Far end or near end disconnect?

[picklemogg]

Hey all- any ideas on this one? We have multiples of Option81 5.0 and CCMS 6.0. We often get reports that CCMS agents are hanging up on customers and pegging as short calls in CCMS. We know that agents are doing this, but does anyone know of a tool that can be used to tell 100% whether the agent is initiating the disconnect vs the customer (far vs near end)? I'm also going to post this in the CCMS forum, but wanted to also post in here. We cannot tell any differences when running an entc on the PBX. Thanks for your help.

Moggs

 

[bigindian65]

no recording package? If you can one that is habitual, might hv to just OBV it from a sup set for a period of time, which is time consuming, but Technology will not replace or administor Human Resources or management.

Mato' Was'aka

 

[PERRYPJ]

Don't you run reports, look for high volume of calls taken and short talk times. This will place you in the ball park

OLD ROLMEN WORKING ON NORTELS AND AVAYA

 

[Tman45]

Yeah I was going to say you should be able to get pretty close by looking at reports - then do a call by call to get more detail.

 

[picklemogg]

Thanks for the feedback thus far, however I guess I must not have been clear on my question.

I can run reports until I'm blue in the face and I have. I can as an example see that a particular agent has let's say 50 short calls (or in our case 50 calls that lasted less than 10 seconds). Of the 50 calls I have no idea the amount of these calls that were disconnected by the agent vs by the customer. I would speculate that a high %, if not all, were terminated by the agent to avoid calls, but I need a method to prove this.

So my question is: is there a tool that can be used to determine if a disconnected call was terminated by the agent vs customer?

Thanks again.

Moggs

 

[bigindian65]

Yes. A Call Recording package.

Mato' Was'aka

 

[picklemogg]

BigIndian65- Thank you for the response, but could you please expand? Is there a particular call recording solution that produces this info? We do have a call recording solution in our environment(NICE)and we cannot confirm this info.

We capture both the call audio and screen data. NICE knows to start/stop recording based on CTI events sent from the PBX across the MLSM. The disconnect CTI event (in the CTI logs) does not ditinguish between a far/near end disconnect. It just recieves the same end call event from the PBX either way. So we cannot determine this from the CTI log portion of our recording system. If I would go ahead and play the call audio back I might hear (we use Nortel agent greeting in our environment not that it really matters, but just throwing it out there), "Thank you for calling Customer Service, this is Frank how can I help you?" As soon as Frank's greeting ends, the call recording stops indicating end of call. There is no way I can tell in this example if Frank initiated the end of the call or if the customer did.

Please let me know if there is something more than what I am thinking here in regards to a call recording solution. Like I said, maybe a different solution may have additional capabilities. Thanks.

Moggs

 

[bigindian65]

I see your problem a little clearer now, I am surprised CTI doesn't distinguish. I wonder would Dch messaging show far end or near end?? I can't remember? Be a lot of traffic to watch though, but might work.

Mato' Was'aka

 

[bigindian65]

hey that should work if it is on a PRI and you turn on both MSGI and MSGO, you will see the call tear down request from one or the other if its from MSGO, then you know it is the agent. (unless you have some serious telco issues going on)

Mato' Was'aka

 

[PERRYPJ]

If you have it narrowed down to one agent, I would take him aside and explain the facts of life to him or her.

OLD ROLMEN WORKING ON NORTELS AND AVAYA

 

[Tman45]

In this case I would silent observe/monitor the particular agent/agents that are suspected so you can make the determination - perhaps run a report early in the morning to see if any are occuring and then listen in to see if you still get short calls pegged then you can hear it for yourself. Wonder if Witness can prove this better than NICE? We are trying to decide which one to use and this could be a deciding factor (although I already know I want Witness) - not up to me though

 

[bajangsa]

I agree with PERRYPJ - if one agent has more short calls than the others we have a human error.
Another thing; you might want to set the short calls to 5 sec instead of 10 in order to catch malpractice og wrong calls.
10 seconds is enough time for a conversation such as "can I speak to Peter?" - "he is in a meeting untill 10 oclock