Keyword-Triggered Replies on WhatsApp
The same questions arrive on your WhatsApp all day. "Price?" "Hours?" "Where are you located?" "Is this in stock?" Each one is a thirty-second copy-paste — but multiply that by a hundred messages, two agents, and a few group chats, and your team spends half its day typing the same four answers instead of closing the conversations that actually need a human.
A WhatsApp keyword reply fixes the dull part of that. When an inbound message contains a word you've defined — price, catalog, book — the right canned answer goes out instantly, before anyone touches the keyboard. Done well, it shaves hours off your week and gives customers a faster reply than any human could. Done badly, it turns your number into a robot that answers the wrong question.
This guide is for teams who handle real volume on WhatsApp and want the predictable questions to answer themselves — without the keyword maze that frustrates customers. We'll cover how keyword matching works, where it beats a chatbot, the patterns that go wrong, and how to set it up on your existing number.
What a keyword-triggered reply actually is
A keyword reply is a simple rule: if an inbound message matches a trigger word, send a specific canned answer. No AI, no decision tree — just a lookup table between words customers type and the responses you've pre-written.
That simplicity is the point. Unlike a WhatsApp chatbot with branching flows, a keyword rule does exactly one thing and you can read the whole logic at a glance. The trade-off is that it only catches what you anticipated. "Price" works; "how much does this run me" might not, unless you've matched on that phrasing too.
Three things define a good keyword rule:
- The trigger — the word or phrase that fires the reply (and how loosely it matches).
- The response — the canned message that goes out, ideally with a clear next step.
- The scope — which chats it applies to, and what happens when no keyword matches.
Get those three right and most of your repeat questions answer themselves. Get them wrong and you ship a frustrating loop. The rest of this guide is about getting them right.
Keyword reply vs. auto-reply vs. chatbot
These three get blurred constantly, but they solve different problems. Pick by how much the customer's message varies.
| Tool | Fires when | Best for |
|---|---|---|
| Greeting / away message | Any first message, or out of hours | Setting expectations, not answering |
| Keyword reply | Inbound contains a defined word | Predictable, repeated FAQs |
| Chatbot / flow | A conversation needs branching | Multi-step qualification or routing |
An auto-reply doesn't read the message — it fires on any inbound or on a schedule ("we're closed, back at 9am"). A keyword reply reads the content and picks an answer. A chatbot strings several of those decisions together into a flow.
Most teams over-reach here and build a chatbot when three keyword rules would do. If a customer's question has one obvious answer, a keyword reply is faster to set up, easier to maintain, and far less likely to trap someone in a menu. Reach for a flow only when the next message genuinely depends on the last one.
Designing keywords that don't backfire
The failure mode is always the same: the rule fires when it shouldn't, or doesn't fire when it should. A handful of habits prevent most of it.
Match on intent, not on one exact word
People phrase the same question ten ways. If your rule only matches the literal word catalog, you'll miss catalogue, products, what do you sell, and list. Map several phrasings to the same response so real language triggers it — not just the one word you happened to pick.
Always handle the no-match case
The single most important rule isn't a keyword at all. When nothing matches, the customer should never hit silence or a "sorry, I didn't understand that" wall. Route unmatched messages straight to a human in your shared inbox, with a short line like "one moment, a teammate is on this." A keyword system that quietly drops the questions it can't answer is worse than no automation.
Keep the answer short and end with a next step
A keyword reply that dumps three paragraphs gets skimmed and ignored. Answer in a sentence or two, then give one clear next action: a link, a question back, or "reply YES to book." The goal is to advance the conversation, not to end it with a wall of text.
Don't stack triggers that collide
If book fires an appointment reply and booking fires something else, a message containing both becomes a coin toss. Keep your trigger list short, review it for overlaps, and resist the urge to add a rule for every edge case — the long tail belongs with a human, not a fortieth keyword.
Keyword replies that earn their keep
Not every FAQ deserves automation. The ones worth a keyword rule share a trait: one unambiguous answer that rarely changes. A few that consistently pull their weight:
- Pricing — "price" or "cost" returns your rate card or a link, then asks what they're after.
- Hours & location — "hours", "open", "where" return the same facts every time.
- Catalog / menu — "catalog", "products", "menu" return a link or your WhatsApp catalog.
- Booking — "book", "appointment", "slot" return your scheduling link or available times.
- Order status — "order", "tracking" prompt for an order number, then hand off to a human.
Questions you should not automate are the ones where the right answer depends on context: "will this fit my use case?", complaints, anything involving money owed. Those need a person, and a keyword reply that fakes an answer does more damage than a slightly slower human one. When in doubt, let the message reach your team and use AI suggested replies — available on Growth and Scale — to draft the response fast instead of forcing a canned one.
Where Tenashi fits
Most keyword-reply tools are built on the official WhatsApp Business API, which means template approvals, per-message fees, and — critically — no access to group chats at all. If your repeated questions show up in broker groups, community groups, or customer groups, an API tool simply can't see them.
Tenashi is the WhatsApp workspace for teams, and it works differently. You link your existing number with a QR scan — no new SIM, no porting — and your whole team works one shared inbox that includes DMs and groups. Keyword-triggered replies, canned replies, and auto-replies sit alongside a real CRM (contacts, custom fields, deals, a drag-drop pipeline) and a support desk with tickets and SLAs, so an automated first touch can roll straight into a tracked conversation a human owns. It's part of the broader WhatsApp automation toolkit, not a bolt-on bot.
Because it runs on your own number rather than the API, there are no template approvals and no per-message fees — a flat price from $18/mo. Sends are ban-protected: human-paced with jitter, typing presence, gradual warm-up, and duplicate/burst guards, so heavy automated replies don't trip WhatsApp's spam signals. That reduces risk; it isn't a guarantee, since WhatsApp's policies can change. And on Growth and Scale, AI can suggest replies, summarize long group threads, run OCR on images, and translate inline — for everything a fixed keyword can't catch.
There's a 3-day Growth trial, no card.
Start your free trial → · See every feature
FAQ
What is a WhatsApp keyword reply?
A WhatsApp keyword reply is an automation rule that watches inbound messages for a defined word or phrase — like "price" or "book" — and instantly sends a matching canned answer. It's the simplest form of WhatsApp automation, handling predictable, repeated questions without an agent typing the same response every time.
Can the WhatsApp Business app set up keyword auto-replies?
No. The free WhatsApp Business app offers greeting and away messages that fire on any inbound, but it can't read a message and pick an answer based on a keyword. For true keyword-triggered replies you need a workspace or automation layer on top, like Tenashi, connected to your existing number.
Do keyword replies work in WhatsApp groups?
Only if your tool can access groups — and most can't. Tools built on the official WhatsApp Business API have no group access at all. Tenashi runs on your own number, so group chats sit in the same shared inbox as DMs and your automations and canned replies can work them too.
How do keyword replies differ from a chatbot?
A keyword reply is a single rule: one trigger word, one canned answer, no branching. A chatbot chains several decisions into a flow where the next step depends on the last reply. Use keyword rules for one-answer FAQs; reach for a flow only when a conversation genuinely needs multiple steps.
Will keyword auto-replies get my WhatsApp number banned?
They can if they fire fast, identical messages at high volume — that pattern looks like spam to WhatsApp. Choose a tool that paces sends like a human. Tenashi adds jitter, typing presence, warm-up, and duplicate guards to reduce that risk, though no tool can guarantee against policy changes.
What should happen when no keyword matches?
The message should reach a human, never silence or an error reply. A good setup routes anything it can't match into a shared inbox where a teammate takes over — ideally with AI suggested replies (Growth and Scale) to draft an answer fast. Automation should cover the predictable and hand off the rest.
Stop retyping the same four answers. Start a free trial, connect your existing number, and let the predictable questions answer themselves — while every message a keyword can't handle lands with a human.