Can Your Clinic Prove Its Marketing Works? We Checked 559 Vet Websites (2026)

The roster is the same 559-clinic screening list behind our advertising study (metro-skewed, about one in seven of the roughly 3,800 practices in Australia). The probe read each clinic’s public homepage code plus, where a Tag Manager container was present, the container’s own public configuration. Twenty-two sites returned no readable homepage and are excluded from every figure rather than counted as tag-free.

Measurement follows marketing intent. Clinics we found advertising on Google in June 2026 almost always have analytics (224 of 243, 92%). Clinics we did not find advertising mostly do not: 45% of their websites carry no Google analytics at all. Once the no-website clinics are included, 52% of the clinics we did not find advertising have no way to measure anything online. The blind spot is concentrated almost entirely in the clinics we did not find advertising on Google. The average clinic deciding whether to start advertising is therefore deciding with no baseline of what its website already does.

Universal Analytics was Google’s standard web analytics tool for about a decade, and Google retired it in July 2023: standard properties stopped recording on 1 July 2023. A UA tag left in the page does nothing except suggest nobody has looked at the site’s code since. The 38 sites carrying UA alongside a GA4 tag have a maintenance job. The 50 sites carrying UA alone have something worse: an owner who may genuinely believe the clinic has analytics, while every visit, call and booking click since July 2023 has gone unrecorded.

This is the inversion at the centre of this report. Most advertisers install analytics, but six in ten have not installed the tag that connects a click to a booking. Without an Ads tag on the site, Google’s system cannot see that a click became a booking-form submission, and the clinic cannot see which campaign, keyword or suburb produced a customer rather than a visitor. It also usually means no remarketing audiences, so a pet owner who visited once is gone.

The honest caveats, because an on-site tag is not the only path: conversions can also be measured through Google’s call-forwarding numbers, imported from GA4, or uploaded offline, and our probe reads homepages rather than every inner page. So 63% overstates total conversion blindness. The floor is the group with neither an Ads tag nor GA4 anywhere: 20 advertisers (8%) with no possible website-side conversion path at all. Between the floor and the 63% sits a large group whose best case is indirect measurement they would need GA4 discipline to use, and 30 advertisers (12%) do not have GA4 either.

One comparison makes the misdirected effort vivid: 21% of clinic websites carry a Meta pixel, 20% a Google Ads tag. As many clinics have tagged their site for Facebook as for the channel where a pet owner types “vet near me” with a sick dog on their lap.

The pattern is easy to create: a site builder or plugin inserts GA4 directly, then an agency adds Tag Manager (or the reverse), and nobody removes the older install. Both tags can fire on every page, so traffic can inflate, engagement metrics can distort, and any conversion counted through the page can register twice. Of the sites with GA4, 84 run it hardcoded only, 154 through Tag Manager only, and 66 have both delivery routes present, 60 of them with the identical ID visibly duplicated.

WordPress powers 76% of the sites where we could identify a platform (303 of 396), so most double-installs sit on WordPress sites. The rate was no lower on the other platforms we could identify: 20% of WordPress GA4 sites, 20% elsewhere, on small numbers. The layering comes from how sites are managed rather than from the platform.

Vet clinics are phone businesses. Many bookings still start with a call, and 84% of these websites carry a tap-to-call link inviting exactly that. Yet the tooling that connects a call to its source, dynamic number insertion from providers like CallRail, WildJar, Delacon or Jet Interactive, is visible on two sites in the entire dataset. Our probe reads homepage code only and cannot see trackers deployed purely inside Tag Manager containers, so treat two as a floor. Even if the true number were several times higher, adoption would still be in the low single digits of a phone-first industry.

The practical consequence compounds the Section 3 finding: for a clinic whose conversions are mostly calls, an Ads tag alone still misses the majority channel. On the website evidence alone, almost no clinic can say “this campaign made the phone ring 30 times”. Google’s call reporting on ads, which the probe cannot see, is the one exception.

1. Sample. The 559 clinics on the roster built for our advertising study (561 businesses there; two duplicate domains merged here). The roster is metro-skewed, so the figures describe this roster rather than the country. It covers about one in seven of the roughly 3,800 practices in Australia (AVA).
2. Probe. One polite fetch of each clinic’s public homepage, with a single retry on timeout, recording the measurement signals in the returned code: analytics IDs, Tag Manager containers, Google Ads tags, the Meta pixel, call-tracking providers, booking links and tap-to-call links. The main run was on 6 July 2026; 36 advertisers whose website address our roster had not recorded were looked up and probed the same way on 7 September 2026.
3. Tag Manager step. For every container found we also fetched its public configuration and extracted the GA4 and Ads IDs inside it. Without that step, 154 sites whose analytics loads only through Tag Manager would have been wrongly counted as analytics-free.
4. Exclusions. Twenty-two sites returned no readable homepage (10 DNS failures, 6 TLS errors, 5 HTTP errors, 1 timeout). They are excluded from every statistic and never counted as tag-free.
5. Direction of bias. Every tag count is a floor, so every “no tag” figure is a ceiling. The probe cannot see tags installed only on inner pages, server-side tagging, analytics tools other than Google’s, or call tracking deployed purely inside Tag Manager. Each section states which way its figure could be wrong.
6. Ads tag absence. A missing Ads tag does not prove a clinic measures nothing: Google’s call-forwarding numbers, GA4 import and offline uploads exist. The report therefore also gives the stricter figure, the 8% of advertisers with neither an Ads tag nor GA4, who have no website-side conversion path at all.
7. Advertiser status. From our June 2026 Google Ads Transparency Center screening, so a clinic that started or stopped advertising since is misclassified. The figures are a point in time.
8. Validation. One roster clinic’s measurement stack is configured by us and served as a known-good check: the probe recovered its container, its Ads tag and its call tracking exactly.
9. Disclosure. We sell the service this report finds missing. That conflict is why every claim is scoped, why the probe’s limits are stated, and why no per-clinic data is published. Aggregates only, because a row-level list of under-instrumented clinics would be a prospecting list dressed as research.

• Google, “Universal Analytics is going away” (support.google.com/analytics/answer/11583528): standard properties stopped processing hits on 1 July 2023.
• Australian Veterinary Association, home page (https://www.ava.com.au/), accessed 6 September 2026: “Across 3,793 practices and a workforce of nearly 30,000 people”.
• Google Ads Transparency Center, June 2026 screening (our advertising study).