What Artera Got Right (And What's Still Missing)

Artera is a patient communication platform serving hospitals and health systems. It does text-based outreach — appointment reminders, care gap notifications, post-visit follow-ups. It's integrated with major EHR platforms. It has raised over $150 million and serves hundreds of health systems across the US.

It is also, in our view, a clear map of where the market is going — and a useful frame for what's still missing.

This isn't a competitive takedown. Artera is serving a real need, doing it well enough to achieve significant scale, and proving that the healthcare communications market has genuine depth. All of those things are worth understanding.

What Artera Got Right

The market is real. This is the most important validation. Artera exists at $100M+ ARR because healthcare organizations desperately want automated patient outreach. They want appointment reminders that actually get through. They want patients to confirm or cancel in advance. They want the operational burden of routine communications taken off their front desk.

If you needed proof that healthcare organizations will pay for automated patient communications, Artera provides it at scale. The market is validated.

The ROI is provable. Artera's customers measure outcomes — no-show reduction, patient portal adoption, staff time savings. The ROI is concrete enough to justify enterprise contracts. Healthcare organizations are conservative buyers; that Artera closes enterprise deals means the value case is defensible in procurement reviews.

Workflow integrations matter. Artera's deep integrations with Epic, Cerner, and Athenahealth are a meaningful part of why large health systems use it. The product connects to the EHR, pulls appointment data, and sends communications automatically. The integration isn't an afterthought — it's the core of the product.

These are lessons worth internalizing. The market pays for automation. ROI has to be measurable. EHR integration is table stakes for hospital-scale buyers.

What's Still Missing

Voice. Artera is a text-first platform. SMS, in-app messaging, email. The premise is that text is the right medium for patient communication.

That premise is partially correct. For asynchronous, low-stakes communications — "your appointment is tomorrow at 2 PM, reply C to confirm" — text is efficient. The patient can act on it at 11 PM. No human attention required.

But text has a ceiling. The 73% of patients who prefer voice for complex interactions — discussing a procedure, understanding a diagnosis, dealing with insurance questions — don't get what they need from a text message. The interactions that actually drive patient decisions happen over the phone. Artera doesn't operate in that space.

AI. Artera is workflow automation, not AI. When a patient responds to an Artera message with an unexpected question — "my insurance changed, will I still be covered?" — the message gets routed to a human. The platform is excellent at handling responses that fit expected patterns. It doesn't handle open-ended conversation.

This is a reasonable design choice for 2019. It's increasingly limiting in 2026. Patient expectations have shifted; people expect software to understand what they're saying, not just match responses to keywords. The practices building on workflow automation platforms are starting to hit the ceiling.

Bot-native API. Artera is built for human administrators. The configuration interface, the reporting, the integration setup — all of it is designed for a human who logs in and manages a campaign. There's no API surface designed for an AI orchestrator to configure and deploy communications programmatically.

This matters more than it might seem. As healthcare organizations build AI-driven operations — where AI orchestrators manage workflows that span scheduling, communications, billing, and clinical documentation — they need their infrastructure tools to be machine-callable. A platform that requires a human to log in to configure each campaign is a bottleneck in an AI-native workflow.

Vertical depth. Artera serves broad healthcare, not specialty-specific verticals. The feature set and the knowledge built into the platform are generalized to work across hospital systems, primary care, specialty practices, and urgent care. That generalization is a strength at the enterprise level (one product for the whole IDN) and a weakness at the specialty level (no dental-specific knowledge, no veterinary-specific integrations, no dental PMS connections).

Specialty practices have specialty needs. A dental recall campaign has different timing, language, and outcome metrics than a primary care cancer screening reminder. The platform that serves dental deeply will outperform a generalist in dental. The platform that serves veterinary deeply will outperform a generalist in veterinary. Horizontal breadth trades depth for scale.

The Lesson for Voice AI Builders

Artera validated the market. It proved that healthcare organizations will pay for automated patient communications, that the ROI is defensible, and that EHR integration is prerequisite, not bonus.

The gaps it leaves are the positioning map for voice AI. Start with voice (higher conversion for complex interactions). Add AI (handle open-ended conversations, not just expected-pattern responses). Build API-first (make the platform machine-callable, not just human-operable). Go deep in verticals (dental-specific knowledge and integrations, not generalized healthcare).

The question isn't whether the market exists — Artera answers that with $100M+ ARR. The question is which layer you own in the next evolution of that market.

Text-first, workflow-automated, human-operated platforms captured the first wave of enterprise patient communications. Voice-first, AI-driven, bot-native platforms are building the next one. The market is proven. The question is whether you're building the next version of what already works, or whether you're building what's missing from it.

---

This post concludes The Telephone Moment series. Read from the beginning: The Phone Call Isn't Dead. It Just Got an AI on Both Ends.