newoaks.ai › Blog › What's the Monthly Cost per Concurrent Voice Call or Per-Minute Charge?

What's the Monthly Cost per Concurrent Voice Call or Per-Minute Charge?

RayJul 15, 2026

A simple answer: most AI voice platforms charge either by usage (roughly $0.10-$0.30 per minute once telephony and model costs are included) or by capacity (a monthly fee tied to how many simultaneous calls you can run). The cheaper option depends less on headline price and more on your monthly minutes, peak concurrent calls, and overage risk.

Understanding Monthly Cost: Per Concurrent Voice Call vs Per-Minute

If you are comparing AI voice agent pricing, the biggest mistake is looking only at the advertised per-minute rate. In practice, your total monthly cost is shaped by four moving parts:

1. Base platform fee or minute bundle

2. Overage rates

3. Telephony costs for inbound/outbound calling

4. Peak concurrency needs during busy periods

For many small and mid-sized teams, minute-based billing is easier to forecast. For high-volume support lines, appointment campaigns, or outbound sales operations, concurrency-based pricing can be more economical because it protects you during traffic spikes.

The Two Main Pricing Models

1. Per-Minute Pricing

With per-minute pricing, you pay for talk time. This may be offered as:

• a prepaid minute bundle
• a monthly included-minute plan
• pure pay-as-you-go usage

Using the current NewOaks AI examples:

• Voice Call Priority: $19 for 90 minutes

- Effective rate: $0.211/minute

- Overage: $0.20/minute

• Voice Call Premium: $89 for 450 minutes

- Effective rate: $0.198/minute

- Overage: $0.20/minute

That spread is small, but it matters. At 450 minutes per month, the larger bundle saves about 1.3 cents per minute versus the smaller one.

2. Concurrency-Based Pricing

Concurrency pricing is based on how many calls can happen at the same time.

If your platform allows 10 concurrent calls, it means 10 callers can actively talk to your AI agent at once. An 11th caller may be queued, rejected, or sent to voicemail, depending on the setup.

This pricing model is common when businesses care about:

• avoiding missed calls during bursts
• supporting contact-center style workloads
• running outbound calling campaigns
• maintaining answer speed SLAs

In these setups, the effective cost is often not obvious until you divide the monthly fee by your actual usage.

The Formula That Actually Decides Which Is Cheaper

To compare plans fairly, use this:

Effective per-minute cost

Total monthly voice spend / total monthly connected minutes

And for concurrency-based plans:

Cost per concurrent line

Monthly platform fee / included concurrent calls

Then pressure-test both numbers against peak demand.

Worked Examples

Example 1: Small local business

A dental office gets about 120 voice minutes per month through an AI receptionist.

If it uses the NewOaks 90-minute option:

• Base: $19
• Extra 30 minutes at $0.20 = $6
• Total: $25/month
• Effective rate: $0.208/minute

If instead it buys the 450-minute plan:

• Total: $89/month
• Effective rate at actual usage: $0.742/minute

In this case, the bigger bundle is clearly the wrong choice unless the office expects usage to rise quickly.

Example 2: Busy clinic with moderate peaks

A clinic uses 600 minutes per month, but only peaks at 3 simultaneous calls.

With NewOaks-style minute pricing:

• 450-minute bundle: $89
• 150 overage minutes at $0.20 = $30
• Total: $119/month
• Effective rate: $0.198/minute

If a competitor offers concurrency-first pricing with a higher monthly floor, that plan may still lose even if it sounds more "enterprise."

Example 3: High-peak call center

A service team handles 3,000 minutes per month, with bursts of 20-30 concurrent calls after marketing sends or weather events.

Here, minute pricing may still be affordable on paper, but the bigger operational risk is capacity. If your platform cannot handle simultaneous load, the real cost shows up as:

• abandoned calls
• lower booking rates
• customer frustration
• higher human overflow costs

This is where paying for reserved concurrency can be rational even if the sticker price is higher.

Telephony Changes the Math More Than Most Buyers Expect

Many buyers compare only the AI layer and forget carrier and phone-number charges. If your setup uses Twilio or another telephony provider, your real bill may include:

• local phone number rental
• inbound call minutes
• outbound call minutes
• SIP or media streaming charges
• recording/transcription storage

For example, Twilio publishes voice call pricing by country and direction and charges vary materially depending on where calls originate and terminate. US local inbound and outbound rates are typically measured in per-minute telecom charges, not bundled into a flat AI fee. See Twilio's pricing pages for current rates and number fees (Twilio Voice pricing, Twilio phone numbers).

Using the current NewOaks examples with Twilio integration:

• 90 minutes: $24/month

- Effective rate: $0.267/minute

• 450 minutes: $94/month

- Effective rate: $0.209/minute

That means telephony can add enough overhead to make a low-usage plan look noticeably more expensive on a per-minute basis.

Real-World Cost Buckets to Check Before You Sign

When evaluating any vendor, ask for these line items separately:

Platform fee

The monthly charge for access to the AI voice system.

Included minutes

How many connected minutes are bundled.

Overage rate

The marginal cost once you exceed the bundle.

Concurrent call limit

Whether the plan caps simultaneous calls.

Telephony pass-through

Carrier costs, phone numbers, SIP, and call routing.

Model or transcription costs

Some vendors pass through speech-to-text, text-to-speech, or LLM usage separately. OpenAI, for example, publishes API pricing separately from telephony, so voice builders often stack model charges on top of carrier charges (OpenAI pricing).

Implementation and support

Onboarding, prompt design, CRM integration, and custom workflows can exceed the recurring usage charge in month one.

Per-Minute vs Concurrent Calls: When Each Wins

Choose per-minute pricing if:

• your call volume is under roughly 500-1,000 minutes/month
• demand is uneven or still being tested
• you do not expect large simultaneous spikes
• you want simpler budgeting
• you are replacing after-hours overflow or a receptionist first

Choose concurrency-based pricing if:

• your business gets bursty traffic
• you run campaigns that create call surges
• missed calls are expensive
• you need guaranteed capacity
• you manage queues like a support desk or contact center

A Better Way to Estimate Your Monthly Bill

Use this 3-step worksheet.

Step 1: Estimate total monthly minutes

Take:

• average calls per day
• multiplied by average call duration
• multiplied by business days per month

Example:

• 18 calls/day
• 3.5 minutes average
• 22 business days

18 x 3.5 x 22 = 1,386 minutes/month

Step 2: Estimate peak concurrency

Look at your busiest 15-minute windows.

If you commonly receive 8 calls within a 10-minute span and average duration is 4 minutes, your concurrency could easily hit 3-5 simultaneous calls, depending on overlap.

If you use Twilio, its usage records and Voice Insights can help analyze actual call patterns after launch (Twilio Voice Insights).

Step 3: Compare three scenarios

For each vendor, model:

• expected usage month
• busy month
• worst-case spike month

This reveals whether a cheap base plan becomes expensive once overages or concurrency upgrades are applied.

What Businesses Often Miss

Cheap per-minute pricing can hide expensive overages

A plan that looks low-cost at 200 minutes can become costly at 900 minutes.

Concurrency limits are operational, not just financial

A low-capacity plan may save money while silently reducing answer rates.

Telephony is not optional math

Even excellent AI pricing can be undermined by carrier fees, international calling, or number rental.

Average usage is less important than peak usage

The average month may fit your plan perfectly, while two high-traffic days create customer-facing failures.

Bottom Line

If you are asking, "What is the monthly cost per concurrent voice call or per-minute charge?" the practical answer is:

• Per-minute AI voice plans often land around $0.20/minute effective cost for smaller bundles once overages and telephony are considered.
• Concurrency-based plans make more sense when peak capacity matters more than raw minute efficiency.
• The right metric is not the advertised price. It is your effective cost per connected minute at your real peak traffic level.

For NewOaks AI specifically, the current published examples favor minute-based budgeting over monthly per-concurrent-call billing. That is a good fit for businesses with moderate or variable call volume, but teams with heavy spikes should still model concurrency separately before choosing a plan.

FAQ

What does an AI phone agent usually cost per minute?

A realistic all-in range is often $0.10 to $0.30 per minute, depending on included minutes, overages, telephony, and whether model costs are bundled or billed separately.

Does NewOaks AI charge per concurrent call?

Based on the pricing examples provided here, no. NewOaks AI is positioned around included-minute bundles and overage pricing, rather than a monthly fee for each concurrent voice line.

Is the larger NewOaks bundle always cheaper?

Not always. It has a lower nominal per-minute rate, but it is only cheaper if you use a substantial share of the included 450 minutes. Low-usage teams may pay far more per actual minute if they overbuy capacity.

Why do vendors publish concurrency limits?

Because simultaneous-call capacity affects service quality. A plan can be inexpensive per minute and still be a poor fit if it cannot handle busy periods without queues, dropped calls, or overflow.

What should I ask a vendor before buying?

Ask for five numbers: monthly fee, included minutes, overage rate, concurrent call cap, and telephony pass-through charges. Without all five, it is impossible to compare plans accurately.

Sources

• Twilio Voice pricing
• Twilio phone numbers pricing
• Twilio Voice Insights documentation
• OpenAI API pricing
• Google Cloud Speech-to-Text pricing

References

• https://www.newoaks.ai/pricing?via=special
• https://www.ringg.ai/pricing
• https://elevenlabs.io/pricing/agents
• https://www.reddit.com/r/AiForSmallBusiness/comments/1ruv0sb/what_does_an_ai_phone_agent_actually_cost_a_small
• https://www.reddit.com/r/SaaS/comments/1toak7k/voice_ai_is_the_rare_b2b_saas_category_where_you
• https://www.reddit.com/r/AiForSmallBusiness/comments/1ruv0sb/%E2%80%A6