Service Cloud Voice Developer Guide

Overview
REST API
Lambda Functions for Amazon Connect
Limits and Scaling
Org Limits
Customer Example
Best Practices for Error Handling
PDF

Newer Version Available

This content describes an older version of this product. View Latest

Service Cloud Voice Customer Example

This example describes how these limits operate in a contact center that uses Service Cloud Voice.

For simplicity, the example discusses only inbound calls, but limits apply to both inbound and outbound calls. Assume that the call center’s call volume is approximately 200 calls per 15 minutes. Suppose the following scenario where an important sports game is starting at 8:00 pm.

  • At 8:03 pm, the internet service provider has a technical issue and service goes down. Customers can’t see the big game.
  • At 8:04 pm, the call center is flooded with calls from upset customers. The number of calls spikes to approximately 1,000 calls in that minute.
  • Assuming that there were no other calls in progress before 8:04 pm, Voice can handle all of the incoming calls because the number of calls is exactly at the 1,000 concurrent call limit.
  • However, if the system detects that more than 50 calls are made in a single second, then the remaining calls will fail. The failed calls are relayed back to the AWS Lambda function that defines the interactive voice response (IVR).

Suppose that in the three seconds between 8:04:01 pm and 8:04:03 pm, there were 210 incoming calls. The 50 call/ second limit takes effect, so the first 50 calls per second are accepted and the second 50 calls fail as follows:

  • 8:03:01 - 60 calls come in. 50 calls are accepted; 10 calls fail.
  • 8:03:02 - 50 calls come in. 50 calls are accepted.
  • 8:03:03 - 100 calls come in. 50 calls are accepted; 50 calls fail.

The failed calls are written to the contact attribute for the Lambda function.