D ReplyTray

Canned Responses: 20 Examples That Don't Sound Like a Bot

Canned Responses: 20 Examples That Don't Sound Like a Bot

Most canned response libraries fail for the same reason: they were written to be safe, and safe reads as hollow. The lines below are the opposite — short, specific, and human enough that you can send them with one word changed.

Acknowledging a delay

  • “Caught your message — I'm on site until 4, proper reply after that.”
  • “This needs a longer answer than I can type now. Back to you this evening.”
  • “Checking with a colleague, will come back with something definite.”

Why they work: they give a time and a reason, not an apology loop.

Delivering bad news

  • “It's not going to work the way you described — here's the version that will.”
  • “We can't do X, but Y gets you most of the way there.”
  • “That's outside what I can do. Here's who has it.”

Answering the “how much” question

  • “$19.90 once — no subscription. Happy to answer anything before you decide.”
  • “Depends on scope: usually one of two numbers. Tell me which case you're in?”
  • “It's on the site, but the short version: one-time, two machines.”

Handling a complaint

  • “That's a fair complaint. Here's what I can do about it today.”
  • “You're right, that shouldn't have happened. Fixed now — and here's what changed.”
  • “I'd rather hear it than not. What would a good outcome look like for you?”

Closing a thread politely

  • “Nothing needed from you — just closing the loop.”
  • “Going to leave this here unless something changes on your side.”
  • “Ping me if it flares up again, I'll keep the notes on file.”

Thanking someone without the corporate voice

  • “This is genuinely useful, thank you.”
  • “Exactly the detail I needed — saves me a round trip.”
  • “Noted and stolen for the next one.”

Declining a request

  • “Not something I take on, but I know someone good.”
  • “Happy to help with A and B; C is out of scope.”

Starting a follow-up

  • “Following up once, in case this got buried.”
  • “No rush — just keeping this on your radar.”

The pattern behind all twenty

Every line does three things: it acknowledges the specific situation, it commits to something concrete (a time, a scope, a next step), and it avoids one stock phrase — “please be advised”, “we apologise for any inconvenience”, “thank you for reaching out”. Those three phrases alone account for most of the bot smell.

Keep the ones that sound like you, delete the rest, and put them somewhere you can reach in one click — a document in another tab will not get used. Here is the system for that.

Want to try it?

ReplyTray docks next to the app you are using — click a line and it is sent.

See how it works

More posts

How to Make AI-Written Replies Sound Like You

How to Respond to Negative Comments Without Making It Worse

Text Expanders vs. Reply Snippets: Which One Actually Saves You Time?

← All posts