AI Receptionist vs. Hiring Part-Time: The Real Math for Canadian Small Businesses
Every small business owner hits the same wall: the phone rings while you're with a customer, under a sink, on a ladder, or in the chair doing someone's hair. You need someone to answer. The question is who — a person, or software. I run a company that sells the software version, so I'll show my work and let you check it.
What a part-time receptionist actually costs in Ontario
The job posting says $20/hour. The real number is never the posting. Wages: 25 hours/week × $22/hour × 52 weeks = $28,600/year. Employer payroll costs (CPP, EI, vacation pay, WSIB): add roughly 12–15% in Ontario → ~$3,700/year. Hiring and training: job ads, interviews, the two weeks where the new hire is learning your business instead of helping it → conservatively $1,500 in your time and direct costs. Turnover: front-desk roles turn over. If you replace that person once in two years, double the hiring cost. Realistic first-year total: $33,000–$35,000. And here's what that buys: coverage for 25 hours a week. Not evenings. Not weekends. Not holidays. The calls that come in at 8 p.m. on a Tuesday — statistically some of your most valuable, because that's when people finally have time to deal with their problem — still go to voicemail. To cover 60 hours a week with humans, you're looking at two part-timers or one full-timer plus overtime. Now you're at $55,000–$65,000 a year before you answer a single after-hours call.
What an AI receptionist costs
Our own AVA plans run $197–$599 a month depending on call volume and features. Take the middle: $399/month is $4,788 a year. It answers on the first ring, 24 hours a day, 7 days a week, in five languages. It books into your calendar, takes messages, answers FAQs from your website, and texts you a summary after every call. I'm not going to pretend it's identical to a great human receptionist. It isn't. A great human reads tone, calms down an angry customer, remembers that Mrs. Chen likes Tuesday mornings. AI is getting surprisingly good at the first two and still learning the third. But compare like with like: a $4,788/year system that answers 100% of calls around the clock, versus a $34,000/year hire who covers a third of your week.
The comparison nobody makes: cost per answered call
Say your business gets 150 calls a month. Part-timer (25 hrs/week, ~$2,850/month all-in): answers maybe 90 of them → ~$31 per answered call. AI receptionist ($399/month): answers all 150 → ~$2.66 per answered call. Even if the AI only handles 80% well enough and you still want a human reviewing messages, the economics aren't close.
When you should still hire the human
Honesty requires the other side. Hire the person when:
Your average sale is very high and very emotional. Funeral homes, family law, high-end renovations — callers in distress or spending $50k+ want a human voice, full stop.
Your callers are mostly elderly and phone-shy. Some demographics still hang up on anything that sounds automated. Know your customers.
The phone is your brand. If white-glove personal service is literally what you sell, a human at the desk is part of the product.
For everyone else — trades, clinics, salons, auto shops, property management — the honest answer is that an AI receptionist handles the routine 90% (booking, rescheduling, pricing questions, "are you open Saturday?") and you handle the exceptions.
The hybrid most businesses actually land on
Here's what we see working best with our customers: the AI receptionist answers everything, instantly, around the clock. Routine calls get fully handled. Anything complex, emotional, or weird gets a warm handoff — the AI takes the details and tells the caller you'll ring back within the hour, then pings you immediately. You get 24/7 coverage for the price of a nice dinner each month, and your human hours go to the calls that actually need a human. That's not replacing people. That's just arithmetic.
Comments