How to calculate average response time
For every customer message that gets a reply, measure the gap until that reply, then average all the gaps for the period. A conversation with a two-minute first reply that then waits a day for the second answer has a fast first response time but a poor average response time.
Averages are skewed by outliers, so many teams also look at the median. Exclude internal notes and automatic acknowledgments, and decide whether nights and weekends count. Email usually runs slower than chat or text, so compare each channel against itself.
Example: an accounting firm in tax season
In March, an accounting firm’s email threads with clients get slower with each exchange as staff juggle returns. The first reply comes the same day, but follow-up answers about which documents to send take two or three days. Tracking average response time per thread reveals the slowdown that first response time alone hides. This scenario is illustrative.
Keeping replies moving in InsertChat
InsertChat’s assistant replies to every message in a conversation, not only the first, on calls, chat, texts, and WhatsApp. For email, it can draft each reply for review or send automatically with a delay and a business-hours option, while complaints, payments, and legal topics wait for your team.
Conversations waiting on a person are marked needs-you, and queues show what is waiting for your reply. See the accounting firm setup.
