Tell me about your experience with CCB and any cool modifications you've done.

[david.macias]

I've done a few installs of CCB, but never really done a serious deep dive into the solution. I recently posted a blog about a modifications I made to CCB to prevent it from being offered when your expected call back time was longer than your open time. This got me thinking about what else others have done. The most common modifications I see are:

- Changing the voice. It's ironic that all the default prompts for Cisco are female voices, but for some reason CCB is a male voice.

- Changing the app to be used for all business units who which to use it, allows to branch off based on Q.

- Only make a callback when the EWT and your call is 1 in Q.

Other than that I've not seen a lot of modifications. Even though I feel that CCB has a lot to be desired I'm surprised there hasn't been much posted on how others are making it better. I would love to hear what others are doing or are we just enslaved to Cisco's way of doing CCB.

~david

[bill.king1]

I think the challenge is that most of the basic type things you might want (do it in Spanish/another non-English language or offer custom messaging based on line of business when it does the callback) all require you to make some level of modifications to some of the CVP scripts that Cisco says cannot be modified.

So if you are able to do something custom/neat, it typically also technically means it will not be supported if you ever run into a problem.

[david.macias]

You actually bring up a really good point. The "limitations" that Cisco calls out are a bit overly restrictive, however I think you can still do plenty without going too far away from the lines Cisco draws. Personally, I see the limitations strictly around the custom elements that Cisco built and everything else is fair game. Now, I've been lucky enough to not have to have asked for TAC help on CCB modifications so when it happens I might be singing a different tune.

david

[ptindall]

As US English hasn't been adopted by the rest of world it was always intended that the caller experience would be modified as required for initial treatment/dialogue before disconnect and the interaction following reconnect.  The intention was only ever to protect the mechanism from being broken by accidental scripting problems, especially during the caller-less phase.

Regarding customisations, by far the most common one I encounter is operation in preview mode such that the agent gets the callerless call and triggers the callback.  This is to handle the situations when preemptive mode is just never going to work reliably due to long call times, small teams etc and customers won't tolerate being placed back in queue (even for a few seconds).  

Another one is use of substitution in the Set Queue Defaults element to make your scripting somewhat simpler.  

Paul

[janinegraves]

Paul, by 'preview mode' do you mean using the AgentRequest api?

[ptindall]

Janine,

Agent Request API by default isn't actually preview mode even though the callback isn't made until the agent goes ready although it can be made to work that way by delivering the callback to CVP and then doing a SIP Refer to the intended callback destination when the agent accepts.

The one mentioned is the affectionately named Courtesy Callback Nordic Edition.  This was a retrofit of the very original CCB proof of concept mode of working where the agent triggers the callback using a signal from the desktop.   As always, there are pros and cons but it was done some years ago specifically to match requirements in a particular region for preservation of position in queue, absolutely zero tolerance of the caller queuing once called back, long and uneven call durations and small teams. 

It wasn't a solution that was intended to be proliferated as it's by no means perfect but as the requirement has come up repeatedly and it has proved to work successfully in the right hands, it spread a bit by stealth and I did eventually concede and make it public on the Tindall workbench twitter channel.

[david.macias]

Paul drops a golden nugget by saying preview and provides no details! Yes, Paul curious about this agent preview as it's something that comes up often.

david

[ptindall]

Couple of links to things mentioned:

https://app.box.com/s/zjfn6vkdvt8h6ch50k2stghrruw1k1jk

https://app.box.com/s/z5k2ziz8yf49biqx9hy4fzpghk9kutox

[david.macias]

@ptindall this is genious. I'm going to have to try this. I will report back. Thank you a million times for sharing this.

david

[mouazyusuf]

I do think the calculations for the EWT is a big challenge , for one deployment a customer has set different prioritization for calls in different queues and this totally mess the EWT calculations for different calls . its also noticeable and there are  big gap between what is calculated and when system actually make the call back . 

[david.macias]

This is 100% the biggest issue. EWT is just not consistent and if you have any sort of unique routing (priority, multiple PQs, steps, etc.) you can toss EWT out the door.

david

[AlexIancu91604]

EWT dynamic calculation was a nightmare for me which was switched to static

Now I am curious if anyone has any experience with limiting CCB callback until the queue position is 1 

[david.macias]

Yeah I've done this before and it worked well enough. Problem is that you still get some callers that are on hold when you have such not predicative talk times.

david

[Sadi20107s]

Hi,

so did you manage to resolved CCB callback issue. if i have 5 calls in ccb queue for every call its saying you are number 1 in Queue.

[david.macias]

This prompt is the callback CVP app. Look through the elements and change it. I removed the "you are number" part of it every time.

david