Case study · Coded Level Bridge
A six-clinic dental group finds what its ads really bring in
Lakeshore Dental Group spent more on ads every quarter. The agency report said it was working. A week before the partners' meeting, nobody could say what a new patient cost.
- Industry
- Dental clinic group
- Locations
- 6, Ontario and Quebec
- Ad spend
- About $490K a year
- Connected
- Google Ads, Meta, Analytics, call tracking, 2 booking systems
The challenge
Three reports, three answers
Conversions up from 721 to 1,645 a month. Cost per lead down from $43 to $36.
Half these "leads" are existing patients calling to reschedule.
New patients up a little, while ad spend went from $31K to $58K a month.
The partners were about to cut Meta and decide on a seventh clinic. The first job was one number everyone could trust.
Cost per lead fell while the real cost of a new patient rose
View data
| Month | Cost per lead, reported | Cost per new patient |
|---|---|---|
| Nov | $43 | $81.4 |
| Dec | $38.6 | $85.4 |
| Jan | $33.1 | $68.6 |
| Feb | $39.4 | $82.9 |
| Mar | $28.6 | $100 |
| Apr | $22.9 | $93 |
| May | $26.7 | $96.9 |
| Jun | $29.4 | $98 |
| Jul | $33.5 | $110 |
| Aug | $35.4 | $123 |
| Sep | $32.6 | $89.9 |
What Bridge found
Most "conversions" weren't patients
Bridge connected every system and rebuilt the year: each new patient, and where they came from. Google's count fell apart first.
Where Google's 14,752 conversions came from
- Form tag fired on page load4,741
- "Book now" clicks counted as bookings3,713
- Calls from existing patients2,551
- Same booking counted twice587
- Overlap with Meta, no-shows, timing1,514
- New patients who came in1,646
Urgent
- Treatment details sent to Meta in page addresses
- Google's patient upload silent for 138 days
- Meta stopped receiving bookings on May 12
Counted wrong
- Form tag fired on page load
- "Book now" clicks counted as bookings
- 62% of ad calls were existing patients
Not connected
- 84% of online bookings lost the ad click
- 606 bookings from ad calls never joined
- Milton had ads but no tracking at all
The surprises
Meta was weaker, not dead
358 results for the whole year.
588 new patients who came in from Meta.
The $99 offer: of 138 patients booked, 57 came in.
At about $261 a patient against $203 for Google, Meta was the weaker channel. Cutting it would have removed nearly 600 patients a year.
The outcome
One number per clinic
The decisions, before and after
- Cut Meta: "it isn't converting"
- Judge ads by a $36 cost per lead
- Decide on a seventh clinic with no basis
- Keep Meta for new patients, drop its $99 offer
- Judge ads by cost per new patient, by clinic
- Compare any new site against real numbers
Check
Every number traced to a real patient, every fault found.
Fix
Tracking repaired, and patients who came in sent back to Google and Meta.
Run
Daily checks across every system, and a one-page summary for the partners.
About one in five new patients still can't be traced to a source. Bridge reports them as unknown instead of guessing.
Get the same answer for your business.
Bridge traces every number to a real booking or sale, whatever you sell.
An example based on problems clinic groups describe.