← All case studies
Google Ads Conversion Tracking GA4 & GTM Call Tracking Multi-Location

$467,500 in ad spend. $4.68 Million in booked work.

Twenty-plus territories across nine states. Thirteen accounts with full reporting access, measured.

$467,500Spent on Google Ads
7,800Quote requests
$4.68MBooked work
$10.00Back per $1 spent

Ad spend and quote requests are measured figures from the client's own Google Ads accounts. Booked work is modeled on published 2026 industry benchmarks, as set out below.

01 The Hook

Twenty-plus franchise territories were about to spend more than half a million dollars on Google Ads. Not one of them could measure a lead.

New territory, new Google Ads account, new website, and nothing underneath it. No container, no Conversion actions, no record of which ads produced a phone call and which produced nothing at all.

We built the measurement layer first. Everything below happened because of it.

02 The Outcome

What $467,500 actually bought.

A lead is not the point. Booked work is the point.

$467,500 Spent on Google Ads
7,800 Homeowners who asked for a quote
3,900 In-home appointments booked
1,170 Jobs sold and installed
$4.68 Million Booked insulation work
$400 to win a job worth $4,000. The job pays for its own marketing ten times over.
$10.00 back for every $1.00 spent.
$2.1 Million in gross profit at a 45% margin.

Per territory, per month: $1,493 spent, 25 quote requests, 3.7 jobs sold, about $14,950 in booked work.

Greater Charlotte Campaign Overview upwork
Campaign performance across the measured territories.
Greater Charlotte Conversions upwork
One Conversion taxonomy, repeated in every account.
03 The Method

Where those numbers come from.

The spend and the 7,800 quote requests are measured facts pulled from the client's own Google Ads accounts. Everything after that is modeled on published 2026 insulation industry benchmarks, using the middle of every range.

What we assumedWe usedPublished industry range
Quote requests that become booked appointments50%35% to 65%
Appointments that become sold jobs30%20% to 45%
Average insulation job value$4,000$2,000 to $8,000
Gross margin on completed work45%Industry standard

Two measured results worth noting. Published 2026 cost per insulation lead runs $60 to $250. This network delivered leads at $60, the very bottom of that range. Published cost per sold job runs $200 to $900. This network came in at $400.

The result holds whichever way you cut it.

Quote requests that become jobsJobs soldAt $2,500At $4,000At $6,000
10%, below industry norms780$1.9M$3.1M$4.7M
15%, used above1,170$2.9M$4.68M$7.0M
20%, strong operator1,560$3.9M$6.2M$9.4M

There is no bad outcome in that table. The weakest cell returns four dollars for every dollar spent. The strongest returns nearly twenty. We report the middle row because the published midpoint is 16.25%, and we rounded it down to 15%.

04 The Reason

Why this had to come first.

Google Ads bids automatically now. It spends the budget chasing whatever you tell it a customer looks like.

Tell it nothing and it optimizes for clicks, because clicks are all it can see. The longer it runs, the better it gets at buying the wrong thing.

Launch first and fix Tracking later, and you have paid for months of an algorithm learning the wrong lesson. That budget does not come back.

Want to know what your own account is teaching Google?

05 What We Built

One architecture, deployed identically in every territory, before the first ad ran.

BlazeByte deployed one Conversion taxonomy across all twenty-plus franchise territories before any campaign launched, so territory seventeen reports exactly like territory one. The build below covers event naming, phone calls tracked as first class Conversions, and the rules that make cross-location comparison possible.

One Conversion taxonomy across the network. Same events, same names, same primary and secondary rules, so territory seventeen reports exactly like territory one. This is what makes cross-location comparison possible at all.
Phone calls tracked as first class Conversions. Calls from ads, calls from the website, and every tap on a phone number. Insulation sells on the phone. A form-only setup would have thrown away a large share of these leads and the owner would never have known they existed.
1,523 invalid traffic sessions held out of the bidding signal. Bots convert faster than people, so an account that counts them trains itself to buy more of them.
Strict primary and secondary classification. Only genuine lead events reach Smart Bidding. Micro conversions and page views stay tracked for analysis and excluded from bidding, so the budget chases customers instead of activity.
Enhanced Conversions and GA4 to Google Ads linkage configured per account, with Conversion values assigned where the lead economics supported it.
GTM Koala Container scaled
The container that every territory was deployed from.
06 The Engagement

Delivered white label for a US digital advertising agency. They owned the client relationship and ran the campaigns. We built and maintained the measurement layer underneath, across every territory, reporting to them rather than to the franchisees.

Agencies bring us in for exactly this reason. Tracking is the part that breaks quietly, and it is the part that decides whether the campaigns above it can be optimized at all.

07 The Count

Why you can trust the count.

Across two years these accounts accumulated 18,304 total Conversion records, including CRM imports, micro conversions, and invalid traffic flags. We report 7,800. The same comparison, reported against recorded, is what the Tracking Loss Calculator runs on your own two numbers.

Removed before countingRecords
Upstream button clicks (the click that opens the form, not the submission)4,556
Duplicate CRM imports mirroring website events already tracked4,354
Flagged invalid traffic sessions1,523

In one territory, form submissions totaled 165 and the duplicate import totaled 165. In another, 278 and 277.

That gap is the architecture doing its job. Every one of those extra records was classified as secondary or excluded by design, which is precisely why the primary lead count stays clean and why Smart Bidding never learned from it.

The difference it makes: a raw dashboard would report leads at $26. The real figure is $60. Most reporting quotes the larger number.

08 The Benchmark

Nine states, 13 territories, 313 territory-months.

Every territory measured the same way, from the month its Tracking went live through August 2026.

Scroll the table sideways to see every column.

TerritoryLive sinceMonthsAd spendSpend / moQuotesQuotes / moCost each
Westside, CANov 202421$36,000$1,7451,18157.2$30
Coastal Carolina, NCMar 202429$26,200$90767923.5$39
Central Jersey, NJNov 202421$34,800$1,68786241.8$40
Mercer & Middlesex, NJJul 202424$16,700$68332313.2$52
North Eastern Carolina, NCMar 202429$22,400$77539013.5$57
Greater Charlotte, NCAug 202424$47,100$1,99472530.7$65
West Florida, FLNov 202421$74,900$3,6311,08852.7$69
North Central NJ, NJNov 202421$10,300$4991477.1$70
Georgetown, TXNov 202421$46,900$2,22463530.1$74
North Chicagoland, ILJan 202431$60,300$1,93272623.3$83
Midsouth, TNMay 202427$56,100$2,06267124.7$84
Superstition Springs, AZSep 202422$16,000$7121777.9$90
Colorado Springs, COSep 202423$19,800$8701968.6$101
NetworkJan 2024313$467,500$1,4937,80024.9$60

These are floor numbers. Several franchisees paused or closed their accounts inside the measured window, and the months column counts calendar time from first campaign. Dormant periods are averaged in, so every per-month figure understates active-period performance. On active months the network runs closer to 32 quote requests per territory per month.

What this is worth to an owner. A territory in California wins a quote request for $30. One in Colorado pays $101 for the same service, in the same brand, on the same Tracking. Bring Colorado to the network average and the same monthly budget returns roughly 6 more quote requests every month, about 11 more sold jobs a year. No extra spend. Just a fix you can only find when every location is measured identically.

"BlazeByte is truly exceptional. A fantastic partner in the analytics and ads space and a wealth of knowledge."

Robin Nickell, koalainsulation.com/f/central-jersey Central Jersey is one of the 13 territories in the table above.

Every location measured the same way. That is the whole trick, and it is what we would build for you.

West Florida Conversions upwork
The spread that only identical Tracking makes visible.
Westside Conversions upwork
One taxonomy, 13 accounts, 313 territory-months.
09 What It Means

What this means for your business.

You do not need to understand Conversion Tracking. You need to know what it puts in the bank.

You find out what a customer really costs you.Not the flattering dashboard figure. The real one, so you can decide what to scale.
Your budget buys jobs instead of clicks.Feed Google real customers and it finds more of them, on the spend you already have.
Every location becomes comparable.You see who wins work cheaply and who does not, and fix the gap instead of guessing.
It scales.This went into 20+ territories. The next one takes days, and reports identically from week one.
$0 extra added to the ad budget.Every gain came from spending the same money on real customers instead of noise.
The one-line version

Your competitors decide on real numbers. Most ad accounts run on figures inflated more than twice over. We make sure the money you already spend goes to homeowners who are actually there.

Methodology note

Ad spend and the 7,800 de-duplicated quote requests are measured figures from the client's own Google Ads accounts. The architecture was built across 20+ franchise territories; figures cover the 13 accounts with full reporting access, from each account's first campaign date through August 12, 2026. Quote requests exclude upstream button clicks, duplicate CRM imports, and flagged invalid traffic. Appointment rates, close rates, job values, and margin are not client-reported. They are modeled on mid-range published 2026 insulation industry benchmarks and are illustrative, not actual client revenue. Lead totals are conservative.

"
BlazeByte is a dream to work with and such an incredible resource! You should hire them immediately.
Robin Nickell koalainsulation.com/southwest-florida

"Haider is our go to for all things google analytics and tracking. Exceptional work, great attention to detail!"

Robin Nickell

Start with a free audit of your ad account.

No cost, no obligation. Your real cost per lead, your real cost per won job, and what the gap is worth in booked work. Response within 24 to 48 hours.

Prefer written findings first?

Get your free audit →

Or email [email protected]

Scroll to Top