Per Minute vs Per Call Answering Service Pricing: Which Billing Model Costs Less, and the Increment Rule That Decides It

Per-minute answering service pricing is cheaper when your calls are short and a meaningful share of them are junk. Per-call pricing is cheaper when your calls are long and mostly real. The break-even sits between about two and seven and a half minutes of average call length depending on which two plans you compare, and the cleanest proof is LexHelper, which sells both models: its 100-call plan is $294 a month while its 100-minute plan is $258, so any call that runs longer than roughly 68 seconds is cheaper billed by the call. What almost nobody checks before signing is the billing increment, and that single rule can move a per-minute bill by 30 percent.

Last updated July 2026. All prices read off each provider's own published rate card.

Which is cheaper, per minute or per call?

Neither, universally. The answer is set by one number you already have and probably have never written down: your average call length. Per-minute billing charges for time, so short calls are cheap and long ones are expensive. Per-call billing charges a flat fee whatever happens, so long calls are cheap and a 30-second wrong number costs exactly what your best lead of the month costs.

The way to settle it is to divide, not to guess. Take a per-call plan's price per included call, divide by a per-minute plan's price per included minute, and you get the break-even call length. Below that length, per-minute wins. Above it, per-call wins. Here is that calculation run across real published July 2026 plans.

Two things fall out of that table. The first is that the break-even is not a market constant. It ranges from just over a minute to nearly seven and a half, entirely depending on which two vendors you happen to be comparing, which is why blanket advice about which model is cheaper is worthless.

The second is the last row, and it is the most useful comparison on this page because it removes the vendor from the equation. LexHelper sells both a per-minute and a per-call rate card for the same service. On its own numbers, any call longer than about 68 seconds is cheaper billed by the call. Since PATLive states that typical answering-service interactions run three to three and a half minutes, most real traffic sits comfortably past that line. On LexHelper's rate card, a firm taking 100 calls a month averaging three minutes would need the 500-minute plan at $1,026 to cover 300 minutes, against $294 for the 100-call plan. That is three and a half times the price for identical service.

How do answering services bill minutes?

Not by the second, and this is where per-minute pricing quietly gets more expensive than the rate card suggests. Providers bill rounded time, and the rounding rule is set by the vendor rather than by any industry standard. PATLive publishes its rule directly: "We bill the first minute in full, and then each 10th of a minute thereafter (in 6 second increments)."

The six-second increments after the first minute are genuinely good, and better than the full-minute rounding some services still use. It is the first minute that costs you. Every connected call bills at least sixty seconds regardless of what happened in it. A robocall the service picks up, a wrong number, a supplier cold-calling your office, someone who says "sorry, wrong department" and hangs up: each one bills a full minute at your plan's rate.

Put that against a plan. PATLive's Starter is $250 for 75 minutes, which prices an included minute at $3.33. If 25 of your 100 monthly calls run under a minute, and that is a conservative share once you count spam and misdials, those 25 calls consume a third of your allowance and $83 of your $250, and not one of them was a customer. On Ruby's Starter, where 50 minutes cost $250 and a minute is therefore $5.00, the same 25 short calls would eat half the plan.

So the three questions to put to any per-minute provider in writing, before you compare a single headline rate, are these. What is your billing increment? Is the first minute billed in full? Does billable time include hold, transfer, and after-call wrap? A provider that answers all three cleanly is usually the one worth buying from, and the answers change your effective rate far more than a ten-cent difference in the published one.

What does per-call pricing hide?

Per-call plans avoid the rounding problem entirely, which is their real advantage. Their weakness is that they charge the same for everything, so your junk-call rate becomes a direct line item. If a fifth of your calls are spam, a fifth of your answering bill is spam.

The other thing to watch on per-call plans is the surcharge menu. Smith.ai's live plans meter the parts of a call that actually produce value: booking an appointment adds $1.50, a conflict check $0.50, taking a payment $1.00, and extended intake $1.50 plus $0.25 for every question past the first five. A firm on the 90-call Basic plan that books half its callers pays roughly $67 a month in booking surcharges on top of the $810. If intake is the reason you are buying the service, price the surcharges in, because they attach to exactly the calls you wanted.

That five-question intake cap is worth sitting with. A real new-client conversation for a professional practice runs ten to fifteen questions. Under a per-call plan with metered questions, the thorough version of the call is the expensive version, which is a strange incentive to build into a service whose whole job is gathering information. It is the same tension per-minute plans create through time, arriving by a different route.

Which model is better for a law firm?

Law firms sit awkwardly in both models, and it is worth being specific about why. Legal intake calls are long by nature: you are capturing a factual narrative, screening for conflicts, checking jurisdiction and statute of limitations, and often calming someone who is having a bad week. Six to twelve minutes is normal. That length makes per-minute billing expensive fast, and on a Ruby Small Business plan a single eight-minute intake call consumes $31.60 of a $395 allowance.

Per-call pricing suits that length better, which is why the legal-specialist providers lean that way, and why LexHelper's per-call card undercuts its own per-minute card so heavily for real traffic. But per-call plans usually pair with an intake cap, and a capped five-question intake on a personal injury inquiry is not intake. It is a message with extra steps.

The practical read for a firm: if you want a message taken accurately and a call transferred, buy per-call from a legal specialist and check the cap. If you want the conversation to actually finish the inquiry, screening the caller against your criteria, collecting the documents, and booking the consultation, then both meters work against you, because both charge more for the thoroughness you are paying for. That is the case for a legal answering service priced flat rather than by usage.

What about flat-rate pricing?

Flat pricing sets the monthly cost by the size of the business and counts nothing at all. No minutes, no calls, no unique callers, no seats. Consultations is priced this way at $49 a month for a solo professional, $129 for up to five, and $299 for up to twenty.

The structural argument for it is that every metered model charges you more precisely when the service is working. Run a campaign that succeeds and the per-minute bill climbs. Ask the full twelve-question intake instead of a cropped five and your minutes double. Screen out the spam and you pay for the screening. Flat pricing removes that tension: there is no way to use the product more that costs more.

The honest limit is volume. A business taking 30 short calls a month is genuinely better off on a small metered plan, and Specialty Answering Service's $159 tier or Rosie's $49 AI plan will cover it for less. Flat rates start winning once call volume is real, once calls need to be long because you want them thorough, or once you need the call to end in a booked and documented consultation rather than a message someone still has to work. Whichever way you go, treat the answering line the way you would any other recurring software cost, where the published rate and the invoice that arrives rarely match unless somebody is actively watching the spend.

How to price your own answering service bill in five minutes

You need three numbers, and your phone system already has all of them. Pull last month's call log and count total inbound calls, total talk minutes, and the number of calls under sixty seconds. Then run each candidate plan.

1. For a per-minute plan: take your total talk minutes, then add one minute for every call that ran under sixty seconds if the vendor bills the first minute in full. That adjusted figure is your billable minutes, not the raw total. Multiply by the effective rate, or check which tier it falls in.
2. For a per-call plan: multiply total calls by the per-call rate, then add the surcharges for whichever calls will trigger booking, payment, or extended intake.
3. For both: repeat the calculation at 150 percent of last month's volume. That is the number that matters, because you are buying this service in order to grow, and the plan that is cheapest today is frequently not the one that is cheapest after the marketing works.

That last step is the one people skip and later regret. Metered plans are priced so that the tier you comfortably fit in today is the tier you will exceed the month a campaign lands, and the overage rate, not the plan price, is what a good month actually costs you. Full published rate cards for every provider named here, including effective rates and overage, are in our answering service pricing comparison, and if you are weighing software rather than a staffed service, the same arithmetic applied to AI receptionist pricing gives a very different set of numbers.