to ask.
Built this way the automation does something narrow and useful: it converts a hang-up into a thread. Everything else hangs off that thread existing, which is why the filtering and the sending number deserve more attention than the copy.
Frequently Asked Questions
What is missed call text back?
It is an automation that sends an SMS to a caller when their inbound call ends without being answered. The text comes from the number they dialled, acknowledges the call and asks a question, so the conversation continues in a thread that can be assigned and answered rather than disappearing into a call log.
Will it text people whose calls were answered?
Only if the filter is wrong. The call trigger fires for answered calls as well, so the workflow needs a condition on call status and direction. Restrict it to inbound calls with a no-answer, busy or failed status, and optionally treat an answered call under about ten seconds as a drop.
Which number should the reply be sent from?
The number the caller dialled. Sending from a different number makes the message look unsolicited and puts the caller’s reply on a line nobody is monitoring. Check that the receiving number is provisioned for SMS as well as voice, because a voice-only number will fail silently in the workflow log.
Do I need carrier registration before turning this on?
For local US numbers, yes. Application-to-person traffic on unregistered numbers tends to be filtered rather than rejected, so the workflow reports a successful send while nothing arrives. Complete brand and campaign registration for each sub-account that will send, and allow time for the review.
How quickly should the text go out?
Close to immediately. The value of the automation comes from reaching the caller while they still have the phone in their hand. A one or two minute delay is defensible if staff often call back manually and you want to avoid crossing wires, but a longer delay removes the reason the message works.
What stops a repeat caller getting the same text four times?
The re-entry setting, combined with a cooldown. Allowing re-entry means a genuine call on another day is answered; a cooldown window means several attempts in a few minutes are treated as one event. Turning re-entry off entirely leaves the most motivated callers with silence.
How do I know whether it is working?
Track the reply rate rather than the send count, then track what happened to the threads that replied. Also ask the client’s team each week whether anyone was texted after a call they took: that question finds a broken status filter faster than any report.
We build your GoHighLevel account for you
CRM, pipelines, funnels, SMS and automation — configured for your business, not left for you to figure out.
- Setup (one-time)$850
- Ongoing (per month)$38.80
No obligation quote · reply within 1 business day
This site is independently operated and is not officially affiliated with, endorsed by, or sponsored by HighLevel LLC. We do not receive referral commissions — this is our own flat-fee implementation service.