VDN Abandons/Skill Abandons

[RGarcia87]

We recently setup all of our "VDN Call Profiles" in 5 second increments. We noticed that we are abandoning many calls in the first bucket (0-5 sec). The "Skill Call Profiles" are setup the same way and there is a big difference in the abandoned calls. In the vectors we are not prompting for digits or have any announcements before the calls queue to the skill. What we do have are "time of day" and "holiday table" lookups, and two "adjunct routing link" steps prior to queueing to the skill.

 

[TykeUK]

The split/skill reports only starts to measure calls after they have hit the "queue to skill" step. If calls abandon before this time there will be differences in the reports.

Can we see the vector in question?

 

[RGarcia87]

I understand that but it shouldn't be that big of a difference. It takes maybe 2 seconds to get to the "queue to" step. Here is the vector...

1 wait-time 2 secs hearing ringback
2 goto step 20 if holiday in table 2
3 goto step 18 if time-of-day is all 12:00 AM to all 6:00 AM
4 goto step 16 if staffed-agents in skill 72 < 1
5 adjunct routing link 2
6 adjunct routing link 1
7 queue-to skill 72 pri m
8 announcement 39176
9 wait-time 60 secs hearing music
10 goto step 18 if time-of-day is all 12:00 AM to all 6:00 AM
11 announcement 39177
12 wait-time 60 secs hearing music
13 announcement 39153
14 goto step 9 if unconditionally
15 stop
16 route-to number 39012 with cov n if unconditionally
17 stop
18 route to number 918002345678 with cov n if unconditionally
19 stop
20 disconnect after announcement 32076.

 

[TykeUK]

I can see what you mean. My guess would be that it has something to do with adjunct routing steps. As the calls hits this step before queing to a skill the switch will apparently send an ASAI message requesting a call route over the specified adjunct link. Unfortunately I have never had much experience with adjunct links so this could be nothing to do with this.

 

[avgindyguy]

I agree with TykeUK. We use Genesys and the adjunct links can create this problem. The delay is because of the MAPD card, which I am assuming you are using. The routing decision must check the link first and wait for a request/decision. Also, what happens if the adjunct link is to be used to route the call? Are you still queuing the call? Also, as an experiement you could change the call profile setup a little. First increment would be 0-2 seconds, next 2-5 seconds, etc. This may help you isolate where the abandons are happening.