One of our key metrics in social response is, of course, Response Time. There are many cases in conversation, though, where we might need to research a question. We might respond to the user to let them know, they respond "Sounds great!", then we snooze and research.
The problem is that I don't think snooze time is factored out of response time. If we snooze a thing for 2 days, and then respond right after it unsnoozes, Response Time still looks like 2 days, totally throwing off averages. It also falsely represents the customer experience as it adds tot he pile of "long tail" 8 hour+ responses in LSW Analytics.
Is there a way to have snooze time factor out of Response Time, or is that a product change idea?
@HeyScottyJ our calculation of response time (TAR) is always from the perspective of the customer and the time it takes for an agent to provide an answer. We also have Handle TIme now available as a way to measure agent efficiency by calculating how much time an agent actually spends on the individual response as well as the conversation.
The scenario that you describe is valid -- see this product idea and related comments for recommendations on how you can account for the use case where a conversation is snoozed to do research or investigation.
Totally understand the logic on the customer perspective and the intent of handle time as a separate metric, but traditionalists need to see traditional metrics to understand stuff.
The use case commented by @LeslieJ on that Idea you linked me to sounds perfecto - we're going to give that a stab. The "thanks" reply is really the problematic one, because our only other recourse is to say "you're welcome, stop messaging me" 😉