Is an answering service right for my business?

It is the right fix when your problem is coverage, and the wrong one when your problem is anything else. An answering service buys you exactly one thing: something picks up when nobody in your business can. If calls are being answered and sent to the wrong desk, that is a routing problem inside software you already pay for. If the work is piling up during the day and the phone is only part of it, that is a hiring problem. If you already cannot service the leads you have, capturing more of them makes the week worse. BotsDontSleep answers on the first ring 24/7 in 90+ languages for $0.60 CAD per answered minute, billed by the second, with no setup fee and no contract — and none of that matters until coverage is the problem you actually have.

What does an answering service actually buy you?

Coverage. That is the whole product, and it is worth being precise about what it means.

Coverage is the difference between a call that reaches nothing and a call that reaches something. The something has to do four jobs on the call itself: find out who is ringing and why, ask the questions your business needs answered before anyone can act, write a record somebody can work from, and decide by your rules whether this wakes a person or waits until morning. A system that only does the first and the third is a voicemail box with better transcription.

The clearest version is a line where the caller will not wait. A worried owner ringing a veterinary clinic at nine in the evening does not leave a message and sit with it; they ring the next clinic, or they drive to the emergency hospital. Same with a burst pipe. On a Hamilton plumbing line, the February call at 1am is the valuable one and the one most likely to be lost, which is an uncomfortable combination to leave to a greeting.

That is coverage as a purchase. It is narrow. Most of the ways a business loses calls are not that.

Which problem do your missed calls belong to?

Write down the symptom before you write a cheque. Four problems wear the same costume, and only one of them is coverage.

Two rows send people to the wrong purchase more than the rest. Routing is the common one: a call that reached your business and reached the wrong desk is a configuration afternoon, and layering a service on top of a broken hunt group just adds an invoice to the same problem. The other is capacity. If the constraint is that you cannot get to the work, a service that captures every call will fill your voicemail with people you then disappoint, which is a worse outcome than the voicemail you started with.

How do you tell a coverage problem from a routing one?

One question does it: did the call reach your business and reach the wrong person, or did it reach nothing at all?

Your call log answers it, and every cloud phone system has one. Export last month, keep the calls that ended in voicemail or an abandoned ring, and sort them into those two piles. Wrong-desk calls are routing. Nothing-at-all calls are coverage. If the routing pile is bigger, fix the routing first and buy nothing — our piece on answering service or phone system software covers which of the two owns which failure.

Then read the timestamps on the coverage pile. They usually cluster in two bands, after close and the middle of your busiest hour, and those are separable purchases. You can forward evenings and weekends without forwarding anything else, which is the smallest version of this and the one made entirely of calls you are currently losing.

When is the real answer to hire somebody?

When the gap is daytime judgment and hands, not hours on the clock.

The arithmetic is worth doing honestly. Ontario's general minimum wage is $17.95 an hour, effective 1 October 2026 and set until 30 September 2027 (Ontario Ministry of Labour, Your guide to the Employment Standards Act). Forty hours a week across fifty-two weeks is $37,336 in base wages, or about $3,111 a month. Statutory on-costs — CPP, EI, vacation pay, WSIB where it applies — add something on top; we use roughly 10 to 15% as a planning estimate and your accountant should give you the real figure for your business.

What that buys is forty of the week's 168 hours, minus holidays, sick days and the time the person is on their lunch. It also buys things an answering service cannot: somebody who can walk down the hall, read a room, chase a supplier, and exercise judgment on the call that does not fit any rule. If the reason your phone goes unanswered is that nobody is doing the front-desk job at all — the bind a paralegal on an Ottawa legal line knows, screening calls that never convert while the billable work waits — the honest answer is a hire, and a virtual receptionist compared against a full-time employee sets out where each one wins.

A hire and coverage are also not mutually exclusive. The common arrangement is a person for the hours you are open and something answering the rest, because the alternative to coverage is not a cheaper person. It is voicemail.

What does getting this wrong actually cost?

Less than most people fear in one direction, and more in the other.

On price, the category is wide. Across twenty providers whose published pricing we read on 15 October 2026, a 300-minute month cost a median of $804 CAD at the ten human-staffed services and $169 CAD at the ten AI services — the full table, with every figure linked to the provider's own page, is our answering service price index. Ours is $0.60 CAD a minute billed by the second, so 300 answered minutes is $180: above the AI median on a steady month, and below most of the group when volume is uneven, because there is no bundle to pay for in a quiet quarter. The rate and the billing rules together matter more than the headline number, since a service that rounds every call up to the next minute turns a 90-second message call into two billed minutes.

The expensive mistake is not the monthly fee. It is buying coverage for a routing problem, leaving the routing broken, and concluding a year later that answering services do not work. That one costs you the fee and the fix.

When is an answering service the wrong purchase?

Five cases. We would rather say them here than after an invoice.

Your volume is small and you answer it yourself. Under roughly twenty calls a month, answered from your pocket, the phone in your pocket is already doing the job. Buy nothing.

Your problem is routing. Covered above, and it is the most common wrong purchase in this category. No service fixes a hunt group.

You already have reception for every hour anything arrives. Test that rather than assume it. Pull the log and look at what lands after close. If it is empty, it is empty, and coverage is not your constraint.

A meaningful share of your callers will not deal with an automated voice. Then capture rate is beside the point: the message is perfect and the caller is gone. Some practices know this about their clientele and they are usually right. Where a person on every call is the requirement, a staffed service is the better purchase, and our Posh comparison sets out what that buys instead.

You cannot service the demand you already have. More captured leads is the wrong direction. Fix the capacity, then come back to the phone.

There is a sixth we will not rule on: where your profession has rules about how calls are recorded, retained or disclosed, that is a question for your own compliance advisor, not for us and not for a vendor's marketing page.

What should you check before you decide?

Call one as a customer, at the hour you actually get missed calls, and try to break it.

Ask for the most common thing you sell and see whether it reaches a real calendar or just sends you a notification to do it yourself. Then ask something it cannot possibly know and listen to what it does — a good agent says it does not know and hands the call on; a bad one guesses, and you want to find that out before a customer does. Say the word emergency and see whether your rules fire. If your callers switch languages mid-sentence, do that too.

Ours is Maryann, on +1 437-494-9110, at any hour. Launch takes about one week once you decide, and the work in that week is mostly you: your hours, your prices, your escalation rules, and a review of real test calls before anything touches a customer. The businesses this suits are the ones that can describe what should happen to a call at 2am — a service advisor at National Auto Repair in Brampton is under a hood rather than at a desk, and the rule for the call that arrives meanwhile is a decision the shop makes, not one we make for them.

If you have not yet counted what you are losing, do that first. A one-week call audit is the cheapest way to find out whether there is a coverage problem at all, and the break-even arithmetic tells you what it has to be worth before the fee makes sense. This page only tells you whether the fee is aimed at the right thing.