Case Study

Central Dry-Cleaning - An AI Voice Agent That Books the Morning Pick-Ups

Company

Central Dry-Cleaning

Industry

Dry cleaning and laundry

Location

Queenstown, serving Queenstown and Wanaka

What We Built

• An AI voice agent, Sophie, answering the shop's enquiry line
• Hotel and business pick-ups captured on the call
• Pick-ups written to Google Calendar and texted to the driver
• Callback tasks in place of live transfers
• A clearly flagged summary email for every call

Client Overview

Central Dry-Cleaning has been cleaning in Queenstown for more than 20 years. It looks after residential customers, hotels (staff uniforms and guest valet), restaurants and government agencies across Queenstown and Wanaka. Much of the work is turned around the same day.

The Challenge

The phone is busy, and the team wanted it off their plate so they could run the plant. Hotel concierges ring most mornings to book a collection, and turning that call into a driver run meant someone at the desk reading the details and texting a driver by hand.

What the team wanted:

Enquiries answered properly: services, prices, hours and holiday closures, from approved information
Pick-ups captured once: the hotel, guest valet or staff uniform, the items and the day, ready for the driver
No incoming calls to juggle: anything that needs a person comes back as a callback, not a transfer
Every call visible while training: so the owner could check what the agent said and correct it

The Solution

The first stage was enquiry handling, pick-up capture and message taking. Calendar entries for pick-ups and texts to the drivers came next, once the basics were working.

1. Enquiries from an approved knowledge base

Sophie answers from a knowledge base built from the shop's website and the owner's onboarding answers. It is updated as calls are reviewed. When a call showed she was turning down suit pressing, a service the shop does offer, the press-only service was added. She also mentions an upcoming listed holiday closure on any call about timing, and points callers outside the area to recommended cleaners in their own town.

2. Pick-ups go to the calendar and the driver

When a hotel books a collection, the summary email leads with "PICK-UP REQUEST" and the hotel name, carries a banner with the details laid out for the driver, and is marked high importance. The pick-up is also written to the team's Google Calendar as part of the morning or afternoon run, and the driver gets a text that says who, when and what in the first line.

3. "Tomorrow" means tomorrow

Early on, a pick-up asked for "tomorrow" was logged for the day of the call. The fix was to stop asking the AI to do date arithmetic. Sophie's analysis now records the word the caller used, and code turns it into a date against New Zealand time. The calendar, the email and the driver's text all use that same date, so they cannot disagree, and a pick-up is never logged on a weekend.

4. Callbacks instead of transfers

At the owner's request, Sophie never puts a caller through. When she cannot resolve something she takes the details, tells the caller the team will come back to them, and the call becomes a callback task on the team's calendar with the transcript attached. It is due when someone is in the shop to make the call: straight away during opening hours, otherwise first thing the next working day.

Day to day

The email always comes first

The summary email is the record the team works from, so it is sent before anything else. The calendar entry, the driver's text and the callback task come after it. If one of those fails, the team still has the email and falls back to the way they worked before, rather than losing the pick-up.

While Sophie is being trained, every call is emailed, including the plain information calls, so the owner can review them all. A filter that holds back information-only calls is already built and can be switched on once the shop is happy with her answers.

A hotel pick-up call

Example flow

Concierge rings the shop
Sophie takes the hotel, the items and the day
Calendar entry created
Placed on the right day's run in Google Calendar
Driver texted
Who, when and what, readable from the lock screen

No copying out by hand

What Changed

The shop phone no longer pulls people away from the plant. Routine questions get answered, and a morning pick-up now reaches the calendar and the driver's phone without anyone retyping it.

Pick-ups reach the driver directly - calendar entry and text, from the call itself
No transfers to juggle - unresolved calls become callback tasks the team works through
One date everywhere - the email, calendar and text always agree on the pick-up day
The agent keeps improving - call reviews feed straight back into what Sophie knows

Key Takeaways

Central Dry-Cleaning is a useful example for any service business that books collections or visits over the phone:

Automate the step after the call: the value was in the calendar entry and the driver text, not just the answer
Let code do the dates: record what the caller said and calculate the day reliably
A callback can beat a transfer: the team calls back when they are ready, with the transcript in front of them
Start with every call visible: filter the inbox later, once the agent has earned it

AI Voice Agents

Sophie answers the phone and books the pick-ups. See how an AI voice agent can answer your calls and pass the work straight into the tools your team already uses.

Explore AI Voice Agents

Is the phone pulling your team off the job?

Let's map what an AI voice agent would answer for your business, and where the bookings it takes should land.

Explore AI voice agents

Free 30-min call · We'll show you 2–3 things to automate first · No obligation