Why Most Lead Volume Estimates Are Guesses

Most businesses pick a lead target the way they pick a lunch spot: it sounds about right, so they go with it. Then the campaign misses, the budget blows, and nobody can explain why — because there was never any math behind the number in the first place.

The problem isn't effort. It's missing inputs. A defensible lead volume estimate needs three numbers most teams never write down: your conversion rate, your average handle time, and your true cost per connected minute. Skip any one of them, and the target is a guess wearing a spreadsheet.

The third input trips people up most. A headline rate like "$0.05/min" rarely means what it appears to mean. According to pricing analysis across AI call center vendors, an advertised "$0.05/min" often covers only the platform layer, leaving you to pay separately for everything else. The real cost stack includes:

• A platform or orchestration fee (e.g., $0.04–$0.05/min on BYOK platforms)
• LLM inference, ranging from $0.003 to $0.080 per minute depending on the model
• Text-to-speech, from $0.015 up to $0.040 per minute for premium voices
• Telephony charges, up to $0.015/min unless you bring your own SIP

Stack those layers and the picture changes fast. Fully loaded AI voice agent pricing typically runs $0.07 to $0.19 per minute once LLM, TTS, telephony, and voice infrastructure are included — not the nickel on the brochure. Per-minute rates across the market actually span $0.05 to over $1.50 depending on whether you're buying raw infrastructure or a managed service. If your forecast was built on the brochure number, your forecast is wrong before the first call dials.

There's a second, subtler trap: not all dialing time costs money. With most AI calling platforms, billing begins when the call connects — ringing, busy signals, voicemail detection, and unanswered calls cost $0. That's good news for your budget, but it means your model needs a connection rate, not just a lead count. If only a third of dials answer, as happened with the IRS's 2025 tax season calls where just 32% of 38.6 million calls were answered, your required dial volume is roughly triple your target conversation volume.

This is why My AI Call Center quotes the full campaign — setup, management, and a locked per-minute rate — before anything launches. The math only works when every layer of cost is on the table from day one.

Work Backward: The Lead Volume Formula

To calculate the number of leads needed, start with your conversion goal and work backward through your funnel. For example, if your target is 50 qualified appointments and your historical conversion rate from connected conversation to appointment is 25%, you’ll need 200 connected conversations to reach that goal (50 ÷ 0.25). This backward approach ensures your lead volume aligns with realistic conversion expectations rather than guesswork.

Next, factor in your connection rate — the percentage of dials that result in a live conversation. If only 32% of calls are answered, as seen in IRS 2025 tax season data where 12.4 million of 38.6 million calls were answered by employees, you’ll need to dial approximately 625 numbers to get those 200 connected conversations (200 ÷ 0.32). Remember, billing starts only when a call connects and audio begins — ringing, voicemail, and no-answers cost $0, so only connected conversation time accumulates charges.

Finally, multiply your required connected conversations by your average handle time to determine total billable minutes. If each qualified conversation averages 4 minutes, you’ll need 800 connected minutes (200 × 4). With My AI Call Center’s per-minute pricing starting at 9¢ for connected time, your total cost would be $72 (800 × $0.09). This method lets you forecast spend accurately while ensuring your outreach volume matches your conversion objectives.

• Define your conversion goal (e.g., 50 qualified appointments)
• Divide by expected conversion rate to get required connected conversations
• Divide by connection rate to get total dials needed
• Multiply connected conversations by average handle time for total billable minutes
• Apply per-minute rate only to connected time — ringing and voicemail cost $0

What Per-Minute Pricing Does to Your Numbers

The cheapest call in your budget is the one that only bills when someone actually picks up. That single rule changes how you calculate leads more than any other variable in the formula.

AI voice agents typically run $0.07–$0.15 per productive minute, compared to $1.33–$2.73 for US human agents and $0.38–$0.82 for offshore agents, according to a detailed pricing breakdown. The same source translates that into per-call terms: a routine 4-minute call costs $0.28–$0.60 with AI versus $3.00–$7.00 with a US agent. In practice, the same outreach budget reaches roughly 10x more leads.

Billing transparency matters as much as the rate. With per-minute models, billing begins when the call connects and audio starts — ringing, busy signals, voicemail detection, and unanswered calls cost $0. So your cost model counts only connected conversation time, not dials. If your connection rate resembles the IRS's 2025 tax season, where only 32% of 38.6 million calls were answered, your dial-to-connect ratio becomes a key input, per market research.

Comparing rates across service models requires looking at the full cost stack, not headline numbers. As one pricing analysis notes, a "$0.05/min" ad may cover only the platform layer while you pay separately for speech-to-text, LLM, text-to-speech, and telephony. Typical all-in ranges:

• BYOK / infrastructure: $0.05–$0.15/min, but you combine several provider bills yourself
• Managed / all-in-one: $0.25–$0.50/min, bundled and easier to forecast
• Business-grade / enterprise: $0.50–$1.50/min, with compliance and integration support built in

Here is the part most teams miss: your early per-minute data is not just a bill — it is the raw material for better pricing later. The conversion rate, handle time, and cost data from per-minute usage become the inputs to an outcome price, such as the $1–$5 per resolution models described in the same breakdown. Run your first campaigns per-minute, capture real disposition data, and you can negotiate outcome-based terms with actual numbers instead of estimates.

This is why My AI Call Center quotes campaigns per-minute — starting at 9¢ per connected minute, agreed before launch — before anyone commits to longer-term structures. Per-minute pricing is the right model when call volume is variable and use cases are still being defined. Treat your first campaign as a measurement exercise, and every later calculation gets easier.

Turning the Math Into a Campaign Plan

A formula on a whiteboard is worth nothing until it survives contact with real calls. The math you did in the last section — conversion rate, handle time, cost per minute — only becomes useful when it drives an actual campaign plan with real data flowing back in.

Start by defining one clear goal per campaign. A campaign that tries to qualify, remind, and survey in the same call produces muddy numbers for every calculation you make afterward. Scope each campaign around a single outcome, so every disposition code maps cleanly to one step in your funnel.

Before you spend a dollar, review your list source and consent records. This is where most ROI calculations quietly fail: a list without clear permission means abandoned calls, opt-outs, and compliance risk that erode your effective conversion rate. My AI Call Center checks list source and consent records before any campaign launches — and tells you plainly if the list won't support the campaign, before you spend anything. The cost math also works in your favor here: with per-minute billing, ringing, busy signals, and unanswered calls typically cost $0, since billing begins only when the call connects and audio starts.

Then run a first campaign to collect real numbers instead of assuming benchmarks. Industry benchmarks are a starting point, not a substitute — for context, only 32% of calls to the IRS in the 2025 tax season were actually answered, a connection-rate reality that can swing your lead volume math dramatically. Your first campaign should be sized to answer three questions:

• What is our real connection rate, and therefore how many dials do we need per conversation?
• What is our actual average handle time per connected call?
• What percentage of connected calls convert to the campaign's single goal?

This is exactly how outcome-based economics are supposed to work: as pricing analysts note, the conversion rate, handle time, and cost data from your per-minute usage become the inputs to your next, more accurate calculation. Each campaign's disposition report — confirmed, qualified, renewed, opted out, no answer — replaces assumptions with measured rates.

When your first campaign is quoted before launch at a locked rate from 9¢ per connected minute, with no per-seat charges or surprise platform bills, your cost inputs stay fixed while your conversion inputs get sharper. For reference, AI voice agents typically run $0.07–$0.15 per productive minute versus $1.33–$2.73 for US human agents, so the budget side of your equation is far more predictable than the demand side.

Run the first campaign, read the disposition report, and recalculate. Your second lead-volume estimate will be built on your own data — and that's when the math starts paying for itself.

A Simple Checklist Before You Launch

Before launching your campaign, a few key checks ensure your lead volume calculation stays grounded in reality. Start by confirming your conversion goal in writing—whether it's qualified leads, appointments, or survey completions—so your team and vendor share the same target. Next, estimate your connection rate conservatively; industry data shows only about 32% of outbound calls are answered by a live person, meaning you’ll need to dial significantly more numbers to reach your connected-minute target. Verify the full cost stack, not just the headline per-minute rate, as My AI Call Center’s pricing starts at 9¢ per connected minute but may include setup or management fees that impact total spend. Check consent records on your list to confirm every contact has approved, permissioned, or reviewed status—this is non-negotiable for compliance and campaign approval. Finally, plan a review point after the first campaign to replace your estimates with actuals: compare projected vs. actual connection rates, handle times, and conversion rates to refine your model for the next wave. Once these steps are complete, you’ll have the clarity to move forward confidently. Plan your campaign and get the full number before approving launch.