A per-minute meter punishes the calls you want

Take a caller who rings a dental practice, describes a chipped tooth, gets offered three slots, picks one, gives a date of birth and hangs up. That is a four-minute call and it is the single most valuable thing the phone did all day.

On Dialzara's Business Lite plan, published at $29 for 60 minutes with $0.48 a minute after that, four minutes of overage costs $1.92. A one-minute wrong number costs $0.48. The good call costs four times the bad one.

On our Starter plan the same four-minute call costs $0.45 once the 200 included calls are used, and so does the wrong number. We are not claiming that is cheaper in every case, because it plainly is not. We are claiming it charges the same for a call regardless of whether the call was worth having, which is a defensible thing for a meter to do.

Rosie's Professional plan is $49 for 250 minutes. If your average call runs four minutes, that is roughly 62 conversations. Our $49 covers 200 answered calls. If your calls run one minute, Rosie's plan gives you 250 and ours gives you 200, and Rosie is the better deal. That is the honest version of this comparison.

A per-minute meter means the vendor earns more when the assistant is slower. Nobody is going to build a deliberately sluggish product to inflate a bill, but incentives do not have to be acted on cynically to shape a roadmap. If a change makes calls thirty seconds longer and slightly more thorough, a per-minute vendor has no reason to resist it and a per-call vendor does.

We would rather sit on the side of that trade where shortening a call costs us nothing. It is a small thing and we are not going to pretend it is the main argument, but it is real and it is worth saying out loud.

Goodcall goes further in the same direction: unique-caller pricing with uncapped minutes, published at $79 for 100 unique customers. On call length they have the cleanest incentive of anyone in this category, and where a customer rings repeatedly about one job, their model beats ours outright.

If your calls genuinely are short, per-minute is better value and you should buy it. A restaurant answering hours and reservation questions might average well under a minute, and on that pattern 250 minutes goes a very long way while 200 calls does not.

The same holds for a business whose inbound is mostly quick confirmations. Do not buy a pricing model because an argument on a vendor's blog was well constructed. Take last month's phone bill, work out your average call length, and multiply.

What we would push back on is buying a per-minute plan without knowing the overage rate, and four of the vendors we track do not publish theirs. A meter you cannot compute is not a price, it is a promise.