Case studies / BookingOps
AI Operations PlatformOur product · Field servicesNov 2025 – Aug 2026

BookingOps

One platform to run a service business: scheduling, dispatch, payments, and AI agents that answer the phone and analyse the numbers.

Built for
Cleaning, HVAC, snow removal, dog walking and other field-service businesses with multiple locations
Platform
Admin, provider and customer web apps, provider and customer mobile apps
Status
Live
The BookingOps landing page: one platform for scheduling, dispatch, payments and AI for service businesses.
5apps: admin, provider and customer, web and mobile
725API endpoints behind the platform
4AI voice agents by role
5,000+commits from a 10-person team
Overview

The short version.

BookingOps runs everything a field-service business does in a day. Owners and office staff work in the admin app, cleaners and technicians use the provider app on their phones, and customers book and pay in the customer app. One organisation can run many locations and switch between them from a single dropdown.

On top of the operations core sits an AI layer: voice agents that answer calls and book jobs, a setup assistant that builds the service catalogue through a conversation, an AI Analyst that answers business questions in plain English, and an inbox reader that turns lead emails into quotes.

The challenge

What was broken.

Service businesses run on a tangle of tools: a calendar for jobs, a spreadsheet for providers, a payments app, a separate CRM and a phone that rings while everyone is out on site. Owners couldn't see which customers were drifting away, providers got job details by text message, and every new location multiplied the mess.

  • Jobs, providers, payments and customers lived in separate tools.
  • Calls went unanswered while the whole team was out on jobs.
  • Owners had no quick answer to "how did last month go?" without an export and a spreadsheet.
How it came together

From idea to production.

  1. Nov 2025

    Operations core

    Monorepo, backend and admin app, starting from cleaning operations: services, bookings, providers and schedules.

  2. Dec 2025 – Jan 2026

    Three sides, five apps

    Provider and customer web apps and the two mobile apps, with payments, payouts, checklists and push notifications.

  3. Feb – Apr 2026

    Growth tools

    Leads, quotes, pricing engine, inventory, payroll, invoicing, subscriptions, coupons, campaigns and multi-location support.

  4. May – Aug 2026

    AI layer

    Voice agents, the conversational setup assistant, the AI Analyst and email lead parsing.

The solution

What we built.

BookingOps gives each group its own purpose-built app over one shared system. The admin sees a live schedule timeline for every provider, customers grouped by churn risk, and the full money trail from quote to invoice to payout. Providers get their day's jobs with room-by-room checklists, before and after photos, maps and navigation. Customers browse services, book a slot and pay, even without an account. When the phone rings, an AI voice agent looks up the customer, checks availability and books the job.

How it works
  1. Step 1Customer calls, books online or emails
  2. Step 2AI agent or customer app creates the booking
  3. Step 3Job dispatched to the right provider
  4. Step 4Provider completes the checklist with photos
  5. Step 5Invoice, payment and payout close the loop
Key features

What we delivered.

Live schedule timeline

Every provider's jobs for the day on one timeline, across locations.

AI voice agents

Operations, sales, quality and customer-service agents answer calls, look up customers, check slots and create bookings.

AI Analyst

Ask "which services grew last quarter?" in plain English and get the answer from the business's own data.

Conversational setup

A setup assistant interviews the owner and builds services, add-ons, checklists and specialisations as it goes.

Provider mobile app

Job checklists, before and after photos, GPS, push notifications and biometric login.

Customers by churn risk

Customers grouped into champions, at-risk and churned so the team knows who to call.

Leads to quotes

Lead forms, inbox parsing of lead emails, quotes and a pricing engine.

Money end to end

Invoicing, subscriptions, coupons, payroll and provider payouts through Stripe Connect.

Inside the product

A closer look.

Where the AI does the work

Intelligent, and trustworthy.

Two paths for every question

An AI router decides whether the owner is setting something up or asking about the numbers, and sends the message to the right specialist agent.

Questions to answers, safely

The AI Analyst writes database queries from plain English, with guardrails that allow reading only.

Chosen for speed

The assistant's model was picked by measurement: about 2 seconds per answer against over 25 for the alternative tested.

Voice that takes action

The phone agents don't just talk. They look up the caller, check real availability and create the booking.

Built with

The stack.

Web apps
Next.js admin, provider and customer apps
Mobile
React Native (Expo) provider and customer apps
Backend
NestJS API on Supabase Postgres, Redis, RabbitMQ, live updates
AI
LangGraph assistant, MCP tools, Pipecat voice with Deepgram, Cartesia and GPT-4o
Integrations
Stripe and Stripe Connect, Twilio voice, Google Maps route optimisation, Firebase push, Gmail lead intake
Results

The numbers.

Build proof

5
apps across web and mobile
Verified
725
API endpoints
Verified
74
screens in the admin app alone
Verified
~2s
AI assistant answer time
Verified

Expected impact for users

0
double bookings across locations
Projected
35%
more jobs per provider per week with route planning
Projected
90%
after-hours calls answered and booked
Projected
5 → 1
tools replaced for a multi-location team
Projected

Verified from project repositories, test runs and release records. Projected modelled estimates of user impact, not measured client results.

Your product next

Got an idea? Let's ship it.

Tell us the workflow that's costing you customers or hours. We'll come back with a prototype plan, a timeline in weeks, and a fixed quote.