[salawat]

>i am absolutely positive, without proof of course, that this is an extremely common practice. my isp does the exact same thing with basically the same wording. "sorry, high call volumes right now. but hey, did you know you can do lots of stuff on the website? go to the website. please use the website".

Look up Erlang numbers for call centers. We absolutely know how to calculate required reps for a desired queue dwell. It is 100% a voluntary decision to degrade the Call Center to push people to web based automation. Consider this your proof. We have the equations. Executives make the active decision to not use them/use them to shift cost burden.

t. Helped implement a Call Center before, and we aimed for sub 5 minute queue dwell at all hours of the day.

[carefulfungi]

Wait time is calculable; but you need an accurate forecast to staff and schedule. When I last worked in this space, forecasts were generated down to 15m granularity and agent work schedules (hours, break times, etc.) were derived from those forecasts.

I wonder how these systems work now...

[miki123211]

What if you get a large number of people calling at very particular times? E.G. what if you're getting far more calls at 09:00 than at 09:15? You can't hire agents just to handle a 15-minute surge.

Erlang's model assumes that the world is static or at least predictable; it doesn't take into account things like the superBowl, a hurricane cancelling 90+% of flights from a major airport, or a much-larger-than-usual number of customers trying to cancel because of a previously-confidential price increase now being publicly announced.

[sdwr]

Baseline demand affects the numbers much more than the unpredictable spikes do. You can come up with edge cases if you like, but the reality is that it all averages out pretty well with large volume.