Skip to content

Mar 27, 2026 · 1 min read

Buy answer-seizure ratio, not per-minute rates

The cheapest route is rarely the cheapest route. A short guide to evaluating carriers on what outbound teams actually pay for — connected conversations.

CarriersSIPOutbound

Outbound teams buy termination on price per minute, because that's the number on the quote. Then they wonder why connect rates dropped after switching to a cheaper carrier.

The per-minute rate is a real cost. It's just not the dominant one.

The unit that matters

What an outbound operation actually buys is connected conversations. Everything else — minutes, channels, DIDs — is machinery for producing those.

So the comparison that matters between two carriers isn't rate. It's cost per connected conversation, which folds in answer-seizure ratio, how often your calls arrive labelled as spam, and how many attempts you burn to get one live contact.

A route at a lower per-minute rate with a materially worse ASR is more expensive per conversation. This is routine, not exotic. And it compounds: the extra attempts consume dialer capacity and list depth as well as minutes.

What to measure

Per route, per destination, over a window long enough to mean something:

  • Answer-seizure ratio. The headline. Sharp drops usually mean something changed upstream that nobody told you about.
  • Average call duration on answered calls. A collapsing ACD alongside a healthy ASR often means the far end is answering and immediately dropping — which can indicate labelling problems.
  • Post-dial delay. Long PDD means dead air before the far end rings. Callers hang up during dead air.
  • Attempts per connect. The composite number. This is the one to put on a dashboard.

Run these per destination, not globally. A carrier that's excellent to one region can be mediocre to another, and a blended average conceals both.

Route selection should be automatic

Once you're measuring per-route quality, the routing decision should follow the measurement rather than a spreadsheet someone updated last quarter.

Practically: at least two carriers per destination, selection weighted by recent measured quality, automatic failover on failure, and a floor of traffic kept on the secondary so you always have current data on it. That last part gets skipped, and then failover night is the night you discover the backup route degraded three months ago.

Number presentation is part of routing

Calling into a market from an out-of-region number costs you answer rate before the call is even routed. Local presence isn't a gimmick — it's one of the larger single levers on connect rate, and it belongs in the same conversation as carrier selection because the two interact. Some carriers handle local number presentation cleanly in some markets and badly in others.

The corollary: don't evaluate a carrier's ASR using the wrong number presentation. You'll blame the route for a problem you introduced.

Reputation is an asset you can damage

Numbers get labelled. Labelling follows patterns of use — call volume per number, duration distribution, how often people hang up immediately.

Which means the way you dial affects the value of your numbers over time. Rotating a small pool hard is a way to burn it. Teams that treat number reputation as an asset to maintain, rather than a consumable to churn through, hold connect rates longer.

The short version

Ask a carrier for a rate sheet and you'll get one. Ask for ASR by destination over the last ninety days, and the response tells you as much as the data does.

Written by Kaushal KumarVICIdial architect & cloud telephony engineer. If this is the sort of problem you're staring at right now, I take a small number of advisory engagements.