Agent Number Effects

Fig. 1 - Schedule Table Columns

When you use the Work Scheduler to allocate the agent hour output from a campaign (Fig. 1, column 2) to a number of time periods, spread over days, weeks or months, you also have the opportunity to change the number of agents you show as logged in (Fig. 1, column 1), in any time period. See Schedule Table - Input and Output Columns - Inputs

On predictive campaigns, there is a variable relationship between the number of agents logged in to a campaign and the talk and wait times achieved. In most cases, this is unlikely to be big enough to bother you. But you should check out the effects by running your campaigns using different agent numbers. The higher the number of agents, the higher the average talk time and the higher the number of results that will be recorded in an hour.

Example

You may have a task at a fixed period during the day to which you allocate

10 agents on Mondays and Tuesdays, and
30 agents on the remaining weekdays

If you treat all of this as a single campaign in Oceanic®, with an average of 20 agents, you will be

overstating what can be achieved early in the week, and
understating results for later in the week

Whether the differences you observe are significant or not will depend upon the other assumptions you make, particularly for the distribution of call outcomes.

If they are significant, then use different campaigns for each set of agents to generate your results, i.e. run a campaign with one set of agents, then change agent numbers and run again. Since you are assuming different time periods, you may also want to take account of different call outcomes as well.