The original outpatient EMRs were built around billing. Athenahealth did not begin as a pure EHR company; its health IT platform grew out of claims, payments, and revenue cycle work before the clinical chart became part of the suite. That history matters because the architecture of a system leaves fingerprints on the workday of every physician who uses it.
Billing became the backbone. On top of that backbone, vendors stacked point-and-click documentation templates. The physician sat in the exam room with a patient in front of them and a screen beside them, clicking through review of systems, physical exam elements, diagnosis fields, orders, billing prompts, quality measures, and plan templates. Dozens of small documentation decisions had to be made while the patient was talking.
That design did exactly what it was built to do. It generated structured data for bills while the patient conversation, the quality of the note, and the patient's ability to remember the plan were pushed downstream. It turned doctors into data entry people, and those of us who have spent the last 25 years in medicine have made millions and millions of clicks pushing data into EMRs so the system could get paid.
The Problem Was Never Just Bad Interface Design
That misses the deeper problem; the workflow is backwards.
The visit is supposed to start with a human story; the patient explains the pain, the medication side effect, the fear about the CT scan, the reason they stopped taking the GLP-1 even though it was working. A good clinician listens for pattern, risk, context, and motivation. Legacy EMRs interrupt that process by asking the clinician to translate the story into billing-shaped fragments while the story is still unfolding.
eClinicalWorks, Athena, Epic, and other EMR-first tools can be useful, especially when a practice needs scheduling, charting, and claims tools in one place. But the core interaction is still tab-heavy and template-heavy. The documentation serves the system more than the visit. You can make the clicks cleaner. You can make the template library bigger. You still have a clinician feeding data into a machine while trying to maintain eye contact.
Ambient Scribe Changes the Center of Gravity
You can say what you want about AI. It is a mixed bag, and medicine should be allergic to breathless software promises. But for EMR workflows, ambient scribe technology is a real break from the old model because it lets the conversation become the source material.
Right now, most ambient scribes are being judged on whether they write good notes. That is fair. The note is the first thing clinicians feel. Grail writes great freaking notes, and that matters because a bad AI note is just another inbox item with better marketing.
The note is the first output. Once the patient conversation is captured, understood, and organized clinically, it can drive the rest of the practice workflow. The easy to use EMR of the future will understand what happened in the room and prepare the work that logically follows.
The Visit Should Create the Work
A patient visit contains the reason for the medication change, the failed alternatives, the home blood pressure pattern, the imaging concern, the follow-up interval, and the exact explanation the patient needs to hear again later.
A next generation EMR should use that material to create prior authorizations for medications, queue orders for staff, prepare prescriptions, draft imaging orders, and generate patient instructions that actually match the conversation. That last part matters. We have all watched patients sit through a careful explanation, nod along, walk out of the room, and forget half of what was said before they reach the parking lot.
The visit should produce a clean note for the chart, a readable plan for the patient, clear next steps for staff, and a reliable way to track forms that need signatures. New patient intake can happen through a secure chatbot conversation with fewer brittle packets of endless fields. This is where ambient scribe becomes ambient clinical workflow.
Revenue Cycle Is Part of This, Whether We Like It or Not
No physician went into medicine because they wanted to master revenue cycle. Still, in the world we actually work in, billing determines whether a small practice survives.
Insurers have asymmetric control of the information. They know the medical policy. They hold the money. They can use AI and automation to deny claims, delay payment, and force practices to spend staff time proving that obvious medical care was medically necessary. The practice is left paying a revenue cycle tax just to collect money for work already done.
If the EMR has the visit conversation and the relevant medical policy, it can help craft stronger claims, support prior authorizations, respond to denials, and document the medical necessity that was already discussed in the room. The system remains predatory. The small practice gets a fighting chance inside it.
What an Easy to Use EMR Looks Like Now
Easy to use used to mean fewer clicks. Modern EMR workflows need a higher bar. A modern EMR should reduce the number of times the clinician has to translate the same clinical thought into different administrative formats.
The next generation EMR should document the visit, run the office, schedule appointments, guide intake, manage forms, protect patient data, and integrate with the systems a practice already depends on. It should make eClinicalWorks, Athena, Epic, and other existing tools less painful while the market shifts. Over time, the old point-and-click template system and the one-biller-per-provider operating model are going to look increasingly strange.
Grail is designed for that shift: an EMR workflow built around the patient conversation, where the clinical story comes first and the administrative work follows from it.
There is plenty in medicine right now that is hard to feel optimistic about. This part is different. For once, the technology can move the work closer to the actual practice of medicine. I still think clinicians should stay skeptical. I also think the old EMR model is living on borrowed time.