"building off-queue schedules instead of on-queue schedules" can you expand on this? It isn't clear to me. Having a schedule as all/mostly off queue would still result in adherence events if the agent went on queue during the off queue scheduled time.
Schedules are plans; agents not adhering to the schedule are out of adherence. Games to hide or cover up adherence events (i.e., by configuration, by schedule changes done in arrears) is deception. My recommendation would be to lower your expectation on what very good adherence is (e.g., instead of 98% maybe you adjust to 96%). You can also weigh conformance higher than adherence when evaluating agents - e.g., you are more concerned with total on queue time given versus scheduled over the actual timing of status changes compared to schedule.
Shops that attempt to circumvent adherence by various means tend to have too aggressive of adherence goals and might have some form of compensation or benefit set at different tiers of adherence for agents and sometimes supervisors. The comp/benefit leads to desires to essentially game or work the system and so it has to be realistic with checks and balances.
------------------------------
Jay Langsford
VP, R&D
------------------------------