02-19-2013 03:43 PM - edited 03-14-2019 11:16 AM
We have a group of reps that receive inbound calls but also make outbound calls. To make the outbound calls they first must go Not Ready selecting a Not Ready reason in CTIOS then make their Outbound calls. After the call they return to a ready status. We are using the "LAA" node to select the Longest Available Agent in ICM. What we found is the following: Agent A and Agent B receive calls, Agent A should be next. Agent A goes not ready, makes a call, then returns to Ready. Next call is delivered to Agent B since technically they have been "Available" the longest amount of time. We want to exclude any Not Ready time from Agent A so they get the call next. Essentially we want the "Longest Agent who has not received an ACD call" type formula.
Anyone have a way we could get that functionality in an LAA node using maybe a custom formula or one of the standard settings in the drop down box? None of the standard ones seem they would work.
Thanks, Bill (Medtronic).
02-19-2013 10:52 PM
Hi,
I am not aware of any standard way of doing this. What I would recommend you to consider: implement agent based routing (instead of the current skill group based one). You can actually tell ICM which rep to pick, using the AgentToAgent node. How you inject this information: well, it can be a VRU script, a database lookup, or an Application Gateway - the last being the fastest.
Example:
Call comes in. Hits the IVR. Customer input is collected, and assigned to a PeripheralVariable or an ECC Variable. The variable is sent to an Application Gateway. Application Gateway checks for the realtime availability of the agents and also the longest idle time (but this time a different calculation is used). Application Gateway sends back an ordered list of agent ID's to ICM. ICM tries to deliver the call to the agents on the list.
I know this is not a place for an advertisement, but anyway, our company has developed an Application Gateway scripting engine, and it could be a perfect solution. Drop me a private message if you are interested.
G.
02-21-2013 02:16 PM
I'm having a hard time finding functionality like what you're looking for built into ICM. Manually building something for this is possible, but nontrivial... If you're using CVP, you could potentially:
You would of course need to account specially for the situations where there are no available agents, or where CVP gives you a peripheral ID and that agent is no longer ready by the time you try to give them the call.
Why wouldn't you want to route to the longest available? Are you trying to fairly distribute inbound calls to agents? It seems like in this scenario you're trying to give more work to the agents performing more outbound calls.
02-21-2013 02:51 PM
Thanks for your inputs, I like the CVP dip into the DB but doubt we will do it, nice idea though.
Here's the real issue we have: CTIOS requires you to be in Not Ready to make an outbound call. These reps are commission based so they not only call customers but serve customers who place inbound calls from various marketing campaigns the company runs. So when they go Not Ready their "Available" time goes to zero and doesn't start timing it again until they go back to Ready.
Now, when an inbound call comes in, the system will send the call to the longest available agent vs the longest agent who has received an ACD call, what I wish is that Not Ready time would not stop the "available" timer but there is no way around that either.
For example:
Aent A takes a call at 2 PM.
Agent B takes a call at 2:05 PM
Agent A finishes call and goes Not Ready at 2:10 PM.
Agent B finishes call at 2:15 PM but stays Ready
Agent A goes back ready at 2:30 PM
New call arrives at 2:31 PM, it will go to Agent B because they are the Longest Available agent, even though Agent A hasn't received an ACD in the longest amount of time.
Hope that makes sense, just want to include Not Ready time as Available time to provide a different routing strategy, to not penalize agents making outbound calls and fairly distribute the calls, again this is all because this group work on commision so fair distribution is desired.
I think the only solution is a custom one, track it in a database or whatever, I can write something in Studio if need be after we get back to the business group requesting it.
Thanks again, Bill.
02-22-2013 01:06 AM
Hi,
did you explore the possibility of enabling Outbound Option (aka "Blended Agent" aaka "Auto Dialer") yet?
Logically, an agent should not go Not Ready unless there is a good reason for it. Going not ready for a sake of an outbound call does not sound like a good reason to me. I mean, if an agent is Not Ready, it usually means he/she's not able to do what he/she's paid for. And as you have explained above, the agent actually tells the system "Hey, I am going to be away for a while" - but in the meantime, they actually continue working.
Of course, OO would not prevent agents from entering Not Ready any time they want, but I think you would get more consistent reporting, since all calls, inbound or outbound will be counted, including their times. And if you choose to implement Predictive or Progressive dialling mode, the agents would be given only live calls, answering machines, fax machines, robots, aliens would be filtered out.
G.
02-22-2013 07:21 AM
How much variance is there in your inbound AHT? If your goal is to evenly distribute time on Inbound calls, or number of Inbound calls, you have a couple options built in. Next Available Agent selection should give you a round-robin call distribution. You could also potentially use the Route Select node, with a Target type of Agent and "Distribute among targets" selected.
Discover and save your favorite ideas. Come back to expert answers, step-by-step guides, recent topics, and more.
New here? Get started with these tips. How to use Community New member guide