MESSAGING WITH APPROVAL AND HAND-OFF

WhatsApp Business AI Automation

Build WhatsApp Business workflows that answer from approved knowledge, collect structured details and hand sensitive conversations to a named person.

Discuss this workflow
01

Fit

  • Customer enquiry intake
  • Appointment and viewing coordination
  • Order or booking updates
02

Architecture

  • WhatsApp Business webhook
  • Consent and identity check
  • Knowledge or workflow step
  • Human gate
  • CRM conversation log
03

Permissions & ownership

  • Meta Business verification
  • Approved templates
  • Restricted CRM and calendar scopes
04

Limitations

  • Template and opt-in rules apply
  • 24-hour service window
  • Not suitable for emergency or clinical decisions
05

Supported events / actions

  • Message received
  • Template response approved
  • Appointment requested
  • Human hand-off
06

Deliverables

  • Conversation map
  • Template inventory
  • Escalation rules
  • Monitoring and transcript policy

What Meta's rules allow, and what they don't

WhatsApp Business is the most rule-constrained channel we work with, and most implementation failures are policy failures rather than technical ones. Outside a 24-hour service window you can only reach a customer through a pre-approved template, opt-in has to be genuine and recorded, and template approval takes time that belongs in the project plan. Business verification with Meta is a prerequisite, not a formality. We scope the template inventory early, because a conversation design that depends on a message Meta will not approve has to be rebuilt from scratch rather than adjusted.

Where the human gate goes

Messaging feels informal, which makes it the channel where an automation is most likely to be asked something it should not answer alone. Anything touching a price commitment, a medical or legal question, a complaint, or a change to an existing booking stops and transfers to a named person with the conversation history attached. The assistant answers from approved material and collects structured details — dates, reference numbers, contact preferences — so that when a person picks the conversation up, the tedious part is already done and the judgement part is untouched.

Process WhatsApp messages with consent and delivery state

Normalize each inbound webhook into an internal record containing the business account, provider message ID, sender reference, received time, message type and conversation context; this makes the workflow independent of vendor payload formatting. Validate the event, deduplicate by account and provider message ID, check the recorded contact preference before any outbound action, and select only an approved template or an allowed in-window reply under the current platform rules. Keep sensitive requests and uncertain intent in a named human queue. Record send and delivery outcomes separately from the conversation itself, retry only transient failures with a bounded policy, and reconcile unknown send outcomes before retrying to avoid duplicate messages. Restrict the log to the context needed for support, and document who owns template approval, failed-message review and credential revocation.

How to measure the pilot

A pilot is accepted against the current process, not an industry promise. Agree the sample, baseline and owner before launch, then review the same measures after the workflow has handled real cases.

01

Baseline before build

Sample real cases from customer enquiry intake, appointment and viewing coordination, order or booking updates. Record the current cycle time, manual touches, exception rate and owner effort before replacing a step, so the pilot has an honest comparison rather than a generic automation benchmark.

02

Pilot success

Review message received, template response approved, appointment requested, human hand-off as observable events. Acceptance covers correct outcomes, time returned to the team, cases completed without repair and the quality of conversation map and template inventory.

03

Alerts and failure conditions

Treat these constraints as monitored failure conditions: template and opt-in rules apply; 24-hour service window; not suitable for emergency or clinical decisions. Repeated exceptions, a growing queue or a write outside the approved workflow pauses automation and creates an alert instead of silently continuing.

04

Human hand-off

The exception keeps its original input, completed actions and failure reason, then routes to a named owner under these controls: meta business verification; approved templates; restricted crm and calendar scopes. A person can approve, correct, retry or reverse the action without reconstructing the case.

WhatsApp Business automation with consent and hand-off

WhatsApp Business automation should use approved templates and valid opt-in, then route each reply under the platform’s messaging rules. Verify identity before account-specific actions and transfer complaints, payment questions or low-confidence answers to a named operator.