Why Per-Minute Pricing Misleads Campaign Budgeting

Many organizations evaluating AI calling focus only on the per-minute rate quoted by vendors, assuming it predicts total campaign cost. This approach overlooks what happens when AI fails to resolve a call and routes it to a human agent—you end up paying for both the AI minute and the human minute. As industry research notes, "Voice AI vendors will quote you about $0.10 a minute. The number is real, but it is the wrong one to budget with." According to industry research, cost per resolved call—not per-minute rate—is the only meaningful metric for determining whether AI calling actually saves money.

The true economics depend heavily on containment rate, or the percentage of calls fully handled by AI without human intervention. In a worked example comparing AI to human agents, break-even containment was just 11% versus onshore humans ($0.73/min) but jumped to 43% versus offshore humans ($0.25/min). Research shows that below about 43% containment, AI makes each call more expensive, not less, when compared to offshore labor. At 50% containment with an offshore team, the cost per call was $0.95 versus $1.00 for all-human handling, while at 70% containment, the savings reached $92,000 monthly for an onshore operation.

• Platform fees often represent the largest cost layer in AI call centers, ranging from $0.05–$0.08 per minute.
• Personnel costs typically make up 60–70% of traditional call center budgets, making retention a key lever for savings.
• Automated reminder systems cost a mean of €0.14 per contacted patient versus €0.90 for manual calls—a 6.4x differential.

My AI Call Center addresses this by quoting the full campaign cost before launch, including setup, management fees, and usage charges, so customers see the expected cost per resolved outcome—not just a per-minute rate that can mislead budgeting decisions. This ensures transparency from the start, aligning with the principle that "cost per resolved call is the only number that tells you whether AI saves money." Industry experts confirm this approach prevents surprises and supports smarter investment in outbound calling campaigns.

The Four-Layer Cost Structure Behind Every Campaign Quote

When a voice AI vendor quotes you "about $0.10 a minute," the number is real — but it's the wrong one to budget with. As industry pricing analysis puts it, "when a cheap AI minute fails and the call goes to a human, you pay for both." That's why a serious campaign quote needs to account for every layer of cost, not just the headline rate.

The four layers behind every campaign quote

A Total Cost of Ownership (TCO) framework breaks campaign costs into four distinct layers. Skip any one of them and your budget will surprise you after launch, not before.

• Platform subscription or management fee — the fixed cost of running the campaign. In AI stacks, this is often the largest single line item, typically $0.05–$0.08 per minute according to cost-stack analysis.
• AI usage charges — telephony ($0.005–$0.022/min), speech-to-text ($0.005–$0.010/min), the LLM itself ($0.005–$0.16/min), and text-to-speech ($0.012–$0.04/min).
• Add-ons — analytics, quality assurance, and compliance features. Real-world examples include Zendesk's QA at $35/agent/month and voice analytics like Contact Lens at $0.015/min, per pricing model research and platform pricing breakdowns.
• Implementation resources — scripting, integration, and setup. Amazon Connect, for instance, adds a $0.005-per-attempt outbound campaigns fee plus a $0.025/min dialer feature charge, as documented pricing shows.

Why the layers matter more than the rate

Hidden fixed fees are where budgets break. A pricing review notes that complicated per-component pricing makes it "difficult for businesses to risk purchasing… when there is no clear view of what size budget to set aside." Compare that with a flat, quoted-before-launch model: My AI Call Center quotes the whole campaign — per-minute rate, setup, and management fee — before anything launches, with no per-seat charges or platform bill.

The comparison stakes are real. In a worked example from containment-rate analysis, AI calling breaks even against onshore human agents ($0.73/min) at just 11% containment, but needs 43% against offshore agents ($0.25/min). Below that threshold, AI makes each call more expensive, not less.

The number that actually matters

That's why cost per resolved call — not per-minute rate — is the metric worth demanding in any quote. Automated reminder systems already demonstrate the gap: €0.14 per contacted patient versus €0.90 for manual calls, a roughly 6.4x differential per healthcare benchmark data. Ask any vendor to show you all four layers, and the full number, before you approve launch.

How Containment Rate Determines Real Campaign Savings

The per-minute rate on a quote tells you almost nothing about whether a campaign saves money. What determines savings is the containment rate — the share of calls the AI completes without a human taking over — and that number changes depending on what you're comparing against.

Consider a worked example from industry analysis of a 40-seat center handling 60,000 calls per month at a 4-minute average. With AI at $0.15 per minute, the break-even containment rate is just 11% against onshore human agents at $0.73 per minute — but 43% against offshore agents at $0.25 per minute. Below that offshore threshold, "AI makes each call more expensive, not less."

The same math applies to manual processes, not just call centers. Healthcare research shows automated reminder systems cost a mean of €0.14 per contacted patient versus €0.90 for manual phone calls — a 6.4x differential that holds regardless of containment, because there is no human fallback to pay for.

This is why the campaign review matters as much as the rate itself. Before launch, you should know your break-even containment against your actual alternative:

• Onshore staff ($0.73/min fully loaded): break-even containment is low, often under 15%, so most structured campaigns clear it easily.
• Offshore staff ($0.25/min): break-even climbs to roughly 43%, so campaign design and list quality decide the outcome.
• Manual processes (Schedulers making reminder calls by hand): the comparison is per-contact cost, where automation already wins by 6x or more.
• No current coverage (after-hours, lapsed databases): the baseline is zero, so anything resolved is pure gain.

Two cautions keep expectations realistic. First, per-minute rates can mislead — as one analysis puts it, "when a cheap AI minute fails and the call goes to a human, you pay for both." If the AI minute rises to $0.30, no containment rate breaks even against offshore staffing. Second, containment gains shrink your team less than the math suggests: an Erlang C example shows 250 calls per hour needing 21 agents drops to only 14 at 50% containment, because remaining calls are longer and harder.

My AI Call Center builds this reality into campaign review: one clear goal per campaign, the full number quoted before launch, and the rate locked so mid-campaign drift cannot erase your margin. At 70% containment in the onshore example, savings reach $92,000 per month — but the honest path there starts with knowing your break-even number first.