Most dental ad reports lead with one number: blended cost per lead, or blended cost per booking. It's clean, it fits in a headline stat, and it's the wrong number to build a report around. A dental practice doesn't run one type of ad. It runs implant campaigns, Invisalign campaigns, cosmetic campaigns, and general checkup campaigns, often in the same account, sometimes in the same ad group. Those procedures don't cost the same to acquire, and they're not worth the same when they book. Averaging them together produces a figure that's technically correct and practically useless.
On our dental agencies page we talk about reporting in cost-per-booking language instead of ROAS, because that's the number a dentist actually reads. But cost per booking still has to be split by what was booked. A practice that books 20 checkups at $50 each and 2 implant consults at $110 each does not have a "$54 average cost per booking" problem or win. It has two separate stories, and a client who can't see both is going to ask why the number moved without understanding why.
What the industry-wide numbers actually say
Start with the baseline. WordStream and LocaliQ's 2026 Google Ads Benchmarks report, which analyzed more than 13,000 search campaigns across 23 industries between April 2025 and March 2026, puts dental at an $8.00 average cost per click, a 5.66% click-through rate, a 10.67% conversion rate, and a $72.97 cost per lead, blending Google and Microsoft Ads data. That's the number most agencies would hand a client as "the" dental benchmark. It's real, and it's also an average of procedures with wildly different economics sitting underneath it.
the blended average cost per lead across all dental search campaigns nationally, per WordStream/LocaliQ's 2026 Google Ads Benchmarks. It's a useful headline. It's not a useful target CPA for any single campaign.
Same practice, four different economics
A June 2026 analysis by Delmain, built on roughly 180 of its own U.S. dental Google Ads campaigns over a rolling 120-day window, breaks cost per lead out by treatment type. General dentistry campaigns averaged a $50.60 cost per lead. Emergency campaigns came in at $77.62. Cosmetic campaigns landed at $119.62. Implant campaigns averaged $107.63. Invisalign and orthodontics campaigns were the most expensive at $212.63 per lead, driven by a lower 3.5% conversion rate against a still-modest $4.57 cost per click. That's a more than 4x spread between the cheapest and most expensive procedure type, inside the same specialty.
the spread between general dentistry cost per lead ($50.60) and Invisalign cost per lead ($212.63) in Delmain's 2026 dental patient acquisition data. A blended report can't show a client both numbers at once, and shouldn't try to.
Think about what happens when those campaigns get averaged into one weekly figure. If a practice runs mostly general dentistry with a small Invisalign push, the blended CPA looks fine, and the Invisalign campaign, which might be badly targeted or bidding into keywords it can't win, gets hidden inside a number that still clears target. Flip it around: a practice running a heavy implant push during a slow general-dentistry month will show a spiking blended CPA that looks like a crisis, when the implant campaign is actually performing exactly as expected for a $107 procedure lead with a five-figure lifetime value attached to it.
Why a single target CPA breaks the report
Most reporting tools, including the earliest version of what we built, ask an agency to set one target cost per booking per client and measure everything against it. That works fine for a single-service business. It doesn't work for a dental practice running four procedure lines with four different acquisition costs and four different lifetime values. Set the target CPA at the general-dentistry number and every implant or Invisalign campaign will look like it's failing, every single week, forever. Set it at the implant number and every general campaign will look artificially cheap, masking real waste.
Either way, the client stops trusting the number. And once a client stops trusting the number in a report, they stop reading the report at all, which is a big part of why PPC agencies lose so many clients over reporting, not performance.
There's a second layer to this that a blended CPA hides even more completely: lifetime value. A general checkup patient might be worth a few hundred dollars a year in recurring cleanings. An implant patient can be worth tens of thousands once the full treatment plan is factored in, and an Invisalign case often runs several thousand dollars on its own. So a $212.63 Invisalign lead isn't actually expensive relative to what it returns, it just looks expensive sitting next to a $50.60 general dentistry lead in the same average. A report that shows procedure-level CPA without procedure-level value attached still leaves the client guessing at whether a number is good or bad.
Reporting that already separates the campaigns you run
NarrateIQ writes a plain-English report every Monday that breaks results down by campaign, not just by blended average, so an implant push and a checkup campaign never get flattened into one misleading number.
Book a free audit call →What a dental report should actually show
A useful weekly report for a dental client shows cost per booking broken out by procedure or campaign, next to that procedure's own target, not a house-wide average. It flags which line is above target and which is below, in dollars, not percentages. And it ties the number back to what the practice actually cares about: a $107 implant lead that converts is worth far more than a $50 general lead, so the report should say that plainly instead of forcing the client to do that math themselves.
This is the same logic behind cost-per-booking reporting generally: dentists don't think in ROAS, they think in chairs filled and cases signed. Splitting that by procedure is just taking the same idea one level deeper, because a chair filled with a checkup and a chair filled with an implant consult are not the same event, financially or clinically. An agency that reports the difference looks like it understands the practice. One that hides it behind a blended average looks like it's guessing.
None of this requires a bigger reporting effort, just a different structure. The same weekly pull from Google Ads and Meta already has the campaign-level data broken out by procedure, most agencies just roll it up into one number before it reaches the client. Unrolling it is mostly a formatting decision, not a data problem, and it's the difference between a report a dentist skims once and one they actually use to decide where next month's budget goes.
If you manage dental clients and you're still sending one cost-per-booking number a week, it's worth checking whether that number is actually three or four numbers wearing a trench coat. It usually is.