Strong evidence (grade B: first-hand account)Owner-operator who built it describes the setup in detail across three posts and gives specific recovery rates; figures are self-reported.

A Salon Owner Built Booking-Abandonment Recovery With Claude; 17.3% of Lost Bookings Came Back

A three-salon owner built e-commerce-style "abandoned cart" follow-up for half-finished online bookings. With email, 16% came back; with texts added, 17.3%, and tracked recovered bookings passed $10,000.

Who this is about: Three-location hair salon business (Hottie Hair, Las Vegas; co-founder Mike Frehner)

Claude CodeSupabaseVercelGitHubLovable (first prototype)Local & serviceAutopilot

Get the prompt

SalesAdded Oct 16, 2026Source: medium.com

What it takes

Cost: From $20/monthClaude Code: Included with Claude Pro ($20 a month billed monthly), Max, Team and Enterprise plans, but not the Free plan. You can also pay for API usage instead.Built by the owner, a software engineer, during a four-month website rebuild; no costs given.Price checked Oct 24, 2026 on claude.com. Prices change; check before you buy.
Setup time: A few weeksOur estimate from the source, for one business.
Skill needed: Developer skillsYou build or run your own code, even if AI writes it.

Explain it simply

When someone starts booking a hair appointment online but leaves halfway, the salon's system now follows up on its own. It sends an email about the exact service and stylist they picked, then a text, and an offer later if needed. The owner, a software engineer, built it himself with an AI coding tool while rebuilding the salon's website.

New to AI? See how AI works

What was done

He prototyped a new site in Lovable, moved the code to GitHub with a Supabase backend on Vercel, and kept building with the Claude CLI; the site and booking flow were rebuilt between late September 2025 and the end of January 2026.

When a visitor picks a service, stylist and time but doesn't finish, the system saves that partial booking.

If the booking isn't completed within a set window, an email goes out about the exact service and stylist they chose; a text follows after a gap, and an incentive appears on a later touch.

Nobody on staff chases anyone; the booking reaches the salon already confirmed.

How it went

In May he showed his co-founder the data: 16% of abandoned bookings came back through the email flow, 17.3% once texts were added, and tracked recovered bookings had crossed $10,000. Her worry had been that the follow-ups were bugging customers.

The owner is a software engineer of 28 years, so this is a developer build, not a no-code setup.
He doesn't say over what period the $10,000 was recovered or how many bookings were abandoned.
Not promotional: he writes about his own salons and doesn't sell the system.

Prompt

PromptWritten by our editors from the source

Add abandoned-booking recovery to our booking flow ([stack: e.g. Next.js + Supabase]).

1. Save a partial booking as soon as a visitor picks a service, a stylist or a time, with their email or phone if given and consent to contact.
2. If it isn't completed within [2] hours, send an email naming the exact service and stylist with a link that reopens the booking where they left off.
3. If still not booked after [24] hours, send one text; on a final touch after [3] days, include [offer].
4. Stop all messages the moment they book or opt out.
5. Track started, abandoned, recovered and revenue recovered per channel on a dashboard.

Show me the plan and the message copy before writing code.

Setup notes

Only message people who gave contact details and consent, and include an easy opt-out in every email and text.
Measure started versus finished bookings first, so you know how many you're losing before you build.

Business fit

Who this helps, and how:

Salons and spas with online booking: win back clients who start a booking and get pulled away.
Local & service
Clinics, trainers and other appointment businesses: send a reminder about the exact service someone was booking.
Local & service

Similar tools

Other tools for this job, and how they differ. Chosen by our editors, never paid for.

Cursor
AI code editor with agents, for owners who'd rather build the same flow in an editor than the terminal.
Codex
OpenAI's coding agent, working in the terminal, editor or cloud; another way to build a custom booking flow.
n8n
A fixed workflow triggered by an abandoned-booking event, with an AI step to write the message; self-hostable.

“Strong evidence” (grade B) means: The person who did the work describes the setup and results in detail. How we grade. Spotted a mistake? Suggest a correction.