HVAC · PLUMBING · ELECTRICAL

Cost per booked job, cut from $332 to $143.

How a Dallas-Fort Worth multi-trade operator replaced cost-per-lead with cost-per-booked-job — and found the revenue its agency couldn't see.

$332 → $143
Cost per booked job
5.4×
ROAS
$25K+
Recovered abandoned bookings
90 days
To full impact

Live client data, Copeland Home Services, prior-6-month and YoY benchmarks.

The Problem

The gap between what they were paying for and what they were getting.

Copeland was running strong ad spend across Google and Meta, but had no way to connect clicks to booked jobs. Their agency reported cost-per-lead. Their ServiceTitan schedule told a different story.

The gap between what they were paying for leads and what those leads actually booked was significant and invisible. Without attribution, every budget decision was a guess.

What We Put In Place

Deployed in under 2 weeks. Zero disruption.

Peach deployed in under 2 weeks. DNI replaced all static tracking numbers. The booking form went live on their service pages.

ServiceTitan integration created a live feedback loop: every booked job traced back to the originating ad click. For the first time, Copeland could see exactly which campaigns, keywords, and creatives were producing revenue.

1
Week 1

DNI + ServiceTitan Integration

2
Week 2

Booking Form Live

3
Week 3

First Attribution Report

The Result

The numbers in the first 90 days.

$332 → $143
Cost Per Booked Job
57% reduction
5.4x
Return on Ad Spend
vs prior 6-month baseline
65%
Online Booking Conversion
vs 10-20% industry average
$25,000+
Abandoned Revenue Recovered
in first 90 days

Based on live client data. Measured against prior-6-month baseline and YoY benchmarks.

In Their Words
"Peach showed us the number we didn't know we were missing. Our agency was optimizing for leads. Peach showed us we were leaving booked jobs on the table every single week."
B
Brandon
CEO · Copeland Home Services · Dallas-Fort Worth

Ready to see your numbers?

Book a 30-minute session. We'll show you what your attribution chain actually looks like.

Book a demo