Chatbot flows and journeys
Bots that ask, look things up in your system, and hand over with context.
Build on a canvas with a live WhatsApp preview beside it, test on your own number, publish a version, roll back in a click. Drips run on the same guard rails as campaigns.
Keyword
“book”
1,240 in
List
What would you like?
68% chose book
Ask
Which date?
98% answered
Call API
GET /slots?date=…
412 ms
Hand over
Front desk, with path
7% dropped at the question. A nudge after four hours brought a third of them back. Timeout branch: 4 h.
What it is
A WhatsApp chatbot builder is a visual canvas for designing automated conversations that run on a business's WhatsApp number: a keyword or a template reply starts a flow, the bot sends messages, reply buttons and list menus, asks questions and saves the answers, branches on conditions and working hours, calls the business's own API for live data, and hands over to a human team with the path summarised. Journeys are the scheduled counterpart: drip sequences of template messages that start on a behaviour and stop the moment the person replies, converts or opts out.
What it does
Every line below is in the product today. The ones marked only here were not found in WATI’s, AiSensy’s or Interakt’s public documentation on 20/09/2026.
Call your API mid-flow
Credentials in an encrypted store, a request in the flow, the answer used in the next message. Bots that know order status or free slots.
Live preview and sandbox
Walk the flow as a fake customer while building, then run the draft on your own number before publishing.
Versions and rollback
Publish v3, restore v2 in a click. Contacts mid-flow finish on their version.
Analytics on the node, branches on time
Entries and drop-offs per step, drawn on the canvas. "No reply in four hours", working-hours routes, a fallback for the unexpected.
Ask, condition, set, hand over
Ask a question and save the answer. Branch on attributes. Set a stage. Hand over to a team with the path summarised for the agent.
Journeys start on behaviour
Opt-in, a tag, a link click, a keyword, a stage change, a new contact, or an event your system posts to the API.
Exit rules and continuous enrolment
Leave on reply, conversion, opt-out, tag or stage. A segment-gated journey keeps enrolling whoever qualifies.
AI to write one
Describe a bot and get an editable flow. Ask the AI to reword a step without touching its logic.
Questions about chatbot flows and journeys
Do I need to code to build a WhatsApp chatbot?
No. Flows are built on a visual canvas with a live preview. The one place code helps is the API-call node, where a developer gives the bot an endpoint to fetch live data such as order status; everything else is drag, type and connect.
Can the bot hand over to a human?
Yes. A hand-over node assigns the conversation to a team through the inbox's routing rules and attaches a summary of the path the customer took, so the agent does not start from zero. An agent can also pause automation for that person for 24 hours.
What is a WhatsApp drip campaign?
A sequence of template messages sent on a schedule, such as day 0, day 2 and day 5, to people who did something specific. Here it is called a journey. It starts on a behaviour, respects the same frequency caps as campaigns, and stops the moment the person replies, converts or opts out.
Is the chatbot allowed under Meta's AI rules?
Yes. Meta's January 2026 policy bans open-ended general-purpose AI bots on WhatsApp but explicitly allows task-scoped automation such as FAQs, bookings and order status. Flows are task-scoped by design, and the AI FAQ bot answers only from text you provide and hands over when unsure.
Talk to us.
In a forty-minute call we go through what you send today, what it costs you, and what a flagged number would cost you. Then we set up a workspace on your own Meta app and build the first campaign with you.