List weighting is one of those things that's hard to get your head around, but once you understand it it's pretty straight forward.
It's the percentage of records from each list are retrieved into the buffer as a percentage. So if your 6 lists are set like this:
List1 - weight = 10
List2 - weight = 20
List3 - weight = 10
List4 - weight = 10
List5 - weight = 10
List6 - weight = 10
total of all weights = 70.
The percentage of records retrieved into the buffer will be like this.
List1 - 14.3%
List2 - 28.6%
List3 - 14.3%
List4 - 14.3%
List5 - 14.3%
List6 - 14.3%
To work this out you do this: <weight> / <total> * 100
* remember, this is records in the buffer (in a retrieved state), not records that will dial.
Now, it's not quite that simple (because Genesys), this is where it gets complicated.
You also have the record buffers and dialling priorities to contend with, and the dialling priorities.
Minimum Record Buffer = Minimum number of records retrieved from a list, per agent.
Optimal Record Buffer = Number of records retrieved per agent when OCS retrieves new records (eg. when it hits the minimum buffer)
Dialling priorities control which records from the buffer get dialled first. By default these are set so that Campaign Callbacks dial first, then Campaign Rescheduled, and the lowest priority are new, or fresh, records.
So, if your Optimal Record Buffer is too low you can end up with a situation where a particular list only has fresh records retrieved but other lists have campaign callbacks and rescheduled. In which case, the list where only fresh records are retrieved will not dial because the other records are high priority. It's also possible that you will experience slow dialling because there are not enough fresh records in the buffer and OCS is waiting until the right time to dial the callbacks and rescheduled. What the record buffer values should be requires experimentation and tuning. I will say this though, do not use the default values for your buffer values, but on the other hand, do not use very large values. Personally, I run 10 and 12 as a starting point, then analyse and tune from there.
I would also recommend not using 6 lists unless you have a very large agent pool. Weighting and priorities are very difficult to predict so the smaller the number of variables the better.
Hope this helps.