Greetings.  Has anyone successfully created a reasonably accurate EWT figure using Architect actions?  I've tested with both queue-based and interaction-based EWTs and they don't accurately reflect the customer's actual wait time from the time the call is queued.

I think that ASA for the interval may be a closer metric, but I am not sure where I can find it to use in an in-queue flow Play Audio action. I am definitely open to better suggestions as well.

Hello Vikki, 

Our Development team has a great write up of what factors affect the EWT, but the formula can be broken down with the following equation:

EWT = PredictedAHT * PositionInQueue / NumberOfAgents

I'm not sure if my colleague @Samuel Jillard has any additional information but I figured I could @ them so see if they have any additional thoughts.

Greetings.  Has anyone successfully created a reasonably accurate EWT figure using Architect actions?  I've tested with both queue-based and interaction-based EWTs and they don't accurately reflect the customer's actual wait time from the time the call is queued.

I think that ASA for the interval may be a closer metric, but I am not sure where I can find it to use in an in-queue flow Play Audio action. I am definitely open to better suggestions as well.

Hello Vikki, 

Our Development team has a great write up of what factors affect the EWT, but the formula can be broken down with the following equation:

EWT = PredictedAHT * PositionInQueue / NumberOfAgents

I'm not sure if my colleague @Samuel Jillard has any additional information but I figured I could @ them so see if they have any additional thoughts.

Greetings.  Has anyone successfully created a reasonably accurate EWT figure using Architect actions?  I've tested with both queue-based and interaction-based EWTs and they don't accurately reflect the customer's actual wait time from the time the call is queued.

I think that ASA for the interval may be a closer metric, but I am not sure where I can find it to use in an in-queue flow Play Audio action. I am definitely open to better suggestions as well.

Whilst all those calls are waiting in queue, whats the handle time of the all the agents?  If they were all theoretically close to completing their calls based off their AHT, your staff could answer those interactions sooner as a group and therefore Genesys would suggest that the time waiting is 2 minutes.  

Hello Vikki, 

Our Development team has a great write up of what factors affect the EWT, but the formula can be broken down with the following equation:

EWT = PredictedAHT * PositionInQueue / NumberOfAgents

I'm not sure if my colleague @Samuel Jillard has any additional information but I figured I could @ them so see if they have any additional thoughts.

Greetings.  Has anyone successfully created a reasonably accurate EWT figure using Architect actions?  I've tested with both queue-based and interaction-based EWTs and they don't accurately reflect the customer's actual wait time from the time the call is queued.

I think that ASA for the interval may be a closer metric, but I am not sure where I can find it to use in an in-queue flow Play Audio action. I am definitely open to better suggestions as well.

Hello Vikki, 

Our Development team has a great write up of what factors affect the EWT, but the formula can be broken down with the following equation:

EWT = PredictedAHT * PositionInQueue / NumberOfAgents

I'm not sure if my colleague @Samuel Jillard has any additional information but I figured I could @ them so see if they have any additional thoughts.

Greetings.  Has anyone successfully created a reasonably accurate EWT figure using Architect actions?  I've tested with both queue-based and interaction-based EWTs and they don't accurately reflect the customer's actual wait time from the time the call is queued.

I think that ASA for the interval may be a closer metric, but I am not sure where I can find it to use in an in-queue flow Play Audio action. I am definitely open to better suggestions as well.

Hi Vikki

As the EWT is relying on cached data of recently completed calls, testing the EWT actions can be tricky.  A way to test it would be using the Estimated Wait Time API whilst viewing calls as they come into a live queue and then verifying how long they waited against the EWT value returned from the API.

If you have consistently tested this in a live queue and the EWT has been inaccurate, I would raise a case with Product Support to investigate what is causing the values to be out.

As for playing the ASA metric via a Play Audio action, I can't currently think of a way to implement this, although maybe others can. If you think this would be a useful feature, I would recommend raising an idea on the Genesys Cloud Product Ideas Lab.

Hello Vikki, 

Our Development team has a great write up of what factors affect the EWT, but the formula can be broken down with the following equation:

EWT = PredictedAHT * PositionInQueue / NumberOfAgents

I'm not sure if my colleague @Samuel Jillard has any additional information but I figured I could @ them so see if they have any additional thoughts.

Greetings.  Has anyone successfully created a reasonably accurate EWT figure using Architect actions?  I've tested with both queue-based and interaction-based EWTs and they don't accurately reflect the customer's actual wait time from the time the call is queued.

I think that ASA for the interval may be a closer metric, but I am not sure where I can find it to use in an in-queue flow Play Audio action. I am definitely open to better suggestions as well.