Most intake workflows were never designed. They accumulated. Someone added a form after a bad outcome, someone else bolted on a portal because the vendor pitched it well, and now your medical assistant spends 40 minutes chasing down whether Mrs. Patel actually filled out the hormone questionnaire before her Thursday consult. If that sounds familiar, you can fix a surprising amount in one focused afternoon.

What follows is the audit I run when a practice tells me intake is "fine" but their MAs are drowning and new patients keep no-showing their first appointment. Block three or four hours, close your door, and work through it in order.

Before you start: pull the actual artifacts

You cannot audit a workflow from memory. Get the real materials on your desk (or one screen) before you open a spreadsheet:

  • Every form a new patient touches, from the "book a consult" web form to the last consent they sign at check-in.
  • The last 20 new-patient charts, opened side by side with your EMR intake fields.
  • Whatever confirmation, reminder, and pre-visit emails or texts go out, in order, with timestamps.
  • Your no-show and reschedule report for the last 90 days, filtered to new patients only.
  • Fifteen minutes of your front desk's calendar. You need to shadow, not interview.

If gathering these takes more than 45 minutes, that is itself a finding. Write it down.

Step 1: Map the actual path, not the intended one

On a single page, write every step a new patient goes through from first contact to the moment the provider walks in the room. Include the invisible steps: the MA copying insurance cards into a folder, the biller checking eligibility, the clinical lead flagging the chart as "ready." You are looking for the real path, so include the workarounds people have quietly built.

In a concierge or compounding practice I worked with last year, the "official" intake had 6 steps. The real one had 19, including two separate places where staff manually retyped allergies from a PDF into the EMR. Nobody had noticed because each person only saw their own slice.

When your map is done, mark each step with three tags: who does it, how long it takes, and whether the patient sees it. The steps the patient never sees are usually where the waste lives.

Step 2: Count the handoffs and the re-entry points

Every time information moves between a system, a person, or a form, you have a handoff. Every handoff is a place where things get dropped, delayed, or duplicated. Count them.

Then count re-entry: how many times does the same piece of information (name, DOB, medication list, pharmacy of choice, card on file) get typed or asked for? For a healthy intake in a modern practice, a patient should give you their medication list once. Twice if you count a verbal confirmation in the room. If you are at four or five, you have a data architecture problem, not a staff problem.

This is where compounding practices tend to bleed the most time. The intake asks about current meds, the pharmacy questionnaire asks again with different fields, the provider asks a third time in the visit, and the pharmacy tech calls to clarify a fourth. Each one is defensible in isolation. Together they cost you an hour per new patient and make the practice feel disorganized.

Step 3: Look at your forms with fresh eyes

Open your intake packet and read it as if you were a 52-year-old executive who booked a concierge consult on their phone at 11pm. Ask honestly:

  • How many fields are there, and how many actually drive a clinical or billing decision in the first visit?
  • Which questions are asked because a provider wanted them once in 2019?
  • Which fields require a desktop to complete (long free-text boxes, upload widgets that fail on iOS)?
  • Is there a single form that, if the patient stops halfway, kills the whole appointment?

In practice, we see completion rates jump 15 to 25 points when a practice cuts its intake in half and moves anything non-essential to a post-visit follow-up. You do not need the patient's full surgical history to run a first consult about testosterone optimization. You need enough to be safe and enough to bill. Everything else can wait a week.

Step 4: Time the eligibility and scheduling loop

Pick five recent new patients and reconstruct the timeline from "requested appointment" to "appointment confirmed on the calendar with insurance verified and forms complete." If your median is more than 48 hours, you are losing patients to competitors who confirm same-day. If your P90 is a week, you are almost certainly losing patients silently and blaming it on "the market."

The most common culprit is a serial workflow where the front desk waits for eligibility before sending forms, or waits for forms before scheduling. Parallelize it. Book the slot immediately, send forms immediately, run eligibility in the background, and only pause the process if something actually fails.

Step 5: Read your reminder sequence out loud

Literally read your confirmation email, your 3-day reminder, your 24-hour reminder, and your day-of text in sequence. You will find one of three problems: they contradict each other, they repeat the same instructions four times, or they ask the patient to click into three different portals with three different logins. Rewrite them as one clean sequence with one link that goes to one place.

What to fix first

You will end the afternoon with a list of 15 to 30 issues. Do not try to fix them all. Rank them by two factors: how much staff time they consume weekly, and how visible they are to the patient. Then start here, in this order:

  • Kill duplicate data entry. If your MA is retyping anything from a PDF into the EMR, that is the first fire. Either fix the form so it writes directly, or stop collecting that field.
  • Shorten the intake packet. Cut anything not needed for the first visit. Move it to a post-visit form or a scheduled follow-up.
  • Parallelize scheduling and verification. Book first, verify second.
  • Consolidate patient-facing touchpoints. One link, one login, one place to upload a photo of an insurance card.
  • Automate the confirmations that a human is doing today. If someone is manually texting reminders, that is a week-one automation, not a year-two project.

Everything else, the nice-to-haves and the "wouldn't it be great if" ideas, goes on a separate list for next quarter. An afternoon audit that produces three shipped fixes beats a six-week project that produces a slide deck.

A note on tools

You can do this audit with any EMR. The fixes are harder in some than others. If your intake, scheduling, eligibility, and messaging live in four separate vendors that do not talk to each other, you will hit a ceiling on how much you can improve without changing the underlying plumbing. That is when it makes sense to look at infrastructure built for how concierge and compounding practices actually operate. If you want to see what that looks like in one system, talk to our team at Qintara.

Frequently Asked Questions

How often should we re-audit intake?

Once a quarter for the first year after any major change, then twice a year in steady state. Intake decays. New staff add steps, new services get bolted on, and forms grow by accretion. A recurring 3-hour audit on the calendar prevents the slow drift.

Should the practice owner run this, or delegate?

The first one, run it yourself. You need to see the actual mess without it being filtered through someone who is trying to protect their process. After that, your operations lead or office manager can run it with a checklist, and you review the findings.

What if our staff pushes back on shortening forms?

Ask them to point to the last five clinical or billing decisions that hinged on the fields they want to keep. Usually two or three fields survive that test. Keep those, cut the rest, and revisit in 60 days. If something turns out to matter, add it back with intent.

How do we know the audit worked?

Three numbers: median time from booking to confirmed-and-ready appointment, new-patient no-show rate, and minutes of staff time per new patient before the visit. Measure them before you start and again 30 days after your fixes ship. If none of the three moved, you fixed the wrong things.