01
Slash command picker
Type "/" in the composer to open a fuzzy search across the entire library scoped to the agent's team. Hit enter to insert; the merge fields populate from the current contact and conversation in under 50 ms.
Build a shared library of reply snippets with merge fields, attachments and keyboard shortcuts. Agents trigger them with a slash command, and the AI layer suggests the right snippet based on the incoming message. New hires sound like five-year veterans on day one.
Pull up any support inbox and you will find the same five answers retyped in slightly different ways: "Yes, we ship to PIN code XYZ, here is the shipping policy link"; "Your order is on the way, here is the tracking link"; "Please share your order ID and registered phone number"; "I have processed the refund, you should see it in 5-7 working days." Every agent has their own slightly off version with a typo, a different tone, or a missing crucial detail. Customers get inconsistent answers, brand voice drifts, and seasoned agents end up coaching juniors by Slack DM on what to copy-paste — which itself becomes a tribal knowledge bottleneck.
The old solution was a Notion doc with reply templates, which works for the first two weeks and then rots. Nobody updates it, nobody can find the right answer fast enough in the heat of a queue, and the merge fields are still manual — agents copy "[ORDER_ID]" and forget to swap in the real value.
A proper canned-reply system is a small piece of software that disproportionately affects quality. Done right, an agent types "/refund" and the right snippet appears with the order ID, customer name and refund amount already merged, the policy PDF attached, and the tone matching the brand. New hires hit production-grade replies on day one, supervisors update the canon centrally, and the team's collective experience stops walking out the door when someone leaves.
Wachat's canned replies live in a shared library scoped per workspace, with folders for teams (Sales, Support, Billing), languages (English, Hindi, Tamil, Bengali) and channels (WhatsApp can use interactive button templates that web chat cannot). Each snippet has a title, a shortcut (e.g. /shipping, /refund-7days), a body with Liquid-style merge fields like {{contact.first_name}} or {{order.tracking_url}}, optional attachments, and tone/voice tags so the AI layer knows when to suggest it.
The composer integrates the library at three levels of agent effort. The lazy path is "/" which opens a fuzzy-search picker over the entire library scoped to the agent's team. The medium path is typing the shortcut directly — /refund7d expands to the full snippet inline, like a text expander. The zero-effort path is AI suggestion: as the customer message arrives, Wachat's AI suggests the top-3 most likely snippets ranked by past usage and message similarity, and the agent presses Tab to accept.
Merge fields pull live from the contact, the conversation, the linked order, the active flow run, the assigned agent and a few well-known constants. They support filters and defaults — {{order.total | currency: "INR" }} or {{contact.first_name | default: "there"}} — so a missing field never produces an awkward "Hi ," reply. The preview pane in the snippet editor renders against a sample contact so authors can validate output before publishing.
Attachments are first-class. Pin a tracking-policy PDF to the /shipping snippet, a refund-process diagram to /refund, a Calendly link card to /book-call, and they travel with the snippet on every channel that supports them. On WhatsApp, the attachment uses Meta's media upload pipeline transparently; on email, it goes as an attachment; on web chat, it goes as a downloadable file card. SabFiles links handle anything too large for native channel limits.
Type "/" in the composer to open a fuzzy search across the entire library scoped to the agent's team. Hit enter to insert; the merge fields populate from the current contact and conversation in under 50 ms.
Liquid-style variables pull from contact, conversation, order, flow run and workspace. Filters handle currency, date, name capitalisation and defaults so a missing field never renders blank or breaks the sentence.
Pin PDFs, images, SabFiles links, Calendly cards or product catalogue items to a snippet. They travel correctly to WhatsApp, Instagram, email and web chat using each channel's native media handling.
As a customer message arrives, Wachat's AI ranks the top-3 most relevant snippets by past usage, message similarity and customer segment. The agent accepts with Tab, edits with Enter, or ignores — and the model learns from the choice.
Scope snippets to teams (Support only, Sales only) or roles (senior agents only for the /escalate-vip snippet). Public snippets are visible to everyone; private snippets are personal drafts that do not pollute the team list.
One logical snippet can have variants per language: /shipping has English, Hindi, Tamil and Bengali versions. Wachat picks the right variant based on the contact's detected language or explicit preference, falling back to English.
Per-snippet metrics: how many times used, by which agents, in what channels, with what CSAT outcome. Identify the snippets that drive high CSAT to promote, and the snippets that correlate with bad outcomes to retire or rewrite.
A footwear brand has 22 canned replies covering size exchange, return pickup, refund timeline, replacement issue, damaged item and quality dispute. /size-exchange auto-populates the order ID, current size and recommended size from the order; the agent reviews and sends in 4 seconds instead of 90.
A vertical SaaS uses canned replies for the first 14 days of every trial: /welcome-day1, /day3-checkin, /demo-booking, /admin-setup, /api-keys. Snippets include personalised Loom links generated by a flow, with the AE's name and Calendly URL merged in. Trial-to-paid conversion is up 18%.
A neobank maintains 35 canned replies for KYC and account-opening scenarios. Each has the regulatory copy reviewed by compliance, with merge fields for PAN, Aadhaar masking, and required document list. New hires can handle Tier 1 KYC queries from week two with zero compliance escapes.
A boutique travel agency uses /itinerary to send a PDF day-wise plan, /visa-checklist for the destination-specific visa docs, and /booking-confirm for the deposit invoice. Snippets are versioned per season and per destination, so monsoon-Goa and winter-Goa get different recommendations.
A coding bootcamp uses canned replies for course access (/reset-password, /lms-access, /class-schedule), submission help (/upload-issue, /grading-timeline) and refund policy (/refund-pre-week2). Hindi and English variants share one shortcut; the contact's preferred language picks the right one automatically.
Canned Replies is included on every SabNode workspace. No separate billing, no extra setup — flip it on from your workspace settings.
Wachat scans your last 90 days of resolved conversations, clusters frequent replies, and suggests 20-40 starter canned replies with shortcuts. Review, edit and approve in bulk — most teams ship the initial library in under an hour.
Open each snippet and replace hard-coded names, order IDs and amounts with merge fields. Preview against a sample contact in the right pane. Save and publish; the snippet is immediately available in every agent's composer.
Pin PDFs, images, Calendly bookings, product cards or SabFiles links. Verify each channel renders the attachment correctly (WhatsApp media, email attachment, web-chat card) using the preview tool.
Enable AI-suggested snippets. The model learns from the team's acceptance and edit patterns over the first 1-2 weeks. Suggested snippet accuracy typically reaches 70%+ after 500 conversations.
Open the analytics view. Promote the high-CSAT snippets, rewrite the low-CSAT ones, and retire any that have not been used in 30 days. Treat the library like a product backlog, not a static doc.
Can't find what you're looking for? Talk to our team.
No credit card. No sales call required. Spin up a workspace, plug in a number, and your team is live in under an hour.