AI Receptionist Payback Period: The Month-by-Month Math

Vendors quote ROI as a multiple — "8x return" — because a multiple sounds impressive and hides its assumptions. Payback period is harder to fake. It asks a single question: how long until the cumulative benefit exceeds the cumulative cost?

For a dental practice, that question is unusually easy to answer, because the benefit is not speculative efficiency. It is a specific set of phone calls that currently go unanswered and would instead become booked appointments. You can count them today.

That makes the analysis honest in both directions. If your practice genuinely answers 98% of its calls and has no after-hours volume, the payback period is long or never, and you should not buy an AI receptionist. If you are missing 15% of calls at a $600 new-patient value, the arithmetic is not close.

1. Monthly unanswered calls

Not total calls — unanswered ones. Pull a 30-day call-detail report from your phone provider and count calls that rang out, hit voicemail during business hours, or were abandoned on hold. A typical single-location general practice taking 600 calls a month finds 70 to 130 of them in this bucket.

2. New-patient share of those calls

Existing patients call back. New patients do not — they call the next practice on the results page. Between 20% and 35% of unanswered calls are usually new-patient enquiries, and those are the only ones with real revenue attached. Be conservative: use 20%.

3. Average new-patient value

Not the value of a single hygiene visit. The lifetime or first-year value of a new patient, which for US general dentistry commonly lands between $500 and $1,200 depending on case mix and insurance profile. Use your own figure from your PMS if you have it — production divided by unique new patients over twelve months is close enough.

If you only change one habit after reading this, make it counting unanswered calls monthly. Practices that track it stop arguing about phone systems and start making decisions from data.

Take a practice with 600 monthly calls, 15% unanswered (90 calls), 20% of those being new patients (18 new-patient calls), and a $600 average new-patient value.

Assume the AI converts only half of those recovered new-patient calls into a kept appointment — a deliberately pessimistic assumption, since the caller already chose to dial you. That is 9 new patients a month, or $5,400 in first-year value against a $599 plan.

1. Month 1: $5,400 recovered against $599 spent. Break-even occurs in the first week.
2. Month 3: roughly $16,200 recovered against $1,797 spent.
3. Month 12: roughly $64,800 recovered against $7,188 spent.

Now stress-test it. Drop the conversion assumption to 25% and the new-patient share to 10%. That is 0.45 new patients a month, about $270 — below the plan cost. Payback never happens. That is the honest failure case, and it describes a practice that already answers its phone well.

The sensitivity tells you what to check before buying: your unanswered-call count is the variable that decides the outcome, not the price of the plan.

A payback calculation is only as good as the cost side. Include everything, not just the headline monthly figure.

• Setup or onboarding fees, commonly $500-$2,500 elsewhere in the market. Wavez charges none, but confirm it in writing with any vendor.
• Per-location charges if you run more than one site.
• PMS integration fees, sometimes billed separately per system.
• Overage on calls or minutes beyond the plan allowance.
• Internal time: a few hours of configuration and transcript review in the first fortnight.

A $2,000 setup fee on top of $599 a month pushes first-year cost from roughly $7,200 to $9,200 and adds about three weeks to payback in the example above. Not fatal, but worth knowing before you sign rather than after.