Recent searches
No recent searches
Average Chat Concurrency - How is this Calculating?
Posted Mar 14, 2022
In both the out of the box, and custom built chat concurrency dashboards don't show a chat concurrency which makes sense.
In the out of the box dashboard, I see a max concurrency of 2-4 depending on the hour, and an average chat concurrency of 0.
When recreating manually, I've been able to drill in and see that the "0" on these average concurrency queries is actually a very low decimal point (think .13 chats per hour).
So....how is the average chat concurrency being calculated? The metric definition states that it only takes into consideration agents which engaged w/ a chat during that time frame, so the minimum possible concurrency should be 1...
See below for screenshots from the pre-built dashboard on chat metrics:
7
15 comments
Luise Kroll
Hi,
same question here.
And it would be nice to have a metric showing the number of Concurrent Chats (not only min, max and average).
Or is there a way/formula to calculate this within the Dataset Chat Engagement?
Cheers
Luise
1
CJ Johnson
I've had a ticket open for multiple days with this question with support still waiting for first reply. I cannot figure out how on earth any of these calculations are being determined, either.
1
Luise Kroll
Thanks for the information, I was wondering if opening a Ticket was a better idea ;)
Would you be so kind to keep me posted?
0
Hannes Merisaari
This community never fails in this sense. I have the exactly the same question, so it was clear that someone had asked already. Following closely with high interest.
0
CJ Johnson
The response I've gotten from Support, is that average concurrency is broken. "We currently have a problem ticket addressing the issue with the calculation for Average chat concurrency."
0
Justin H
It is true that we have a ticket open with our Dev team to further investigate this issue. Please keep an eye on our Product Announcement page for a resolution announcement for this: https://support.zendesk.com/hc/en-us/sections/4405298833818-Announcements
-2
CJ Johnson
The latest update I received:
1
Andrew Lundy
Hello, has a fix for this been deployed or can we assume the data is still incorrect?
0
Dave Dyson
The dataset mentioned in CJ's comment is still under development.
-2
Milton Lobo
Is there any news yet on the new dataset?
Also, can someone clarify - is it all metrics that are inaccurate or just avg?
1
Brian Emond
Hey!
We're looking to measure this as well. Is there any update on whether or not this Dataset is accurate? It appears to be meaningless, but I can't determine for sure.
0
Wojciech Smajda
Hi,
I see there’s some confusion regarding how the average chat concurrency is calculated, so let me provide some clarification.
The 'average chat concurrency' metric is designed to measure the average number of simultaneous chats an agent is involved in during a given hour. Here's how it’s typically calculated:
In essence, if an agent is engaged in one chat for the entire hour, the average concurrency equals 1. If they're engaged in two chats for the whole hour, it would be 2, and so on.
When an agent participates in chats that don’t span the entire hour, or when there are gaps with no chats, it's possible for the average to be a fraction. For example, if an agent handles two chats but each one lasts only 15 minutes within the hour, the concurrency average might look something like 0.5 for that hour because the agent is not concurrently chatting for the entire time frame.
The 'average chat concurrency' isn't simply a headcount of all chats within the hour but a representation of how much of that hour was spent in active chat based on engagement duration. This is why you may observe figures like 0.13 in your dashboard, which suggests that, on average, agents are not simultaneously engaged in chats for much of the hour.
I hope this helps in understanding the concurrency calculations in your dashboard. If you have any more questions or need further clarification, please don't hesitate to ask.
-1
Agnese Ahuna
Hi!
Would this mean that if for example an agent does 6 chats (engagement time for each is 15 min) in the first 45 minutes and then in the last 15 minutes of the hour goes on a break, the concurrency would be 6x15/60 = 1.5 chats instead of 2 chats only because they weren't active the whole hour? Or does it limit to time agents are “online” and ready to take new chats?
1
Aine Brennan
Hi Wojciech , this was really helpful but it would also be great to get an answer to the questions above
1
Annie Baker
Wojciech Smajda, I would also like to know the answer to the question that Agnese posted.
I would also like to know if there are any plans from Zendesk to further expand data in the concurrent chat dataset?
For example, when forecasting, it would be beneficial to see how many concurrent chats our support and customer service groups can average. We would also need this metric to be based on agents' available time and not just by the hour.
0