The pricing model for an answering service matters more than most contractor owners realize when they sign up. Two services with similar published rates can produce wildly different monthly bills depending on whether the model is per-minute or flat-rate-with-overage. This post walks through which model saves contractors money in three real-world scenarios.
How each model actually works
Per-minute model:
- Base monthly fee (often $250-$350/mo).
- Every minute of every call counts.
- Most vendors round up to the nearest minute; some round to 6 seconds.
- Overage rates: typically $1.85-$2.00/minute for trades-friendly live services.
Flat-rate model:
- Base monthly fee tied to an included ceiling of calls or minutes.
- Usage above the ceiling is billed at a published overage rate, per call or per minute.
- Common in AI receptionists: OnCrew is $49 (150 minutes) / $149 (500 min) / $349 (1,500 min), then $0.49 per minute on Starter, $0.39 on Pro, $0.29 on Multi-Truck.
Hybrid: Some live services have a flat base + per-call rate after a small included pool (e.g., $300 base / 30 included / $11.50 per-call overage at Smith.ai live tier).
The three scenarios
Scenario 1: Solo HVAC tech, 60 calls/month, 4 min avg
| Model | Calculation | Monthly cost |
|---|---|---|
| Per-minute (Ruby) | $250 + (60 × 4 × $1.85) | $694 |
| Flat-rate base + minutes (OnCrew $49 plan, 150 minutes included) | $49 + (90 × $0.49) | $93.10 |
| Flat-rate (Goodcall) | $79 + minimal | ~$80 |
| Per-call live (Smith.ai) | $300 + (30 × $11.50) | $645 |
At low volume, flat-rate AI is decisively cheaper. Per-minute live is paying for capacity you don't use.
Scenario 2: Mid-size plumbing shop, 300 calls/month, 5 min avg
| Model | Calculation | Monthly cost |
|---|---|---|
| Per-minute (Ruby) | $250 + (300 × 5 × $1.85) | $3,025 |
| Flat-rate base + minutes (OnCrew $349 plan, 1,500 minutes included) | 1,500 minutes on 1,500 included | $349 |
| Per-call live (Smith.ai live) | 300-call plan, all calls included (rechecked 2026-09-22) | $2,100 |
| Per-minute (PATLive) | 1,000-minute plan: $1,499 + (500 × $1.69), rechecked 2026-09-22 | $2,344 |
At mid volume with mid-duration calls, the gap is still large: OnCrew's minute plan costs $1,339 against $2,100 to $3,025 for the live options, roughly 36% to 56% less.
Scenario 3: Storm-season roofer, 600 calls in a surge month, 4.5 min avg
| Model | Calculation | Monthly cost |
|---|---|---|
| Per-minute (Ruby) | $250 + (600 × 4.5 × $1.85) | $5,245 |
| Flat-rate base + minutes (OnCrew $349 plan, 1,500 minutes included) | 2,700 minutes on 1,500 included | $697 |
| Per-call live (Smith.ai live) | $2,100 + (300 × $8.50) | $4,650 |
Surge month is where the categories diverge most violently. OnCrew's $349 plan runs $697 at $0.29 a minute past its 1,500 included minutes; per-minute live runs $5,000+.
When per-minute actually wins
Honest scenarios where per-minute live is the better economic choice:
- Very low call volume with very short call durations. Under 30 calls/month, 1-2 minute calls each. The per-minute math is tiny; the per-month flat fee on AI is the dominant cost.
- Brand-experience differentiator. If a polished live voice on every call is a marketing asset that produces meaningful conversion lift, the cost difference can be justified.
- Outbound work bundled. Live services often handle outbound qualifying, voicemail returns, and CRM updates. AI services don't bundle outbound. If you'd otherwise hire a part-time person for outbound, the live service replaces that headcount.
When flat-rate AI saves money
Almost every other scenario:
- Mid-to-high call volume. Anything above 100 monthly calls favors flat-rate dramatically.
- Call durations 3+ minutes. Live per-minute billing accelerates with intake-heavy calls, and $1.85 a minute runs up faster than $0.29 to $0.49 a minute past an AI plan's included minutes.
- Peak-week surges. AI overage at OnCrew is $0.29 to $0.49 a minute; per-minute live overage at $1.85-$2.00 a minute scales faster with the same call length.
- Configured intake schemas. AI runs the schema consistently; per-minute live agents on a generic script don't capture the same fields.
- Forecastable monthly cost. A flat base with a published overage rate makes budgeting predictable; per-minute live bills every minute at a higher rate.
The rounding-rule trap
Most per-minute live services round up to the nearest minute on every call. A 31-second call becomes a 1-minute call. A 61-second call becomes 2 minutes. Over 300 calls, the rounding adds 10-15% to the bill.
Some services round to 6 seconds, which is much friendlier at scale; OnCrew adds up the month's answered-call time and rounds up once per billing period. Confirm which your vendor uses before signing.
The "unlimited" trap
Flat-rate plans marketed as "unlimited" usually have an AUP fair-use cap. Read the cap. If your peak month could brush it, you're back to per-call uncertainty, just hidden in legal language instead of pricing.
Honest published included-call or included-minute pricing publishes the overage rate. "Unlimited" with no published cap is marketing.
A simple decision rule
Run this calculation:
- Your normal monthly call count × your average call duration = monthly minutes.
- At $1.85/min, that's per-minute cost.
- Compare against the flat-rate plan's base plus any overage at that volume (for OnCrew, the minutes past 150, 500, or 1,500 at $0.49 on Starter, $0.39 on Pro, $0.29 on Multi-Truck).
If flat-rate is more than 30% cheaper at your normal volume, switching saves money even before considering surge weeks. If flat-rate is within 10% at normal volume, the surge math is what tips the decision.
For more, see the contractor answering service cost benchmark, the best answering service for contractors 2026 guide, and the contractor missed-call cost benchmark.
FAQs
Why do live services use per-minute pricing?
Because live agents are paid by the hour. Per-minute pricing passes the labor cost directly to the customer. It's a structurally honest model for live services even when it's economically uncomfortable.
Why do AI services use flat-rate?
Because AI cost-per-call is much lower than live agent cost-per-call, and the flat-rate model lets vendors offer predictable pricing while building margin. It's the more customer-friendly model when the underlying cost structure supports it.
What if I want both?
The blended pattern works: live service for daytime relationship calls, flat-rate AI for after-hours and overflow. The forwarding rule is time-of-day.
Can I switch mid-contract?
Depends on the contract. Month-to-month plans are switchable any time. Annual contracts have cancellation terms. Confirm before signing.