On some of our calls (but not all), we set a certain participant data key, with a variety of different values.
What would be the best way to retrieve these calls for analytics purposes?
Essentially, for calls that go through our IVR(s) we set a "LastIVR" value indicating their progress, and the value is a four digit code representing their position in the IVR (2000, 2001, 2002 etc). If they transfer to ACD we append "true" at the end so 2000 would become 2000true.
We want to be able to get a count of these values, so we can tell that (e.g.) 500 people had LastIVR value "2001", and 100 people had LastIVR value "2001true".
I know I can look at a particular call and find this value in the participant data value, but, what is the best way to do this on a large scale?
"Don't" according to support.
We were informed that is a mis-use of those attributes and they should only be used explicitly for routing purposes, there are no APIs designed to facilitate large-scale reporting based on them (even when used as legitimate routing instruments we then want information about).
We have to download the entire conversation history into an external system to do that sort of reporting due tot he complexity of ours.
All that said, for very limited usage there is a process by which you can promote an attribute into a property and then search it but it has to be done on a segment level of a conversation
But it's not intended as a scaleable solution for introduction of complex data, for that level of collection we've had to create an internal API that gets called to record tracking data in our own warehouse.
Thanks - that's not quite what we had expected to hear, we were advised to implement in this way!
I guess we can actually do this in Excel by exporting the Interactions list and just doing a few formulas on the participant data... Not dynamic but might work for what we need.