Vision therapy benchmarks barely exist in published data

VT-specific benchmarks are the thinnest-published segment in eye care PPC. The category is emerging in digital marketing data; the public aggregators (LocaliQ, WordStream, Patient10x) report at the optometry or ophthalmology category level, which is too broad to anchor a VT account against. The data that exists is practitioner-level, anecdotal, and not externally validated.

The honest answer for a VT-specific account is that there is no published benchmark to compare against, and any agency citing one is either using a proxy (general optometry, pediatric specialty) without flagging the substitution or reading a number off a deck that someone copied from a competitor’s site without attribution. The right approach is to anchor on adjacent categories with appropriate hedging, healthcare baseline as the floor, ophthalmology category as the ceiling, and live-account validation as the source of truth. This piece walks through what those anchors are, how to interpret them for VT, and how to use them without being misled by them.

What makes a benchmark match the comparable

Benchmarks work when they are matched to the right comparable, and they mislead when they are not. The comparable has at least five dimensions. First, category, healthcare versus the specific eye care subspecialty, optometry versus ophthalmology versus refractive versus VT. Second, sub-specialty, a cataract CPC and an optometry CPC are not interchangeable even though both fall under “eye care.”

Third, account size, a $5K monthly account does not behave like a $50K monthly account, both because Smart Bidding has different signal volume to work with and because the larger account typically has access to more sophisticated structure. Fourth, market density, a New York City CPC is not a rural Nebraska CPC and treating them as the same number produces nonsense in both directions. Fifth, bidding strategy, an account on Manual CPC and an account on Maximize Conversion Value will produce different CPC distributions even with identical other variables.

A LASIK CPC of $45 is fine and within range; an optometry CPC of $45 is broken and indicates either a broad-match leak or a broken Quality Score. The same number means different things in different categories, and an agency that cites a benchmark without specifying the comparable is using the benchmark decoratively, not analytically.

Directional VT benchmarks with appropriate hedging

Directional VT benchmarks, with the caveat that they are anchored on adjacent-category data and should be treated as starting hypotheses, not targets. Specialty-service tCPA sits at $85 to $125 as the floor (Ryze, April 2026) for accounts where the conversion action is qualified lead rather than form submission. Healthcare average CPC sits at $5.64 (LocaliQ 2025) as the broader baseline, with eye care typically at or above this number depending on market density.

Parent-searcher consideration window for VT runs 4 to 12 weeks, which means the attribution window in the conversion action settings must match. An account using a 30-day attribution window for a category with a 60-day average consideration cycle is systematically undercrediting top-of-funnel campaigns and overcrediting brand and retargeting. CVR on matched symptom-stage pages typically falls in the 5 to 12 percent range; CVR on treatment-stage matched pages typically falls in the 15 to 25 percent range, with the upper end requiring credentialing surfaced above the fold and the lower end indicating a generic-template page.

Use these ranges as range anchors, not single targets. A VT account converting at 18 percent on treatment-stage pages is performing well; an account converting at 8 percent on the same pages is leaving meaningful conversions on the table.

The five ways benchmarks mislead

Five ways benchmarks mislead practice owners in QBR conversations. First, the benchmark is cited from 2022 without flagging the age. Healthcare CPCs have risen 40 to 60 percent over three years (Patient10x, August 2025), which means a 2022 benchmark is meaningfully lower than current reality and makes any current account look bad against the stale comparable. Second, the benchmark blends across unrelated healthcare verticals, oncology, dermatology, eye care, primary care, into a single number that means nothing for any specific specialty.

Third, the benchmark is reported as a single number rather than a range, which obscures the actual distribution. CVR for ophthalmology averages 18.29 percent (LocaliQ 2025), but the distribution runs from 8 percent to 30 percent depending on landing page quality, and treating the average as a target either overpromises or underdelivers depending on which side of the average the specific account sits.

Fourth, the benchmark ignores market density. NYC, LA, and Chicago metros run 200 to 400 percent above national averages on CPC (Patient10x, August 2025), and a benchmark that does not adjust for market is functionally useless for an account in any major metro. Fifth, the benchmark mixes brand and non-brand into one CPC figure, which obscures the fact that brand is typically 80 to 90 percent cheaper than non-brand and the blended number masks meaningful non-brand inefficiency.

How to use benchmarks without being misled

Use benchmarks as range anchors, not targets. Pull the LocaliQ healthcare baseline at the start of each quarter as the floor reference. Apply a 2 to 3x multiplier for major urban metros where the account operates. Separate brand and non-brand data before comparing against any benchmark, because the blended figure tells you nothing about either component. Re-check against fresh data quarterly, not annually, because the underlying CPC and CVR distributions shift faster than annual refresh cycles can keep up with.

For each campaign, write down what the comparable is before pulling the benchmark, category, sub-specialty, account size band, market density tier, and bidding strategy. The act of specifying the comparable surfaces most of the cases where the benchmark on hand does not actually match the account, which is the case more often than not. The agency that cannot match a benchmark to a comparable in writing is using benchmarks as decoration; the agency that can is using them as analysis.

Maintain a running benchmark log per account, updated quarterly, with source URLs and publication dates next to each figure. The log forces source discipline, prevents the recycling of stale numbers, and creates a paper trail the practice owner can review without having to take the agency’s interpretation on faith.

Why most cited benchmarks are decorative, not analytical

Our view, most “industry benchmarks” you see quoted are recycled from 2022 without attribution. If your agency shows you a benchmark deck without source URLs and publication dates next to each figure, treat the deck as decorative rather than analytical. The agency is performing rigor without practicing it, and the performance is convincing only to the extent that the practice owner does not check. An attributed benchmark looks specific, LocaliQ 2025 healthcare CPC of $5.64 with the publication date and methodology link, Patient10x August 2025 LASIK CPCs of $20 to $80 with the geographic split disclosed, Ryze April 2026 specialty-service tCPA of $85 to $125 with the service definition specified. VT-specific data is the thinnest segment in the category, low aggregator interest and practitioner-level numbers not yet pooled, so the right move when benchmarks are missing is for the practice to publish its own first-90-days numbers rather than borrow CPCs from cataract or LASIK accounts that share none of the searcher economics. We compile current sources with publication dates in our specialty eye care PPC benchmarks report.