Will an answering service help me book more appointments?

Yes, and in one specific way: an answering service books the appointments you are losing now, not the ones you already catch. Every call that hits voicemail at seven on a Sunday evening, rings out while your hands are full, or arrives second while you are already on the phone is a booking that was available and did not happen. BotsDontSleep answers that call in under one second, 24/7, writes the appointment into your calendar during the call, and charges $0.60 CAD per answered minute billed by the second. If you already answer nearly every call that comes in, expect no lift at all, and read on for how to tell which of those two you are.

Where do the extra appointments actually come from?

From calls nobody is answering today. There are four kinds, and they are worth separating because only two of them are about opening hours.

The closed hours. A business open nine to five on weekdays is unattended for 128 of the 168 hours in a week. Calls arrive in that 76% of the week regardless, and for most trades they are the calls with the least patience behind them: a blocked drain, a cracked windscreen, a tooth that started hurting on Saturday night. Our guide to what happens to calls when your business is closed works through the four things that can happen to one of those calls.

The hands-busy hours. These are open-hours calls that nobody could get to. A service advisor is under a bonnet, a stylist has colour on their hands, a hygienist is mid-appointment. The business is open, staffed and still not answering, which is the category owners find most irritating, because they were standing right there. It is also the one a shop never measures, since a ringing phone in a busy bay leaves no record. Shops taking service bookings in Calgary describe the same pattern every spring.

The second caller. One line, one person on it, and the next caller gets a busy signal or a ring-out. An AI receptionist takes simultaneous calls without a queue, so the ninth caller at 8:02am on a Monday has the same experience as the first.

The caller who will not wait for a callback. This is the one people underestimate. A caller who leaves a voicemail has not committed to anything, and in the gap before you ring back they are free to call the next business on the list. Booking on the first call removes that gap entirely.

How many extra appointments is that, in your own numbers?

Here is the arithmetic on a planning example. None of it is a measurement of anybody's business, and the point of writing it out is that you can substitute your own four numbers.

Say 50 inbound calls a week. Twelve of those arrive outside opening hours, and eight ring out while everyone is occupied, so 20 calls a week go unanswered. Roughly half of unanswered calls at most businesses are not enquiries at all, which leaves about 10 real ones. If one in three of those would book when a person picks up, that is three bookings a week and about 13 a month.

One in three is a planning assumption, not a finding. It is the figure we use when nobody has the data yet, and your own number will be higher if your calls are mostly inbound demand and lower if they are mostly existing customers checking on a job. Replace it as soon as you have a month of your own call records to count.

The cost side of that example is small and exactly knowable. Twenty answered calls a week at an average of two minutes is 40 minutes, about 173 minutes a month, which is $104 at $0.60 a minute. Thirteen bookings for $104 works out at about $8 of answering cost per booking. Whether that is a good trade depends entirely on what one appointment is worth to you, which is the only number in this calculation that you already know.

Is a message as good as a booking taken on the call?

No, and the difference is the whole mechanism. A message is a second chance to lose the job. A booking written into the calendar during the call is finished.

A taken message produces a callback, and a callback has to find the caller free, willing and still shopping. Some of them will be none of those by the time you dial. A booking completed on the call skips all three conditions, which is why the ability to write into your calendar matters more than the quality of the voice doing it.

This is also where staffed services and AI ones price differently. Smith.ai, for instance, lists appointment booking as a $1.50-per-call add-on on top of its call charges, read from its own pricing page on 15 October 2026, and our Smith.ai comparison covers where a human on every call is still the better buy. On our rate card the booking is not a separate line: it is part of the answered minute.

Which option books more appointments: voicemail, a staffed service, or AI?

Staffed figures are the cheapest published route to 300 answered minutes at nine services, read from their own pricing pages on 15 October 2026 and converted at the Bank of Canada rate where a provider quotes US dollars. The median of those nine was $804 CAD. Every row, with its source and the date it was read, is in our answering service price index; check the provider's own page for current figures.

What does each extra appointment cost to capture?

Divide your answering bill by the bookings you would not otherwise have had. On the planning example above that was about $8. It falls as volume rises, because the per-minute rate does not change and the short calls get cheaper in aggregate: a 20-second wrong number costs about 20 cents, and it costs the same whether it is your fourth call of the month or your four-hundredth.

What makes this arithmetic unusually clean is the absence of a bundle. There is no plan to exceed, no included-minute pool to run down, no overage rate waiting at the end of a busy month. $0 setup, $0.60 CAD per answered minute billed by the second, no contract and no minimum, and the full rate card has no second page.

Two bookings a month from a $104 bill is a good trade for a dental practice and a questionable one for a business where an appointment is worth $40. Run it with your own appointment value before you run it with ours.

Which bookings will it finish, and which will it hand to a person?

It finishes the ones with rules it can follow: a named service, a duration, a calendar with real availability, and a policy about deposits, double-bookings and how far ahead you will take work. Dental booking calls in Montreal and driving-school lesson bookings are both of this shape, which is why they convert well.

It hands over the ones that need a judgement call: a quote that has to be seen before a time can be offered, a job whose duration depends on what is found on arrival, anything medical, legal or financial beyond scheduling. In those cases it collects the details, books a callback, and alerts whoever is on your escalation list. It never invents an answer, and it does not offer a slot it cannot honour.

When will an answering service not book you a single extra appointment?

Three situations, stated plainly, because they are common.

You already answer almost every call. If you take 20 calls a month and reach 19 of them, the phone in your pocket is doing the job and there is no lost volume to recover. Nothing here creates demand. It only converts calls that already exist, and if none are being missed, there is nothing to convert.

Your constraint is capacity, not intake. If your calendar is full for six weeks and you are turning work away, more enquiries arriving is not an improvement — it is more declining to do. Fix the bottleneck first. An answering service will faithfully book you into a diary you cannot deliver on, which is worse than voicemail.

Your bookings genuinely require a site visit to price. Custom fabrication, structural work, complex insurance matters: if no appointment can be offered until someone has looked, the AI's job shrinks to intake and triage. That is still useful, but it is not booking, and you should not buy it expecting a booking lift.

There is a fourth, less comfortable case. If what you are really unhappy with is how your front desk handles callers, replacing the person with software changes who is unhelpful rather than whether anyone is. Where a human voice on every call is the actual requirement, a staffed service is the honest purchase.

How do you check whether it actually booked more?

Measure it against your own baseline, not against a vendor's claim. This takes about a month and six steps.

1. Count your missed calls for one week before changing anything. Your phone bill or carrier portal shows unanswered inbound calls. Write down the total, and how many fell outside opening hours. This single number decides whether to continue, and most owners find it higher than they expected.
2. Write down what one appointment is worth. Average invoice, or average first-year value for a repeat customer. Keep it conservative; the argument is stronger with a low figure that holds up.
3. Point one slice of the line at the agent, not the whole line. After hours is the usual choice, because the alternative there is voicemail and nothing is at risk that was not already being lost.
4. Set the booking rules before launch. Service list, durations, how far ahead it may book, what needs a deposit, what must come to a person. Doing this takes an afternoon and determines the entire month's results.
5. Run it for a full month and count bookings from that slice only. Not total bookings, which move for a dozen reasons. Only the appointments created on calls the agent answered, which appear in the call summaries with a time attached.
6. Compare the two numbers. Bookings from that slice, multiplied by your appointment value, against the invoice for the minutes. If it does not clear comfortably, the honest conclusion is that you were not missing much, and that is a useful thing to have paid $100 to learn.

Call the demo line before any of that and give it your awkward booking call. Maryann answers at +1 437-494-9110 at any hour, and judging a booking flow is much easier when you are the caller.

Setting this up takes about a week. We take your hours, services, booking rules and escalation list, tune the agent against them, and you listen to real test calls before a customer hears one. It runs on your existing number, across Canada and six time zones, at $0 setup and $0.60 CAD per answered minute billed by the second.