Case Study ยท Optimization Literacy

3 months of silence.
Then $7,600 in 30 days.

Hustle Training 360 had a marketing machine that printed booked assessments for years. Then they rebuilt their website into the most beautiful thing you have ever seen, and it broke everything. This is what happens when you change the winner, and how I brought it back.

$0

Invested

$0

Sales Recovered

0.00x

Return on Ad Spend

0

Days to Turnaround

Google AdsMetaLanding PagesMeasurement
The Client

Hustle Training 360

A Florida in-home personal training company run by founders Sean and Nikki. For years, their marketing was a machine they trusted: put money in one end, get booked assessments out the other.

Industry

In-home personal training, Palm Beach County, Florida

Team

~20 trainers, sent to clients' homes for 1:1 assessments

The winner

For years, one paid-ads funnel quietly printed booked assessments on repeat

Before

They had a winner

For years their paid ads did one job and did it well. Money in, booked in-home assessments out. Predictable, unglamorous, and profitable. It was the machine that quietly carried the whole business, and it was winning.

The Change

Then they built the most beautiful website you have ever seen

Modern, gorgeous, everything the old one was not. There was just one problem with it.

The day it went live, the machine stopped.

Three months, almost no sales. Not because the new site was ugly. It was stunning. It broke everything because it quietly replaced a proven winner with an unproven one, and swept away what little measurement they had in the process. Beautiful is not the same as working.

The lesson

Never
change the winner

When something works, you protect it. You do not redesign it, reinvent it, or improve it because it feels dated. Change the winner, and you lose everything that made it win. That is what optimization literacy actually means: knowing exactly what works, and refusing to break it.

The Blind Spot

And they never saw it coming

The tracking underneath had never been properly deployed. So when the winner broke, nothing sounded the alarm. The numbers looked vaguely fine right up until the bank account said otherwise.

Google's dashboard even read 90%+ healthy the whole time.

A glowing score on an account that was quietly dying. That is the trap: a healthy-looking dashboard is not the same as a healthy business. If you cannot measure what your winner is actually doing, you cannot protect it, and you will not notice the day it stops.

The Diagnosis

I found the winner they had changed

Most people would have treated this as a creative problem or a budget problem, and burned another three months testing ads on top of it. I went looking for what actually changed.

01

Reconstructed what happened

With barely any measurement in place, I rebuilt the signal from UTM parameters and Hotjar session replays to see what the site change had actually done.

02

Pinned it to the redesign

The collapse traced to one specific date. It matched the day the beautiful new site shipped, exactly.

03

Named the winner they had changed

The old funnel was the winner. The gorgeous new site had quietly swapped it out. That was the whole problem, and nobody had connected the two.

The Fix

I rebuilt the winner. This time, measured.

Not a prettier funnel. The proven one, restored, and finally instrumented so it could never break in silence again.

Restored the winner

Rebuilt the proven funnel the redesign had swept away, instead of chasing a prettier new one that had never earned a dollar.

Deployed tracking properly, finally

Installed the pixel and a real four-tier event hierarchy: view, search intent, click-to-call, then lead. So the winner could never go dark unnoticed again.

Killed 70% of wasted spend

Cut the panic budget down to something sane and stopped pouring money into a black hole.

Tamed the $670 target CPA

Ran a deliberate data-gathering phase, then handed the algorithm clean signal to learn from.

Culled 600 keywords to ~20

Kept only the terms that actually drove qualified assessments.

Value-based lookalikes

Built audiences off their real customer list instead of guessing.

The Result

$1,500 in. $7,600 out. 30 days.

The machine came back to life. Within the turnaround window the paid channels started producing booked assessments again, and the first real sales landed within days of the fix.

0 mo

Of Zero Sales Before

$0

Sales Recovered

0.00x

Return on Ad Spend

0

Days to Turnaround

"We're back to getting leads like we used to."

"It's been a good investment."

Sean, co-founder, Hustle Training 360

There is a detail I love about this one. They had been flying blind for so long that when the tracking came back and the sales started landing, they almost did not trust it. That is what three months in the dark does to a business. The winner was always there. It just needed protecting.

How It Went

Five weeks, June to July 2024

  1. Day 1

    Walked into the dark

    The machine was silent and there was almost no measurement to explain why.

  2. Day 6

    Found the change

    Traced the collapse to the exact day the beautiful new site shipped.

  3. Day 9

    Rebuilt the winner

    Restored the proven funnel and, for the first time, deployed tracking properly.

  4. Day ~20

    Sales come back

    The first real ad-sourced closes in months land as the machine wakes up.

  5. Day 30

    $1,500 to $7,600

    Roughly 5x, after three months of nothing.

Do you actually know what your winner is?

Most businesses cannot tell you which piece of their marketing is the one that works. So they redesign it, refresh it, and accidentally kill it. If your numbers fell off a cliff and nobody can say why, that is usually the reason. Let me find your winner and protect it.

Book a Discovery Call

Figures reflect the recovery period, June to July 2024, and are specific to this engagement. Every account is different. Results depend on budget, offer, and market.