The measurement doesn't make sense in any respect and also puts the onus of errors in scheduling/forecasting on the agent.
One thing that might help you is to use true calls/hour...to get this measurement take for each agent their # of calls answered in a day, then divide by (total login time-total wait time) the total login and wait should come from the switch report (why use spreadsheets when your $pendy switch already monitors and tracks this info). This removes any variance in login time (for whatever reason) and removes the onus of a mistake in schedule/forecast from the agent (if there is a large amount of wait because of overstaffing, the agent is not at fault ). This measurement gives snr mgmt their calls/hour figure and gives your agent a better end of the deal. hope this helps...
------------------
forceguy_2@hotmail.com