Small business

An AI receptionist for a small business, where every call is a first impression

Start free pilotFree pilot. About 14 days to a working line, no call limit.

An AI receptionist suits a small business for a reason that runs opposite to the usual pitch: low call volume raises the stakes on each call rather than lowering them. Across 15 weeks at a Calgary landscaping company we measured 80 calls (about five a week) and 69 of them came from people who had never called before. At that volume there is no margin for a missed call, because there is no second call from the same person to catch it. AnswerAI builds one line for one business, answers every call including up to twenty at once, and charges a flat monthly fee with no per-call meter.

The figures below are first-party and measured. They come from one company over one season and are published as exactly that: a real distribution from a real small business, not a category average.

What does a small business's call log actually look like?

Sparse, and overwhelmingly made up of strangers. This is one company across 15 weeks, published with their agreement, and it is the clearest small-business call distribution we have.

  • 80 calls in 15 weeks, about 5.3 a week.Measured, first-party: Lammer Enterprises, April 21 - August 6, 2026. Full scope on /results/lammer.
  • 69 of the 80 were first-time callers: 86% had never rung before.Measured, first-party. 69 ÷ 80 = 86%.
  • 100% answer rate across the window.Measured, first-party. Every call that arrived was answered.
  • 76 seconds average call length, long enough to establish a job and book a visit, short enough that a minute-metered plan barely registers it.Measured, first-party.

Why does low volume make each call matter more?

Because the calls are not repeats. A business taking four hundred calls a month has a customer base ringing it; a business taking twenty has a market discovering it, and the two failure modes are nothing alike.

At 86% first-time callers, almost every ring is somebody's first contact with the business. A missed call in that population is not a delayed conversation; it is a person who has never spoken to you, has no relationship to fall back on, and has a list of other numbers. There is no second attempt to catch, because they were never a customer to begin with.

This inverts the intuition that quiet phones do not need coverage. A high-volume business missing 5% of calls loses a slice of a base that will largely ring again. A small business missing 5% of eighty calls loses four strangers a season, each of whom was, at that moment, actively trying to spend money with it.

It also changes what quality means. A large operation can absorb a call handled poorly, because the same customer will ring back next month and the impression averages out. A first-time caller forms their entire view of the business from one conversation, and it is frequently the only one they will ever have with a decision on the line.

The uncomfortable corollary is that the cheapest way to answer a phone is the worst fit for exactly this situation. A generic template that handles the common case well and the unusual case badly is a reasonable trade at four hundred calls a month. At twenty, the unusual case is a meaningful fraction of your entire season.

What goes wrong when a small business buys the cheap option?

Four failures, and they share a cause: plans priced for low volume are built for low complexity, and a small business is frequently not simple at all.

  • The plan is sized for minutes you do not useAt five calls a week almost every entry tier covers you, which makes price look like the only variable. It is the variable that matters least at this volume: the monthly difference between the cheapest and dearest AI plan is smaller than one lost customer.
  • The template does not know your one weird thingEvery small business has one: a service you no longer offer, an area you will not travel to, a question you must never answer with a number. A configured template handles it if you thought to configure it, and a stranger finds it if you did not.
  • Nobody is watching the callsAt scale, a bad answer shows up in a metric. At five calls a week it shows up nowhere: it is one transcript nobody reads, and the customer simply does not come.
  • The owner is the escalation pathIn a small business the person the line escalates to is usually the person on a roof or under a sink. Deciding what genuinely warrants interrupting them is a build decision, and getting it wrong in either direction is expensive.

What does a small business actually get?

Stated as what you get. Flat pricing matters more at this size than at any other, because a small business's worst month for cash is frequently its busiest month for calls.

  • BillingOne flat monthly fee. Unlimited calls, no meter, no overage line. No setup fee. The build is the product, and it is included.
  • UpkeepIncluded in the monthly. Every call reviewed, and the person who built it keeps tuning it.
  • The busy dayNo practical ceiling: twenty calls at once is the same as one.
  • The buildCustom software written for one business, against your own call log.
  • Who fixes a wrong answerA wrong answer is a rule we fix, not a model you wait on. The person who built it is the person who answers about it.
  • Time to a working lineAbout 14 days, built with you, starting with a free pilot, no call limit.

When should a small business not buy this?

There are real cases, and a page that cannot name them is not worth trusting on the cases where it does fit.

If your phone genuinely rings two or three times a week and you answer it every time, you do not have a coverage problem and a built line will not create value that was not already there. Smith.ai publishes a free AI tier at 25 calls a month, and at that volume it is a sensible place to start rather than anything we would build.

If your calls are almost entirely from existing customers who know you, the first-impression argument above simply does not apply to you. The stakes on a missed call are genuinely lower when the caller will ring back, and you should weigh this the way a high-volume business would.

And if the value of the work is small enough that a lost enquiry costs less than a month of any plan in this category, the arithmetic does not work no matter how it is presented. That is worth doing on the back of an envelope before talking to anybody, including us.

Where it does work is the shape Lammer has: modest volume, mostly strangers, and enough value in a booked job that four lost enquiries in a season is a number the owner would rather not think about.

At eighty calls a season with eighty-six per cent of them strangers, one missed call is a customer who never speaks to you at all. That is the scale AnswerAI's flat monthly fee is built for, because a small business's busiest month is usually its tightest.

Nick Lovett, Founder, AnswerAI

Count last season's calls and how many were from people you had never spoken to. That second number is the one that decides this.

Start free pilot

Questions small business owners ask

It depends on how many of your callers are strangers rather than on how many calls you get. In the clearest small-business log we have measured, 86% of 80 calls over a season came from first-time callers, a population with no reason to ring twice. If your calls are mostly existing customers who will call back, the case is genuinely weaker.

One flat monthly fee, agreed in writing before anything is built, with no per-call or per-minute meter and no setup fee. Flat matters more at small scale than at any other, because the month your phone is busiest is frequently the month your cash is tightest, and a metered plan charges most exactly then.

Sometimes it is. If you answer every call today and your enquiries are low-value, the arithmetic will not work and a free tier elsewhere is a better starting point. The volume where this earns its place is where enough calls come from strangers that losing a handful a season is a number you would notice.

That is what the build session is for, and it is the main reason a small business is a poor fit for a template. The service you stopped offering, the neighbourhood you will not travel to, the question that must never be answered with a price: those get written down as rules before the line takes a call, rather than discovered by a stranger.

Whoever and whatever you decide, which in a small business is a more consequential decision than it sounds. The line does not judge what counts as urgent; you define the triggers, and it applies them the same way at 7am as at 7pm. Most owners set this deliberately narrow at first and widen it once they have read a few weeks of transcripts.

Answer the strangers.

The pilot runs about 14 days with no call limit. If your call log says you do not need this, we will tell you; it is a shorter conversation than finding out in month three.

Start free pilot

Last updated