BioFuse runs IV therapy, weight loss and TRT clinics across four markets in Michigan and Idaho. Their business runs on appointments booked through their website, which is exactly what the previous agency never bothered to measure. They spent about $75,000 in twelve months and reported page views as results.
BioFuse's business runs on appointments booked through their website. For ten of the previous agency's final twelve months, the account had no tracking on the booking flow at all. About $75,000 was spent in a year, and the monthly reports counted page views as conversions.
We rebuilt the account from twelve overlapping campaigns to eight with one purpose each, fenced the spend in with negative keyword lists, put end-to-end tracking on the booking flow, and separated the four markets so each clinic's spend is reported on its own.
In the ten weeks the previous agency's tracking existed, it recorded 21 bookings against about $17,500 of spend, $846 for every appointment. In our first 103 days the same website recorded 568 bookings at $42 each.
Bookings recorded in our first 103 days
About 5.5 a day
Cost per booked appointment
95% lower
of the top search spend we inherited was buying nothing
About $7,700 of the top $11,600 went to terms with zero bookings or leads. Now fenced off with negative keyword lists.
Cost per search click
51% lower
The client's account data, shared with their agreement. Baseline, timeframe and method are stated below.
URLs across four sitemaps
treatment pages, priced
clinics, two states
Method: counted from biofuse.net and its sitemap at capture, 10 Sep 2026. Build evidence, not a performance claim.
The previous agency let broad match run unattended. Single clicks the client paid for:
| Search term | Cost | Clicks | Leads |
|---|---|---|---|
| “vitamin infusion” | $156 | 1 | 0 |
| “iv therapy that takes insurance” | $98 | 1 | 0 |
| “vivid iv health grand rapids” (a competitor’s name) | $95 | 1 | 0 |
| “iron drip near me” | $92 | 1 | 0 |
| “do b12 shots work for energy” | $67 | 1 | 0 |
$156. One click. No lead.
“Nugenix” is an over-the-counter testosterone booster sold in drugstores, not a service BioFuse offers, and not a customer looking for a clinic. Across three variations of the term the account spent about $344 for 7 clicks and zero leads.
The Traverse City campaign bought the term “game day traverse city”, a local sports bar, for about $293 across 9 clicks. It produced one recorded conversion, which under their tracking setup could have been a page view.
The account had duplicate, overlapping campaigns for the same services in the same cities. The term “iv therapy near me” was bought by seven different campaign and ad group combinations at the same time, costing about $1,150 for 77 clicks, roughly $15 per click, in a market where we now buy clicks at $3.11. Their Demand Gen campaign spent about $1,100 and recorded zero conversions of any kind.
This is the one that mattered most, and it is not a subtlety; it is the basic job. BioFuse’s business runs on appointments booked through their website. For ten of their final twelve months, the account had no tracking on the booking flow at all. Not one completed booking was recorded. Every optimization decision made on that account for the better part of a year was made blind to the only outcome that pays the bills. What the monthly reports counted as “conversions” instead: 2,517 “Start Booking, page view” (someone landing on a page), 1,408 “Local actions, website visits” and 1,217 “Local actions, other engagements”. In October 2025 they promoted the page view to a primary conversion, which is what Google’s bidding then chased, so the budget was being steered toward people who opened a page and left. That month the account reported 323 conversions. 305 of them were page views. The reported cost per lead looked like $24. It was not a real number, and nothing in the report the client was sent would have told them that.
They optimized an account for a business that books appointments online without ever measuring a booking.
Booking tracking existed for exactly ten weeks of the previous agency’s tenure, because they didn’t put any in place until April 2026. Here is everything their tracking ever recorded, against our first 103 days:
| Previous agency (Apr 1 – Jun 10 2026, their only tracked period) | Lumin (Jun 11 – Sep 22 2026) | |
|---|---|---|
| Bookings recorded | 21 | 568 |
| Spend | $17,500 | $24,100 |
| Cost per booking | $846 | $42 |
| Bookings per day | 0.3 | 5.5 |
| Bookings per $1,000 spent | 1.2 | 23.6 |
| Previous agency (12 mo to Jun 10 2026) | Lumin (Jun 11 – Sep 22 2026) | |
|---|---|---|
| Spend | $75,300 | $24,100 |
| Cost per search click | $6.40 | $3.11 |
| Search clicks per day | 31 | 69 |
| Search spend per day | $199 | $216 |
If you open this account today, Google will tell you it produced 1,726 conversions in July. It didn’t. The booking flow fires a conversion at each of five steps, so one person booking an appointment can register five times. We report the fifth one, the completed booking, and we’re fixing the counter. An agency that wanted a better-looking case study would have used the bigger number.
Methodology. All figures come from BioFuse’s Google Ads account and can be verified in it; dollar figures are rounded. Cost per click and clicks per day cover Search campaigns only and are measured identically in both periods. “Bookings” counts completed appointments booked on the BioFuse website, measured by the same tracking in both periods. The previous agency put no booking tracking in place until April 2026, so their figures cover April 1 to June 10 2026, the only period in their tenure where bookings were recorded at all.
Seventy-five treatment pages, each with its price published rather than gated behind an enquiry
Separate in-clinic and mobile booking paths on the same treatment page
Seven clinician bios, so a prospective patient can see who administers the treatment
A hundred-post library answering the questions people ask before booking a first drip
Click-to-call in eight places on the homepage, for the patients who would rather phone
Website, local and AI search, ad account, or all three. Read-only access, real findings on your own accounts, and a person who walks you through them. You keep the findings either way.
