How Much Do Missed Calls Actually Cost a Dental Practice?

Most practice owners can tell you their chair utilisation and their no-show rate. Very few can tell you their missed call rate, and it is usually the cheapest number to fix.

This post is the arithmetic, done honestly, with the assumptions stated so you can replace them with yours.

The calculation nobody runs

Start with three numbers you can get this week:

1. Calls that arrive while nobody can answer. Your phone system or carrier can usually export this. If it cannot, count them by hand for five days: every call that rang out during treatment, at lunch, or after close. 2. What fraction of those callers were patients rather than suppliers. Typically the large majority, but measure it — suppliers and labs should be excluded. 3. Your average first-visit value. Not the single treatment — the first appointment plus the hygiene and follow-up it tends to produce.

Then:

`` missed calls per month × share that are patient enquiries × booking rate you would have had if you answered × average first-visit value = monthly revenue leak ``

Put example numbers in it, and mark them as examples:

• 25 missed calls a month
• 80% are patients (20 calls)
• 60% would have booked (12 appointments)
• 1,200 average first-visit value

That is 14,400 a month, or 172,800 a year, from calls that already reached your practice and then left.

Two things usually happen when a practice runs this: the missed-call number is higher than expected, and the booking rate is the number nobody has ever measured.

Why the leak is structural, not personal

A front desk is not failing when it misses calls. It is being asked to do two jobs that cannot be done at the same time:

• Be present with the patient in front of them
• Answer the phone the moment it rings

During a check-up, a crown prep or a difficult conversation about cost, the phone is simply not going to be answered. That is not a staffing failure; it is a scheduling collision. Adding a person covers the collision for as many hours as you hire, and then you are back to the same arithmetic at 7pm and on Saturday morning.

What people actually try, and what it costs

Voicemail. Free. Also the worst-performing option: most callers do not leave a message, and a message left at 6pm is a booking decision deferred to the next morning, when the caller may have already booked elsewhere.

A human answering service. Roughly per-minute or per-call, and quality varies enormously. The failure mode is not price — it is that the person answering has no access to your schedule, no knowledge of your policies, and no ability to answer the two questions that matter most: do you have anything this week, and what will it cost.

A part-time receptionist for extended hours. Predictable, and expensive the moment volume is uneven. Saturations and holidays still happen.

An AI voice agent bound to your own schedule and policies. Answers after hours and during treatment, answers from your price list and preparation instructions, and books into the real calendar. The honest constraint: it must be given your actual knowledge, or it will invent an answer about your opening hours.

The four numbers that decide it

Do not compare monthly subscription cost to "saved salary" — that comparison flatters every option. Compare:

• Metric — Missed call rate before and after What it tells you — Whether the coverage actually changed
• Metric — Share of calls resolved without a human What it tells you — Whether the agent is doing the job or just forwarding
• Metric — Bookings created outside opening hours What it tells you — The revenue that did not exist before
• Metric — Cost per booked appointment What it tells you — The only comparison that matters across options

The third row is usually where the answer is. Calls that arrive at 8pm on a Sunday were never going to be answered by a receptionist, so any bookings from them are new revenue rather than displaced work.

Where this is the wrong purchase

Be honest about these:

• Your missed call volume is genuinely low. If it is three a month, a subscription costs more than it returns.
• You need clinical judgement on the phone. Triage questions about pain and urgency should reach a person.
• Your patients are long-standing and expect to hear a familiar voice. That is a relationship asset, and an agent can erode it if it is deployed as a replacement rather than as overflow cover.
• Nobody will maintain the knowledge base. An agent answering from stale prices is worse than voicemail.

A note on patient data

Anything that handles patient conversations touches sensitive information. Before deploying, decide what the agent is allowed to capture, where the transcript is stored, how long it is retained, and who can access it. This is a policy question for the practice, not a feature to be assumed.

Disclosure: I work on RingBot, an Agent-as-a-Service platform whose voice agents are built through an AI Builder, bound to your own knowledge base, and able to book into a calendar and hand off to a person with a structured summary. The arithmetic above applies to any option — including a human answering service.