Thank you so much for taking the time to answer some of those questions, I really appreciate it, you guys rock :)
That EWT accuracy reporting would be AMAZING. I didn't see the Idea already so created one under Inbound, if that category needs to change, happy for it to be updated. -
https://genesyscloud.ideas.aha.io/ideas/INB-I-1417Just have a few follow up questions
My Questions around Priorty was less about the priority call, but rather the subsequent calls.
- Because these prioritized calls get cached, and will have a much lower actual wait time, and with only 7 completed calls being looked at, it doesn't take many of these calls for the median call's awt or the lower bound to be much lower than a 'normal' call. So what can we do to avoid that?
- Although priority is the adjust time since queue entry, we also have an API now that can retrieve and adjust the priority of a call inQ, so would it now be possible cache that priority value at the start of the call in queue and use that to either only include similar priority calls or adjust the AHT further based on the difference in priority and Actual Wait Time
And finally.
Still trying to wrap my head around the calculation, so please correct the below if incorrect which I assume must be because I get odd results
Adjusted AHT is Actual wait time of the 4th cached call (median of 7 calls) multiplied by the Agents in Q at the time the 4th cached call entered the queue, all that divided by the position of the 4th cached call when it entered the queue
For:
AWT.M = AWT for the median cached call at time of answer
PIQ.M = PIQ for the median cached call at entry
AG.M = Agents on Queue for the median cached call at entry
PIQ.C = PIQ for the current call
AG.C = Agents on Queue for the current call
Combing AHT into the EWT formulae looks like this then:
(((AWT.M * AG.M)/PIQ.M) * AG.C) / PIQ.C
If I plug the numbers in
Of AWT.M = 100, AG.M = 8, PIQ.M = 5,
Then, AHT = 160 So EWT = (160 * AG.C) /PIQ.C
With PIQ.C = 10, AG.C = 5
Then EWT = 80
So with less agents in queue and a higher position in queue, the EWT is less than the Actual wait and AHT than for a call with more agents and closer to being answered.
And if I change nothing but reduce the Position In Q for the current call to 2, then the AHT doesn't change, but EWT goes up to 400.
This doesn't seem right, so I must have something wrong.
Ultimately we are trying to understand why EWT seems to be inaccurate often enough that we get customer complaints despite rounding up and using a range eg if EWT =11 minutes we say 20 to 30 minutes, and what we can do to improve that accuracy.
------------------------------
Anton Vroon
------------------------------
Original Message:
Sent: 01-27-2022 14:21
From: Matt Lawson
Subject: Genesys Cloud CX Q&A Episode: Estimated Wait Times
When Matt posted AppFoundry's Brick Breaker game, he had no idea how long it would take for someone to break his score or if anyone ever would, but @Anton Vroon answered the call quickly and as his prize Anton got to pick a Q&A Episode dedicated to the topic of his choice, which was: Estimated Wait Times. Matt got connected with two EWT experts and they are here to give a comprehensive introduction to the topic and then answer questions that came in from community members.
Want a shot for your chance to pick a Q&A Show Topic? Keep checking out the Genesys Community for contests. We have one right now to see who can guess what day the Genesys Cloud CX Community will hit 3000 members! Cast your best guess for a shot to win a prize from our upcoming prize wall.
Hope you enjoy the episode!
Video Chapters:
#GenesysCloudCXQAEpisode
------------------------------
Matt Lawson
Genesys - Employees
Online Community Manager
------------------------------