Different measuring levels

There are many ways to summarise, and different focuses. You might think that the question “How many calls were answered?” is easy to answer, but the answer depends on what focus you have or at what “level” you are focusing.

Example:

One caller calls an access number, enters queue A, which first calls ready agent 1 (who rejects call) and then agent 2 who answers (after 40 sec in queue). Agent 2 then transfers the call to queue B. Queue B calls agent 3 who does not answer, and then agent 4, who answers (after 3 minutes in queue B). Finally, agent 4 tries a consult call to agent 5, but agent 5 does not answer, and the caller and agent hangs up.

• On an Overview level we might say that this incoming call (one caller) is answered after 40 seconds wait time since (at least) one agent answered.
• On a queue level, we have one caller that entered 2 queues and was answered in both queues. (Different queue time in the 2 queues)
• On the agent level, 5 different agents were called, 5 had one call offered, but only 2 agents answered.

The number of conversation events with a given queue_key is not the same as the number of queue events for the same queue_key, since there may be 0 or several conversation events (agents called) for each queue event for the same call_id, that is, for one caller in queue.

So, how many calls did we have, and how many were answered?