Key takeaways
- New-patient conversion is booked new-patient visits from calls divided by eligible new-patient calls. Decide and write down what counts as eligible.
- Tag every new-patient call with an outcome for a sample period, or listen to a fixed sample each month; both work if done the same way each time.
- Most lost callers fall into a few reasons: no answer, a date too far away, an unresolved insurance question or a promised callback that never came.
- Track the rate monthly with the reasons beside it, and check it by payer and by provider.
Why this rate deserves its own line
New patients make up a modest share of visits but most of a practice’s growth. In 2019, new patients accounted for 16.8% of US physician office visits, about 173.9 million visits[1]. Most of them started with a phone call. A practice can spend on marketing, referrals and reviews, and still lose a large share of the resulting demand at the moment someone picks up, or fails to pick up, the phone.
Patients research before they call. In a Kansas survey of 284 patients, 57% reported reading online reviews of their physician[2]. By the time a new patient calls, they have usually chosen you over alternatives. Losing that call wastes the effort that brought it in.
Define the conversion before you count
The rate is simple: booked new-patient visits that came from calls, divided by eligible new-patient calls in the same period. The work is in deciding what eligible means. Write the rules down and keep them fixed.
- Include callers who are new to the practice, or new to a provider if your system treats that as a new-patient visit, and who want an appointment.
- Exclude callers you cannot serve by policy, such as a plan the practice does not accept or a service it does not offer. Count them separately; a growing number may be a business signal.
- Count a booking made on the same call. Decide whether a booking made on a follow-up call within seven days also counts, and track it separately so you can see how often the first call fails.
- Record the days to the appointment offered. A booking three months away is a conversion on paper, but many of those patients will not arrive.
Tag the outcome of every new-patient call
| Outcome tag | Meaning | Counts as converted? |
|---|---|---|
| Booked | Appointment made on this call | Yes |
| Booked on callback | Appointment made on a follow-up within seven days | Yes, reported separately |
| Date too far | Caller declined the first dates offered | No |
| Insurance question unresolved | Caller could not get a clear answer about coverage | No |
| Will call back | Caller wanted to check something and did not book | No |
| Not eligible | Plan not accepted or service not offered | Excluded from the rate |
| Abandoned or voicemail | Caller never reached a person | No, counted from phone data |
If the phone system supports call tags, add these as dispositions. If not, keep a simple paper or spreadsheet log for two weeks each month. Do not forget the callers who never reach anyone: take abandoned calls and unreturned voicemails from new numbers in the phone report and add them as lost.
Or listen to a sample
Pick a fixed sample
If your system records calls and your notice and consent practices allow it, choose a fixed number of new-patient calls each month, spread across days and times. Twenty is enough to see patterns in a small practice.
Score each call the same way
Note the outcome tag, the first date offered, whether insurance was answered on the call, whether two times were offered, and whether the caller was asked for their details before any long wait.
Write one line per lost call
Why did this caller not book? Keep the language plain: date too far, insurance confusion, put on hold too long.
Share examples, not names
Play one good call and one lost call at the next team meeting, without identifying the staff member.
Where conversion is lost and how to fix it
Date too far away. If the first offer is weeks out, many callers look elsewhere. Nationally, 28% of medical groups said new-patient wait times were longer in 2026 than in 2025[3]. Check third next available new-patient slots by provider; often one provider has open time that the scheduler is not offering because of template rules.
Insurance confusion. Callers who hear “we’ll check and call you back” often do not book. Give staff a current list of accepted plans and products, and a script for the common uncertain cases. If verification is needed, book provisionally and verify before the visit.
Payer differences. A meta-analysis of 34 secret-shopper studies found Medicaid coverage was associated with a 1.6-fold lower likelihood of successfully scheduling a primary care appointment and a 3.3-fold lower likelihood for specialty care, compared with private insurance[4]. If your practice accepts Medicaid, split the conversion rate by payer. An unintended gap is easier to close once you can see it.
Source: Inquiry: The Journal of Health Care Organization, Provision, and Financing (via PubMed Central), 2019[4]
Show the data
| Item | Value |
|---|---|
| Primary care | 1.6× |
| Specialty care | 3.3× |
Unreturned callbacks. When a caller wants to check their calendar, take their number and agree a time. A promised callback that comes the next afternoon has often lost the patient. Among adults using AI for health information, 18% said a major reason was not having a regular provider or not being able to get an appointment[5]; patients who cannot book will find other routes.
Calls outside office hours. New patients often call in the evening or at lunch. Some practices cover those times with an answering service, and others use an SI Employee such as Ava to answer new-patient questions and pass the request to staff for booking the next morning. Whichever you use, tag those calls and measure how many become bookings.
Connect the rate to where callers came from
Conversion becomes more useful when it is split by source. Ask every new-patient caller how they heard about the practice, and record one of a short list of answers: referring clinician, insurance directory, search or map listing, review site, friend or family, or other. Keep the list short so staff use it consistently.
Two patterns are worth checking for once the split exists. Referred patients arrive with a reason to choose you, so a low rate for referrals usually points to a process problem, such as missing referral paperwork or an authorization delay. Callers from directories and search may be comparing several practices, so check whether their losses cluster around the first date offered and to whether the insurance answer is clear.
The split also changes marketing decisions. If a source brings many calls but few bookings, the problem may be the phone handling rather than the source. If a source brings few calls but a high rate, it may deserve more attention. Without conversion by source, a practice can easily cut a channel that was working and keep one that only looked busy.
Keep the source question at the end of the call, after the booking, so it never delays the appointment.
A script that converts
“Thanks for calling. Are you a new patient with us? Great. Can I take your name and the best number in case we get cut off? … We’re in network with that plan. The first new-patient visit I have with Dr. Patel is Tuesday the 14th at 9:20, or Thursday the 16th at 2:40. Which works better?”
The script does three things. It takes a callback number early, so a dropped call is not a lost patient. It answers the insurance question directly. And it offers two specific times rather than asking an open question about availability, which shortens the call and makes booking the easy choice.
Example: reading a month of data
Example: a fictional gastroenterology practice tags 240 new-patient calls in a month. It excludes 20 callers whose plans it does not accept. Of the remaining 220, 132 booked on the first call and 11 booked on a callback, for a conversion rate of 65%. Of the 77 lost callers, 30 declined dates more than six weeks away, 22 hung up on hold or reached voicemail, 15 said they would call back and did not, and 10 left with an unresolved insurance question.
Example, continued: the practice opens two additional new-patient slots each week for the provider with the longest lag, adds a provisional booking option for insurance questions, and starts working the callback list at 11 a.m. and 3 p.m. It tracks the same tags the following month to see which change moved the rate.
Track it monthly
- Conversion rate, with first-call and callback bookings shown separately.
- Top three loss reasons and their counts.
- Median days to the appointment booked.
- Conversion by provider and, where relevant, by payer.
- Calls excluded as not eligible, as a business signal.
Conversion links to first call resolution, average speed of answer and, once the patient arrives, registration accuracy.
Frequently asked questions
How do you calculate new-patient call conversion rate?
Divide new-patient visits booked from calls by eligible new-patient calls in the same period. Exclude callers you cannot serve by policy and count them separately.
What is a good new-patient conversion rate?
Definitions differ too much for a reliable national figure. Measure your own baseline consistently and work on the largest loss reason first.
Should abandoned calls count against conversion?
Yes. A new patient who hangs up in the queue is a lost booking. Add abandoned calls and unreturned voicemails from new numbers to the lost count.
How many calls do we need to sample?
For a small practice, a fixed sample of about twenty new-patient calls a month, spread across days and times, is enough to spot patterns.
References
CDC National Center for Health Statistics (2019). National Ambulatory Medical Care Survey: 2019 National Summary Tables (opens in a new tab) ↑
Kansas Journal of Medicine (via PubMed Central) (2022). Millennials Seeking Healthcare: Examining the Degree to Which Patients Utilize Online Resources (Kansas Journal of Medicine, 2022) (opens in a new tab) ↑
Medical Group Management Association (MGMA) (2026). New-patient wait times largely hold flat in 2026 as some groups add providers in bid to meet demand (opens in a new tab) ↑
Inquiry: The Journal of Health Care Organization, Provision, and Financing (via PubMed Central) (2019). Medicaid Patients Have Greater Difficulty Scheduling Health Care Appointments Compared With Private Insurance Patients: A Meta-Analysis (Inquiry, 2019) (opens in a new tab) ↑
KFF (2026). KFF Tracking Poll on Health Information and Trust: Use of AI For Health Information and Advice (fielded Feb 24 - Mar 2, 2026) (opens in a new tab) ↑