Most teams pick a reply channel out of habit. The message came in by email, so the reply goes out by email. That is right maybe two thirds of the time, and the other third is where in-app messaging earns its place. The test is short.
Does the answer end in a click?
If the customer has to be inside the product to finish — change a setting, retry a failed payment — answer in-app. Email asks them to read the instruction in one window, switch to another, and hold four steps in their head while they do it. An in-app message puts the instruction beside the thing it describes. Kenji Morita measured this on his own team over six weeks: the same instruction sent in-app was completed by 71 percent of recipients within a day, against 44 percent when it went by email.
Email wins when the answer has to be kept
Anything the customer will need again — an invoice breakdown, or a security answer their compliance team will ask about in March — goes to email, because email is the archive people actually search. An in-app message that vanishes when the widget closes is a poor place to put a reference number. If the answer contains something they will quote back to you in a month, send it somewhere they can find it.
Do not send both
The instinct when unsure is to send both, and that is the worst of the options. The customer now has two threads open, replies to whichever is nearest to hand, and the agent finds the reply in the channel they were not watching. Pick one and say which one out loud. ‘I have sent the full breakdown to your email’ is nine words and it prevents the split.
Where the choice gets made
This is a routing decision, not an agent preference, so it belongs in the workflow rather than in a wiki nobody opens. Send password and settings questions to an in-app queue by default, billing to email, and let agents override either. Defaults settle about 80 percent of it, and the arguments stop.
Leave a Reply