Shipping times
/shippingGreat question! Orders ship within 2 business days. You'll get a tracking link by email the moment it leaves the warehouse.
Most used
💬 Engagement & inbox
Every inbox has its greatest hits — shipping times, pricing, “do you do collabs?”. Saved replies (canned responses, if you grew up on help desks) turn those into a shared library your whole team inserts, personalizes, and sends — the same right answer, whoever’s on the queue.
Workspace reply library
Shared with the whole team/shippingGreat question! Orders ship within 2 business days. You'll get a tracking link by email the moment it leaves the warehouse.
Most used
/refundSorry it didn't work out. We accept returns within 30 days — just reply with your order number and we'll take it from there.
Exact wording
/collabThanks for reaching out! For partnerships, email partners@… with your media kit so the right person sees it.
Routes leads
/pricingPlans start at … and everything's listed here: [link]. Happy to help you pick — what are you mainly looking to do?
Ends with a question
Illustrative library — write yours once, insert it anywhere the inbox reaches.
The playbook
The trap is saving everything. A good library is small, accurate, and boring on purpose: it holds the answers where being exactly right matters more than sounding spontaneous.
Shipping windows, refund policy, warranty terms, opening hours. One wrong improvised answer costs more than the whole library took to write.
Pricing page, booking link, support portal, media kit. Stop excavating them from your own bio every time.
“For partnerships, email…”, “DM us your order number”. Replies that move the conversation to the right place, politely and identically.
A fast, warm “we're on it” for complaints buys your team time to investigate — and it should never be improvised while stressed.
Three or four variants for praise and shout-outs, so your thank-yous don't read like a bot stuck on a loop.
Declining free-product requests, off-topic pitches, and spam collabs. Hardest reply to write in the moment; easiest to save.


Snippets vs. AI drafts
Saved replies shine when the answer never changes. For everything context-dependent, the same inbox offers AI-drafted replies written in your voice — generated from the actual thread, waiting for your edit and tap. The AI reply generator guide covers that side in depth.
| Refund policy wording | Saved reply |
| “Where's my order?” follow-up | Saved reply + edit |
| Nuanced product question | AI draft |
| Complaint with history | AI draft, human review |
| Full support queue at scale | Customer Support Agent |
At the far end sits the Customer Support Agent, which drafts across the whole queue — draft-first by default, per-run credits shown transparently, and it stands down in any thread the moment a human types.
Honest setup
Turn the unified inbox on for the channels where questions actually arrive — it's opt-in per channel, so nothing is pulled until you say so.
Skim your last few weeks of DMs and comments and list the questions that appear three or more times. That list is your library, nothing more.
Write each reply once, properly: correct facts, the right link, and a tone check. This is the version everyone will send.
Work the queue and insert as you go — personalize the first line, keep the body. Retire or reword anything you keep editing before sending.
When the un-cannable questions outnumber the canned ones, add AI drafts — and eventually an agent — on top of the library.
Saved replies are one piece of a managed queue — the inbox management guide shows how triage, assignments, and response-time targets fit around them.
Yes — the library lives at the workspace level, so every teammate working the inbox inserts from the same set. Update a reply once (a price, a policy, a link) and everyone sends the corrected version from then on. No more three people keeping three private notes files.
Always. Inserting a saved reply drops it into the composer as editable text, not a locked template. The healthy habit is a one-line personal touch on top of the canned body — the facts stay consistent, the greeting stays human.
Saved replies are exact words you wrote once and reuse verbatim — perfect when precision matters, like refund policies or pricing. AI drafts are generated fresh per conversation in your voice — better for anything context-dependent. Most teams use both: snippets for the fixed 20 questions, AI drafts for the long tail.
They work anywhere the unified inbox does: Facebook Page DMs and comments, Instagram professional DMs, comments, and mentions, and YouTube comments. One library serves every inbox channel, so the answer is the same wherever the question came from.
Not directly — agents like the Customer Support Agent draft from your business profile and workspace brand voice rather than reading the reply library. Saved replies stay a human tool: your team's canonical answers, inserted and personalized in two clicks. And agents stay draft-first by default, so even the routine responses they prepare wait for a human tap unless you deliberately grant auto-send.
The inbox translates incoming messages so you can read them, and you can reply with a translated response. You don't need to duplicate your whole library per language to handle the occasional thread in French or Portuguese.
Open the unified inbox, save your first five replies, and feel the queue get lighter today.