Client Case Study

How Join Piper Generated $456K in Revenue on $46.8K in Ad Spend in 30 Days

A GLP-1 weight loss provider had a Google Ads account that looked fine on the surface. Our audit found 6 structural issues that were wasting spend and misreporting results. We fixed all 6 and delivered a 9.74x return inside the first month.

Client: Join Piper (now Polly)
Industry: Weight Loss / GLP-1 Medications
Service: Google Ads
Products: Semaglutide & Tirzepatide
Timeline: 30-Day Engagement
Website: joinpolly.com
Client: Join Piper (now Polly)
Industry: Weight Loss / GLP-1 Medications
Service: Google Ads
Products: Semaglutide & Tirzepatide
Timeline: 30-Day Engagement
Website: joinpolly.com
$456K
30-Day Revenue
9.74x
Return on Ad Spend
864
Total Conversions
6
Critical Issues Fixed
The Challenge

An Account That Looked Healthy but Wasn't

Join Piper sells Semaglutide and Tirzepatide subscriptions, GLP-1 weight loss medications sold through a direct-to-consumer model. With average order values around $500, every conversion is real money. They came to us with a Google Ads account that had been running for months and appeared to be performing fine.

Our audit said otherwise. The account had 6 structural problems that were either wasting spend or misreporting results. Duplicate conversion tracking was the worst of them: a single user action was being counted multiple times, which inflated reported conversion numbers and sent every optimization signal in the wrong direction. You cannot make good decisions on bad data.

Beyond tracking, targeting was limited almost entirely to branded keywords: people who were already searching for Join Piper by name. That is not customer acquisition. That is capturing demand that already existed. There were no in-market audiences, no custom intent targeting, no ad extensions, no negative keyword structure, and the landing page was sending paid traffic to an account creation screen instead of a lead form.

The Strategy

Fix the Foundation. Then Scale.

The rebuild had two phases: fix what was broken, then build what was missing.

Fixing what was broken meant cleaning up duplicate conversion tracking so every optimization decision was based on real data. We restructured ad copy with responsive search ads and dynamic elements, redirected landing page traffic to a proper lead form, and built a 128-keyword negative list to filter out searches that would never convert: job seekers, DIY researchers, competitor brand traffic.

Building what was missing meant expanding keyword targeting from brand-only into high-intent category searches: people actively researching Semaglutide, Tirzepatide, and GLP-1 weight loss options. We layered in in-market audiences, custom intent audiences built from relevant search behaviors, and remarketing to recapture visitors who did not convert on their first visit. Performance Max ran alongside search to extend reach across the full Google network.

Account Audit Conversion Tracking Fix Keyword Expansion 128 Negative Keywords Performance Max Remarketing Responsive Search Ads
Before vs. After

6 Issues Fixed in 30 Days

Issue Before Creekside After Creekside
Conversion tracking Duplicate: single actions counted multiple times Deduplicated, accurate attribution
Keyword targeting Brand keywords only Brand + category + high-intent GLP-1 searches
Negative keywords None 128-keyword exclusion list
Ad copy Static, no dynamic elements Responsive search ads with dynamic insertion
Audience targeting None In-market + custom intent + remarketing layers
Ad extensions Missing entirely Sitelinks, callouts, and structured snippets
Landing page destination Account creation screen Dedicated lead form
Google Ads account metrics, October 6 to November 4, 2025: $46.8K spend, 864 conversions, 974.46% ROAS

Google Ads account metrics, October 6 to November 4, 2025: $46.8K spend, 864 conversions, 974.46% ROAS

$456K
30-Day Attributed Revenue
$46,800 in ad spend at a 9.74x ROAS produced $456,000 in attributed revenue across 864 conversions in the 30 days following the account rebuild.
9.74x
Return on Ad Spend
974.46% actual ROAS over the full 30-day period. The client's stated threshold for scaling was 5x, and we delivered nearly double that within the first month.
864
Total Conversions
864 qualified conversions in 30 days, averaging 29 per day. With accurate tracking in place, every conversion represented a real customer action.
6.14K
Total Clicks
6,140 clicks from expanded keyword targeting, reaching active GLP-1 buyers well beyond the branded searches the account had been limited to.

Maintain over 5x and keep scaling.

— Gabriel Vaturi, Founder, Join Piper
9.74x

$456K in Revenue on $46.8K in Ad Spend in 30 Days

In the 30 days following the rebuild, Join Piper generated $456,000 in attributed revenue on $46,800 in ad spend. That is a 9.74x return, nearly double the client's 5x threshold for scaling. The performance improvement did not come from spending more. It came from fixing the six structural issues that had been wasting spend and misreporting results since the account launched.

The Takeaway

The Audit Was the Product

The first deliverable on this engagement was not a campaign. It was the audit. Every dollar of performance improvement came from identifying and fixing problems that already existed, not from introducing something new.

Duplicate conversion tracking was the most damaging issue because it corrupted every downstream decision. An account that appears to be converting at twice its actual rate will make the wrong optimization choices every single time. You pause the campaigns that look expensive but are actually working. You scale the ones that look efficient but are not. The entire account becomes a machine optimizing toward phantom performance.

Once the foundation was clean, scaling was straightforward. By the end of the first month, the client's goal was simple: stay above 5x ROAS and keep increasing the budget. We were at 9.74x.

Services Used In This Case Study
CREEKSIDE MARKETING
Have a Google Ads account that looks like it's working? Let's find out what's actually happening underneath.