Per-Call vs Per-Minute Pricing for Insurance Agencies: When Each Wins

The Pricing Models in Play

The Three Layers Where Pricing Happens

Insurance agencies pay for telephony three times: at the telecom carrier layer (the actual minutes), at the dialer/CCaaS layer (the platform that runs the floor), and at the lead-vendor layer (the upstream traffic). Each layer has its own pricing logic, and each is moving differently in 2026. Treating them as a single line item is how agencies overpay.

Per-Call: When the Number of Conversations Is the Cost Driver

Per-call pricing makes sense at the lead-vendor layer, where the value is the conversation, not the duration. A live-transfer Medicare lead is worth roughly the same to the agency whether the conversation is 8 minutes or 28 minutes — what matters is that a qualified prospect was on the line. The vendor's economics also align: they paid to generate the inbound traffic and they want to be paid for the connection, not for the agency's handle time.

The trap with per-call pricing is the definition of "billable call." If the contract counts a 30-second hang-up as billable, you're paying for noise. The well-written per-call contract specifies a minimum talk-time threshold (typically 60–120 seconds), excludes calls that immediately disconnect, and credits back calls where the prospect doesn't meet the agreed qualification criteria. Without those terms, per-call lead pricing can produce a cost-per-acquired-policy that destroys a campaign before the principal notices.

Per-Minute: When Talk Time Is the Cost Driver

Per-minute pricing dominates at the telecom carrier layer because the underlying cost — interconnect with the PSTN — actually scales with minutes. Industry telecom pricing has trended down for years (sub-cent per minute for high-volume domestic outbound is broadly available now), but agencies that haven't renegotiated since 2022 are often still paying 1.5–2.5 cents per minute. On a 50-agent floor running 8 hours of talk time per day, that's $1,500–$2,500 per month in pure carrier-rate inefficiency.

Per-minute also makes sense at the dialer layer when usage swings hard with season. An agency that runs 30 agents off-season and 80 agents during AEP would pay for 80 seats year-round on a flat per-seat plan, even though half of them only matter for 90 days. A hybrid platform fee + per-minute model lets the cost track actual usage. The math turns dramatically in favor of usage-based pricing once peak-to-trough volume ratios exceed 1.8×.

Per-Seat: When Predictability Beats Optimization

Per-seat pricing wins when agency volume is steady, agent count is stable, and the principal values budget predictability over per-unit optimization. A consistent 35-agent agency calling a stable book is the textbook per-seat case. The cost is fixed monthly, the math is easy, and minor month-to-month volume variations don't change the bill. That predictability has real value, and operators who chase pure usage-based optimization sometimes underestimate it.

Where per-seat starts to fail: heavy seasonal swings, blended in-office and remote teams where some seats sit idle, multi-tenant agencies running sub-agencies on different schedules, and any operation where agent headcount routinely flexes. In those cases, the per-seat fee is paying for capacity that isn't being used.

The Hybrid That Works for Most Insurance Agencies

For most insurance agencies — those running mixed inbound and outbound, with seasonal swings, across multiple verticals — the right structure is a small platform fee that covers the floor's baseline access, plus per-minute pricing on actual usage. This is what wallet-based per-minute billing models look like, and they tend to win on three dimensions:

A Worked Comparison

Take a hypothetical 40-agent agency: 30 baseline producers and 10 seasonal hires for 90 days during AEP. Average talk time per agent: 4.5 hours per day, 22 days per month. Annual baseline talk-minutes: ~1.8 million; with seasonal lift, peak-quarter minutes add another ~325,000. Three pricing scenarios:

These are illustrative numbers, not vendor quotes. Real agency comparisons need to fold in specific carrier termination rates, dialer feature differences, and seasonal-headcount details. The point is the shape — a hybrid model usually wins for an agency whose volume varies meaningfully across the year.

What to Ask Vendors at Renewal

At every dialer or telecom renewal, ask three questions and let the answers shape the negotiation. First: "What does my bill look like with 10% lower volume? With 30% higher volume?" — sensitivity to usage swings. Second: "What's the all-in cost-per-talk-minute including platform, support, recordings, and any surcharges?" — flushes out hidden line items. Third: "What happens if I want to terminate three months in?" — exits matter, and most contracts make them painful. The full vendor-evaluation rubric in our 15 vendor questions piece is built around this framing.

Tying Pricing to Pacing

Pricing decisions interact with pacing decisions. A predictive-dialing operation generates more total minutes than a click-to-call operation; per-minute pricing will look more expensive on a predictive floor. But the apples-to-apples comparison is per-policy-acquired, not per-minute. We work through the pacing math in our dialer pacing math piece — the gist is that the right pricing model has to match the operational mode the agency has actually chosen, not the mode the vendor is pushing.

Key Takeaways for Agency Operators

Agencies pick pricing models out of habit; the right model depends on call mix, not vendor preference. An afternoon spent re-running the math against last year's actuals usually identifies 8–15% in cost savings without any change in operations. That's higher-margin work than most production initiatives.