The PIMS landscape through the booked-patient lens

Every comparison of veterinary practice management software scores the same five things: ease of use, features, support, reliability, price. Those are the right things to score if you are choosing software to run a practice. They are the wrong things if you want to know whether next year's marketing spend can be proven or will stay a matter of faith.

This is the one criterion I am qualified to apply, because my work depends on it. Google Ads for a veterinary practice either produces booked patients or it doesn't, and the only place the answer lives is the practice management system: the new-client record, the phone number on it, the date of the first visit. If those three things can leave the PIMS and meet the click and the call that preceded them, attribution is possible. If they can't, every agency report you will ever read is an estimate wearing a decimal point.

Most practice owners chose their PIMS on scheduling, inventory, and lab integration. Almost none knew they were also choosing whether their marketing would ever be measurable. This post is the evaluation that choice deserved.

What I am not doing

I am not ranking these systems on how good they are. I have never run a practice on any of them, and I have never pulled a report from one myself. My seat, for thirteen years, has been the other end of the export: the practice sends me the new-client list, and I try to match it to the clicks and calls that came before it. That vantage point tells you nothing about how a system is to use and a great deal about how hard it is to get data out of, because the file arrives with the story of what it took to produce. The recurring story is that it was difficult. For pricing, ease, and what the interface is like at 4pm on a Saturday, read the people who evaluate those things; several of them do it well.

I am also not telling you to switch. Migration timelines get quoted in weeks and lived in seasons. Nobody should change the software their whole team runs on to make an agency's reporting easier. What you can do is know which category you are in, and ask the right questions before you renew.

Disclosure: I sell one piece of this stack. My work is easier when your data is reachable, so I have a stake in the answer. Weigh it accordingly.

The five questions

Everything below is graded against these, and only these.

1. Can the practice pull its own new-client list — name, phone, first-visit date — as a file, today, without asking the vendor? This is the floor. Manual export is how attribution actually gets done in most independent practices, and it has to be possible for anything else to matter.
2. Can the practice get its own data programmatically, on its own authority? Meaning: a documented interface the practice itself can use, without becoming the vendor's "partner," without a signed agreement, without a six-week queue.
3. Are the terms of access public before you sign? Rate limits, what data is reachable, what it costs. If the answer is "after the NDA," the answer is no.
4. Who owns the vendor, and does the owner sell something else that benefits when your data stays inside? A software-only company has no reason to wall your records off. A diagnostics company, a distributor, or a private-equity roll-up has several. This is the ownership question in another costume.
5. Where does the data physically live? A database on a server you own is reachable by definition. A database in a cloud the vendor operates is reachable exactly as far as the vendor decides.

The finding that surprised me

I expected the modern cloud systems to be open and the 1998-looking server systems to be locked. It is closer to the reverse, and the reason is structural.

AVImark and Cornerstone keep the practice's database on a computer in the practice. Neither publishes an API; both vendors route sanctioned integration through partner programs with signed agreements. But the database is right there, on hardware the practice owns, and an entire industry of analytics and communication vendors reads it directly with a locally installed agent. That is how the reporting dashboards, reminder platforms, and attribution tools that support twenty-plus PIMS actually work on the legacy systems: not through the vendor's front door, but through the practice's own basement.

The cloud systems have no basement. The only door is the API, and whoever controls the API controls the door. That makes the question of who controls it — question four — the deciding one for every cloud PIMS on this list.

The ledger

Ownership and access facts are current as of September 1, 2026; sources are listed at the end. Anything marked not verifiable means I could not confirm it from public documentation and will not guess.

01Tier one: reachable on your terms

Provet Cloud (Nordhealth). Publicly listed on Euronext Growth Oslo as NORDH, founder-controlled through a dual-class share structure (Class B shares carry ten votes). Software only: no diagnostics business, no distribution arm, no pharmacy. The vendor's own language is "open REST API" and "100% data ownership," and it publishes its prices. Of the nine systems here, this is the only one whose owner has no adjacent business that benefits from your data staying put, and whose access terms are printed — which is what owning your own data looks like when a vendor commits to it in public. US business is small but growing — the 2025 annual report shows US ARR up 41.7% — which means less US-specific tooling exists around it than around the IDEXX and Covetrus ecosystems.

Shepherd. Founded and led by Dr. Cindy Barnes; no outside institutional investor has been announced. Acquired Hippo Manager in March 2025. Shipped a user-accessible OpenAPI in January 2025 — the practice itself, not a partner, can connect. Pricing was published at a flat per-doctor rate until mid-2026 and is now quote-only. Multiple attribution and analytics vendors have built direct connections.

Digitail. Venture-backed: a $23M Series B in November 2025 led by Five Elms Capital, with Atomico and Partech participating. Investors want growth, not walled data, which is a different incentive from a diagnostics parent. Public developer documentation, OAuth 2.0, and a self-serve tier for clinics connecting their own systems — the documentation explicitly names "BI dashboards" and "CRMs" as intended uses. Two honest caveats. Access is "reviewed and approved by the Digitail team," so self-serve is not quite self-serve. And the documentation says plainly that the API is not for "one-time data exports or ad hoc reports," which is exactly the manual path most practices use; that path lives in the reporting UI instead.

02Tier two: reachable on theirs

ezyVet (IDEXX). Acquired by IDEXX in 2021. The API is real and well documented — OAuth 2.0, published rate limits (60 calls per minute on most endpoints), public reference docs. The friction is entirely about permission. Production access requires IDEXX Partnerships approval; a practice wanting its own data goes through a "Private Integration" application that the developer portal says can take "several weeks" to answer. SMS and payments are excluded from the API by policy. Owner incentive: IDEXX sells diagnostics, and ezyVet is deeply bundled with them. The system is open; the gate is not yours.

IDEXX Neo. Same owner. Two access surfaces, neither self-serve: per-clinic API keys generated in the app and emailed to an approved partner, and a separate read-only data API with a documented 1,000-item cap and 5-second query timeout. No public developer portal, no sandbox, technical documentation shared after partner approval under partner terms. For a practice, the realistic path to its own data is the reporting screen.

Instinct. Growth-equity backed (Mainsail Partners, 2022), acquired VetMedux — owner of Plumb's — in January 2024 and ScribbleVet in January 2026. Instinct's own integrations page says it has a Partner API "so that our customers and third-party platforms can develop their own integrations," with access requested by email. That is a real door for the practice. What is not public is what the API returns, what it costs, or how long approval takes — the terms live behind the request. In fairness, Instinct is an emergency-and-specialty system, and ER growth runs on referring-vet relationships more than on search ads; the measurement question matters less here than anywhere else on this list.

03Tier three: reachable through the back door

Cornerstone (IDEXX). On-premise, database on the practice's own server. IDEXX publishes the data models for its Data Services API openly, which is more than most vendors here do; what it does not offer is access. Integration runs through an integration request form and a partnership review, and IDEXX's own documentation says batch data "are refreshed from practice locations every 24-72 hours." But the practice owns the box, and that is why so many tools support Cornerstone: they read the database locally. Owner incentive is the same as ezyVet and Neo.

AVImark (Covetrus). On-premise. Covetrus went private in October 2022 in a roughly $4 billion buyout by Clayton, Dubilier & Rice and TPG — private equity, with a distribution and pharmacy business attached, which is exactly the change-of-landlord risk a practice never gets a vote on. Covetrus Connect is positioned as "the only authenticated and supported way to integrate with Covetrus practice management software"; the partner program involves an application, a reported six-to-eight-week wait, and a signed agreement before documentation. As with Cornerstone, the practice's database sits on the practice's hardware, and the local-agent ecosystem reads it. Covetrus's product attention is visibly on migrating this install base to Pulse, which is worth knowing before you invest in an AVImark-specific measurement setup.

04Tier four: closed

Covetrus Pulse. Same PE owner as AVImark, cloud-hosted, so the back door is gone. No public API. Covetrus Connect is the sole sanctioned integration layer, behind partner application, signed agreement, and NDA; per-practice credentials issued one clinic at a time; scopes and rate limits unpublished. It is the only system on this list in this tier, and the combination is what puts it here: private-equity owner with adjacent pharmacy and distribution businesses, cloud-only, and access terms invisible until after you've signed something. If you are on Pulse, the reporting screen and a manual export are your instrument.

Not scored

Koi. DVM-founded cloud PIMS; too little public documentation of data access to grade honestly. When that changes, so will this entry.

ImproMed, DaySmart Vet, Vetspire, NaVetor, and others. Outside the nine. Vetspire in particular is largely a corporate-group system.

What the tiers actually mean for you

If you are in tier one, the door is yours. Attribution is an engineering task, not a negotiation.

If you are in tier two, the door exists and someone else holds the key. Ask the vendor, in writing, what a private integration for your own data costs and how long approval takes. Get the answer before you renew, not after.

If you are in tier three, you are better off than you think. The systems that look oldest are the ones your own analytics vendor can read most directly, because the data never left your building. The measurement setup is unglamorous — a local agent, a scheduled export — but it works, and it works without asking permission.

If you are in tier four, the honest position is that marketing on a closed cloud PIMS can be measured manually and cannot be measured automatically. That is a real limit, not a reason to switch. Nobody should re-platform a practice for an agency's convenience. Run the manual export monthly and treat any vendor who promises automated attribution on your system with the skepticism the terms of access justify.

The manual path, since it works everywhere

Practices running most of the systems on this list have handed me new-client exports, including practices on the closed ones. I have not verified the export path on every system myself and will not claim to; what I can say is that the file has always eventually arrived, and that "eventually" carried real front-desk hours more often than not. A monthly file of new clients with phone numbers and first-visit dates, matched against call-tracking logs and click timestamps, is how attribution gets done in most independent practices in 2026. It is slower than an API and it is entirely within the practice's control, which matters more.

What a vendor's answer tells you

Three questions to send your PIMS vendor. The answers grade themselves.

1. Can I export my complete client list, with phone numbers and first-visit dates, myself, today? If the answer involves a support ticket, that is a no.
2. What does programmatic access to my own data cost, and how long does approval take? If the answer involves a partner program, you are in tier two or below.
3. Who owns your company, and what else do they sell? If the answer is a diagnostics company, a distributor, or a fund, weigh every "integration" they offer against what it does for the parent.

Method and limits

Ownership facts come from filings, press releases, and investor pages; access facts come from each vendor's own developer or integration pages, read directly, as of September 1, 2026. I did not rely on review-site sentiment for anything in the ledger. Where a third-party summary pointed me to a vendor page, I cite the vendor page. I did not receive access, demos, or compensation from any company named.

What this evaluation cannot tell you: what any system costs (mostly quote-only), how it feels to use, how good support is, or whether it will still be owned by the same people next year. It also cannot tell you what a partner-gated integration would return, because by definition that is not public.