Loading
Loading
A system for gyms, salons, detailers, and similar shops that texts back every missed call, follows up on open quotes until they close, asks satisfied customers for reviews at the right moment, and invites past customers back before they drift away. The owner approves every message before launch.
Solution blueprint: a system Callimachus Forge can architect for your operation. Not a case study, and not a claim of past deployment.
At a glance
Stack candidates
n8n · GoHighLevel · Twilio · Stripe · Calendly · Gmail · Google Sheets · Slack
System run
A simulated run of the exact steps documented on this page, from first trigger to final handoff. Nothing here is live client activity.
Simulation · Loops
The problem
The phone rings while a stylist is mid-cut or a detailer is elbow-deep in a truck, and it goes to voicemail. Most callers do not leave a message. They call the next shop on the list. Quotes go out by text and then sit, because following up feels like nagging and nobody tracks who went quiet. Reviews only get asked for when someone remembers, so the Google profile grows slowly while competitors pull ahead. And the regulars who quietly stopped coming are invisible until the slow month arrives. None of this is a skill problem. It is a follow-through problem, and follow-through is exactly what a busy service business has the least capacity for.
These businesses run lean. The person answering the phone is often the person delivering the service. Leads arrive from Google, Facebook, a website form, and word of mouth, then land in a mix of voicemail, personal text threads, and social inboxes. Booking might live in dedicated software, a paper book, or the owner's head. Margins depend on keeping the schedule full, and the schedule is filled one conversation at a time. Everyone agrees the follow-up work matters, and it is still the first thing dropped on a busy day because it has no deadline attached.
The manual process
The automation opportunity
Every one of these follow-ups is a predictable message triggered by a predictable event: a call that was not answered, a quote that got no reply, a job that was just paid for, a regular who is overdue. A system can watch for those events and send the right message at the right time, written the way the owner would write it. When the customer replies, the system answers simple questions, offers booking times, and hands anything sensitive to a person. The owner reviews the templates once. After that, the follow-through happens every day, whether or not the shop is slammed.
Architecture
Every step is typed: triggers, AI decision points, business logic, actions, data, and the human checkpoints that keep judgment where it belongs.
AI components
Human checkpoints
A missed call, a sent quote, a completed payment, or a web lead form fires a webhook into the orchestration layer. The event carries the customer's phone number, the source, and the event type.
n8n looks the number up in the CRM and pulls name, service history, open quotes, last contact date, and opt-out status. Unknown callers get a new contact record tagged with the lead source.
Rules check opt-out status, quiet hours, and whether the contact is already in an active sequence. The event routes to one track: missed-call text back, quote follow-up, review request, or win-back.
A language model drafts a short message in the shop's voice using the pulled history. It names the service, references the open quote, or thanks the customer by name. Templates set the tone, the model fills specifics.
The message goes out as a text from the business number, with email as fallback. Missed-call texts send immediately. Follow-ups and win-backs send on the schedule the owner approved.
When the customer replies, a language model classifies the intent: book, question, not now, or unhappy. Booking intents get open times or a booking link. Routine questions get answers from an approved FAQ.
Price negotiations, complaints, and anything the model cannot classify with confidence route to the owner's phone or Slack with the full thread attached. The sequence pauses until a person responds.
Every message, reply, booking, review click, and escalation is written back to the CRM contact and a running log. The nightly sweep and the weekly digest both read from this log.
Each night a job scans the log for quotes with no reply, customers past their usual rebooking interval, and stalled threads. Qualifying contacts feed back into the router for the next touch.
Once a week the owner gets a plain-language summary: new leads answered, quotes still open, reviews requested, customers reactivated, and conversations waiting on a human.
Engineering notes
Benefits are qualitative by design. We do not publish numbers we cannot verify.
The free audit maps your actual tools and process against this architecture, and tells you honestly what is worth building.