Skip to main content
Voice AI Pods and Provider Connections

Split Concurrency Between Inbound and Outbound Calls

Decide how many of a Pod's call lines answer incoming calls and how many dial out, and how campaigns share the outbound lines.

Written By voiceaiwrapper

Last updated 26 days ago

Concurrency is how many calls a Pod (an engine or provider connection) can run at the same time. Each Pod has a total number of call lines, and you decide how many of them answer incoming calls and how many dial out. Getting the split right means inbound callers always reach an agent while your outbound campaigns still have room to dial.

Changing the split takes about a minute.

How concurrency works

Every Pod has three numbers:

NumberWhat it means
TotalThe most calls the Pod can run at once. You can't change it on the split screen.
InboundLines kept for incoming calls.
OutboundLines kept for calls your campaigns and agents place.

Inbound and outbound together can't be more than the total. When you connect a Pod, VoiceAIWrapper splits the total evenly. If the total is an odd number, the extra line goes to outbound.

Where the total comes from depends on the engine:

EngineWhere the total comes fromHow to change it
VAW NativeSet by VoiceAIWrapper for your plan. When your plan changes, the total changes with it and your split keeps the same proportions.Contact support.
VapiThe Vapi Concurrency Limit you type when you connect (default 10). Make it match the limit in your Vapi organization.Edit the Pod, change Vapi Concurrency Limit, then set the split again on the next step.
Retell AIRead from your Retell AI account when you connect.Click Refresh next to Concurrency on the Pod's card on the VoiceAI Pods page ("Sync concurrency from provider").
ElevenLabsSet from your ElevenLabs subscription tier when you connect: Free 4, Starter 6, Creator 10, Pro 20, Scale or Business 30.It's set once, when you connect.

When a Retell AI refresh finds a different total, the split resets to an even split. Check your inbound and outbound numbers afterwards.

Change the split

  1. In the header, open the Pod selector (it shows your current Pod's name).
  2. For the Pod you're in, click the pencil icon (Edit). For any other Pod, click the โ‹ฎ menu on its row, then Edit. You can also click the pencil icon in the VoiceAI Pods card on the Account Dashboard.
  3. The Update VoiceAI Pod drawer opens on Update Credentials. Click Next to continue to Set Concurrency. Clicking Next saves the credentials as they are, and on Vapi, Retell AI and ElevenLabs Pods it checks the keys with the provider again. Leave them unchanged unless you mean to update them.
  4. Type a number in Outbound or Inbound. The other field updates automatically so the two always add up to the Total, and the bar underneath shows the split as percentages.
  5. Click Update Concurrency. You'll see "Concurrency distribution updated successfully".
Header Pod selector โ†’ pencil (Edit) on a VAW Native Pod โ†’ Update VoiceAI Pod drawer on the Set Concurrency step. Outbound and Inbound filled in (for example 4 and 6), the read-only Total in the middle, the coloured split bar with percentages, the Website chats field with its description, and the Update Concurrency button.

How to choose a split

  • Mostly inbound (a receptionist or support line): give inbound most of the lines. Lines you give to outbound can't answer incoming calls, even when no campaign is dialing.
  • Mostly outbound (lead follow-up, reminders): give outbound most of the lines, but keep at least one inbound line if leads can call back.
  • Both: start with the even split and adjust after a week of real traffic. The Live operations section of the Pod Dashboard shows the calls running on the Pod right now.

Widget chats on VAW Native

VAW Native Pods have one more field on the Set Concurrency step: Website chats. It's the number of chats your website widgets can have open at once. Chats use no call lines, so this number is separate from the split.

  • Default: 20.
  • Range: 1 to 500.

When the limit is reached, new visitors can't start a chat until an open one ends.

Voice calls started from a website widget are different: they use the Pod's inbound lines, the same as phone calls. So do test calls you make in the browser from an agent's page. Test calls to a phone number use outbound lines.

What happens when every line is busy

SituationWhat happens
Inbound phone call on VAW NativeThe caller hears a busy signal. The call is logged with the end reason Concurrency limit reached.
Inbound phone call on Vapi or Retell AIVoiceAIWrapper doesn't hand the call to your campaign's agent.
Inbound phone call on ElevenLabsElevenLabs connects the call to the agent assigned to the number, so ElevenLabs' own limit applies, not your inbound split.
Voice call from a website widget on VAW NativeThe widget can't start the call while every inbound line is busy.
Browser test call from an agent page on VAW NativeYou see "Concurrency limit reached โ€” end an active call or test and try again."
Test call to a phone number on VAW NativeYou see "All outbound lines are busy right now. Try again in a moment, or raise the Pod's outbound concurrency limit."
Outbound campaignThe campaign waits and dials the next lead when an outbound line frees up.

On VAW Native, Call Logs hide these refused calls by default, and the call count says "missed calls are hidden". To see them, use the Missed calls filter, or filter by the end reason Concurrency limit reached.

Share outbound lines between campaigns with Campaign Priority

All outbound campaigns on a Pod share the Pod's outbound lines. On some accounts, outbound campaigns also have a Campaign Priority (1-10) field that decides how those lines are shared. A campaign with a higher number gets more of the calls: a campaign set to 10 gets about twice as many calls as one set to 5 while both have leads to dial. New campaigns start at 10.

To set it:

  1. Open the campaign and go to Settings.
  2. Open Outbound behavior.
  3. Enter a number from 1 to 10 in Campaign Priority (1-10).
  4. Click Save Settings.
An outbound campaign โ†’ Settings โ†’ Outbound behavior panel, showing the Campaign Priority (1-10) field with a value entered and its description underneath. Use an account where the field is visible.

Priority doesn't reserve lines. A campaign with priority 1 still uses every free line when no other campaign has leads waiting.

If you don't see the field, your outbound campaigns share the Pod's lines without a priority setting.

Frequently asked questions

The total for VAW Native is set by VoiceAIWrapper based on your plan. You can move lines between inbound and outbound, but to change the total, contact support through Need help? โ†’ Help Chat.

Your provider has its own limit. If the total in VoiceAIWrapper is higher than the limit in your provider account, the provider rejects the extra calls. For Vapi, set Vapi Concurrency Limit to the same number as your Vapi organization. For Retell AI, refresh the total from Retell AI (see the table above).

No. Chats are limited by Website chats only. Voice calls from a website widget do use inbound lines.