Hi Team,

I have a scenario where I want to create a phantom call to a queue on behalf of a real call and get expected wait time periodically. While this phantom call is waiting in the queue, a new real inbound call arrives then I wan to join this real call with the phantom call waiting in the queue based on some criteria e.g. CLI?

How can I achieve this using TSAPI? Is there a better way to achieve this functionality?

What licensing is involved if any?

First of all phantom calls require an advanced TSAPI license.

So you would have this phantom call be the "expected wait time" solely based on how long it was in queue? I would suggest taking time stamps of each call that comes in/goes out of queue and base your expected wait time on that.


I don't know of a way to join two calls in vector processing (this is my answer for all three of your other questions). I'm not quite sure what you are trying to accomplish by doing that so you will need to explain further, and I can tell you if there is a way to accomplish it.

Hi Craig, thanks for your reply.

Its like a "secretary" call. I want to retain queue position on behalf of an actual caller whose phone number i get from a databse and want to inform the actual caller (through other non-telephony means) to call-in when his call is about to be answered. This is different than avaya call back application.

The CTI application should receive expected answer time or queue position (or any other acceptable criteria) every x seconds so that it can inform the actual caller (through other non-telephony means) to call in.

Does it make sense or am I talking out of Avaya or Telephony world?

There is probably more than one way to go about this. You will definitely want to have a lab to setup a proof of concept for this scenario. Here are my thoughts:

1. A caller must notify your application by some means that they want to call in
2. Make a phantom call (Put the callers CLI in the userInfo, aka UUI) to the queue to hold the place in line
3. Your application is keeping track of the metrics (monitoring the VDN), or using vustats with DMCC to know when to notify the caller to call in
4. Your application notifies the caller to call in to a specific number
5. That number goes to an adjunct route vector. Here your application holds on to the call until the phantom call is sent to an agent
6. When the phantom call is transferred to an agent the caller is sent via direct agent calling to the agent and the phantom call is disconnected. *This part would definitely benefit from some proof of concept testing. *You put the callers CLI in the UUI now you need to use it to make sure the correct caller is sent to the agent. Most likely will need to write it to a database with agent ext. Then the adjunct routing application would have to poll this database for the agent extension using the calling number.


Number 3: To keep track of the queue metrics you would place a monitor on the VDN and keep track of each call when it gets queued and leaves the queue. From that you will know how many are in the queue and you can calculate EWT. The following thread has information on using vustats: http://www.devconnectprogram.com/forums/posts/list/6868.page#p32068. You can use your metrics to notify the caller of what they need to know at any given point.

Also, if the caller notifies you that they want to call in and there is on 1 or two callers in queue I would imagine you have them just call in and bypass the phantom call altogether.

There are definitely other questions that will arise out of this. If you would like further in depth assistance from DevConnect, including testing some of these steps, please open a DevConnect technical support ticket. If you are at one of our upgraded membership levels we can provide you further assistance.

Hi Craig,

Your reply was absolutely helpful. Appreciate your deep insight into the CTI technology and time to throw some light on it.

I'll try to work on these lines and will also consider working with DevConnect Support if need be.

Thanks again.

Hi Craig,

Just a quick question. Can Single Step Conference feature be used for what I am trying to achieve here?

And is it available in TSAPI or just DMCC?

Any thoughts on it?

With the state the calls will be in with this scenario I don't think so. But, please explain the call flow you are thinking about and i will let you know.

Hi Craig,

Is it possible that I can have your email address? I'll send you the detailed requirements.

Thanks.

Please open a DevConnect technical support ticket under the TSAPI interface and mention my name and they will assign it to me. Please note that if you are a registered member I will only be able to assist at a very high level.