Resources · WhatsApp
WhatsApp Inbound
Automation Guide
for Singapore Businesses.
The complete reference for automatically receiving, qualifying, routing, and logging every inbound WhatsApp enquiry within 90 seconds — PDPA-compliant, built on the WhatsApp Business API and Make.com.
What Is WhatsApp Inbound Automation?
WhatsApp inbound automation is a pipeline that automatically receives, acknowledges, qualifies, and routes messages that customers send to your business first — within 90 seconds, 24 hours a day, without manual handling. Unlike outbound automation (which initiates contact after you capture a lead elsewhere), inbound automation fires the moment a stranger messages you first.
When a prospect taps your WhatsApp link from a Facebook ad, Google Business Profile, or referral and types "Hi, interested in your services," an inbound automation pipeline:
- 01Detects the message in real time via a webhook from the WhatsApp Business API
- 02Sends a personalised acknowledgement within 90 seconds — with PDPA opt-in consent
- 03Runs the message through an AI qualification layer to extract intent, budget, and timeline
- 04Logs the structured lead data to your CRM automatically
- 05Routes the conversation — Telegram alert for hot leads, booking link for warm, nurture for cold
This is distinct from the free WhatsApp Business app, which supports only static away messages and labels — with no webhook, no CRM integration, and no conditional routing. Full inbound automation requires the WhatsApp Business API(Meta’s cloud-hosted API) connected to a workflow platform like Make.com or n8n.
Why Manual WhatsApp Handling Is Costing You Deals
Singapore’s service market is hyper-competitive. Prospects searching for a financial advisor, property agent, or clinic typically message 3–5 providers simultaneously. The first to respond professionally wins the conversation.
391%
more likely to qualify a lead contacted within 1 minute vs 30 minutes
MIT / InsideSales.com
80%
drop in qualification probability after the 5-minute window closes
InsideSales.com
3–4 hrs
average Singapore SME response time to inbound WhatsApp enquiries
Lumora client audits
There is also a volume problem. A typical Facebook or Instagram ad campaign generates 30–150 WhatsApp enquiries per week. Without automation, someone must read each message, ask qualification questions, copy details to a spreadsheet, and decide the next step — 8–15 minutes per lead, degrading in quality as volume rises.
Inbound automation collapses this to under 90 seconds per lead, regardless of volume or time of day. The team member sees only pre-qualified, structured lead summaries — not a stream of raw messages to triage.
The 5-Stage Inbound Automation Pipeline
Webhook Receipt
The moment a message arrives at your WhatsApp Business API number, Meta fires a webhook to your Make.com scenario. This is instantaneous — there is no polling or delay. The raw JSON payload contains the sender's phone number, WhatsApp display name, message body, and timestamp.
90-Second Personalised Acknowledgement + PDPA Consent
Within 90 seconds, an automated message goes out that addresses the prospect by first name (if available), references the service or page they likely came from, and includes your PDPA opt-in consent request. This is not a generic 'Thanks for contacting us' — it is a warm, specific message that reads like a senior team member sent it personally. The consent request is mandatory: PDPA requires explicit opt-in before any follow-up sequence fires.
AI Qualification
A Claude (via Anthropic API) or GPT-4o module reads the prospect's reply and extracts structured data: full name, company or profession, service interest, budget range, timeline, and a lead quality score (HOT/WARM/COLD). If the message is too vague, the AI sends one targeted clarifying question. This replaces 8–15 minutes of manual reading and qualifying per lead.
Smart Routing
Based on the AI output, a Router module branches the pipeline: HOT leads (ready to buy, clear intent) fire an immediate Telegram alert to the relevant sales rep with a one-tap WhatsApp deep link; WARM leads receive a Calendly booking link; COLD or unqualified leads get a generic resources response and a low-priority CRM tag. The sales team sees only the HOT alerts — not every message.
CRM Logging
Every inbound message — regardless of outcome — creates a record in Notion, HubSpot, or Airtable with: contact name, phone number, message body, AI-extracted fields, lead score, PDPA consent status, source (trackable via UTM parameter on your WhatsApp link), and timestamp. Within 90 days, you have clean data on which channels drive your best leads and what qualification signals predict deal closure.
The Full Tech Stack
A production-grade WhatsApp inbound automation pipeline for a Singapore SME uses the following tools. All are pay-as-you-go with no long-term contracts — total platform cost is typically SGD 80–200 per month.
WhatsApp Business API (Meta Cloud API)
Receives inbound webhooks and sends messages. Requires a dedicated business phone number registered through Meta Developer Console. Pricing: ~USD 0.00–0.02 per inbound conversation; ~USD 0.04–0.06 per business-initiated message.
The orchestration layer — connects all other tools. Make.com is recommended for most Singapore SMEs (visual, no-code, 1,500+ connectors, free tier). n8n is better for self-hosted PDPA data-residency requirements or very high volumes.
Claude (Anthropic API) or GPT-4o (OpenAI API)
The AI qualification layer. Claude is recommended for Singapore deployments — zero-data-retention policy, strong instruction-following, and well-suited to Singlish and mixed-language messages. Cost: ~USD 0.003–0.015 per qualification call depending on message length.
Notion, HubSpot, or Airtable
CRM destination — each inbound lead creates or updates a row automatically with all AI-extracted fields. Notion is the most common choice for Singapore SMEs (flexible, affordable); HubSpot for teams that need pipeline views and sales workflows.
Telegram (Lumora alert channel)
Sales rep notification channel for HOT leads. A Telegram message with the lead summary and a WhatsApp deep link fires within 90 seconds of the inbound message. Rep taps the link to open the WhatsApp conversation immediately.
Calendly or custom booking page
Appointment booking for WARM leads routed to a discovery call. Make.com sends the Calendly link in the WhatsApp message; a separate Make.com scenario fires on booking confirmation to update the CRM row and send a Telegram alert to the assigned team member.
PDPA Compliance for WhatsApp Inbound Automation
Singapore’s Personal Data Protection Act applies fully to automated WhatsApp systems. Every Lumora build includes the following compliance controls as standard — they are not optional add-ons.
Consent before follow-up
The first automated message must include an explicit opt-in consent request before any follow-up sequence fires. A simple 'Reply YES to receive further information from us' is sufficient. The Router checks for this confirmation before continuing.
Opt-out handling
When a contact replies STOP, UNSUBSCRIBE, or equivalent, a dedicated Router path flags the contact as Do Not Contact in the CRM and halts all active sequences immediately. This is enforced at the Make.com level — not relying on a team member to notice.
Data minimisation
Only collect what you need for the stated purpose. Standard inbound qualification collects: name, phone number, service interest, budget range, and timeline. Do not request NRIC, income documentation, or sensitive personal information in an automated flow.
Retention limits
WhatsApp conversation logs and CRM records for non-converting leads should have a defined retention period stated in your Privacy Policy. Lumora recommends 12 months for lead data from non-converting contacts. Make.com scenario logs auto-delete after 30 days by default.
Regulated industries
For MAS-regulated financial advisors: the automated system must not constitute financial advice. It should route, qualify, and book — then hand off to a licensed advisor for all substantive discussion. For MOH-regulated healthcare: the system must not ask clinical questions or make treatment suggestions. Build these as hard constraints in your AI system prompt.
Industry Applications
WhatsApp inbound automation has the highest ROI in industries where speed-to-lead is a direct revenue driver and enquiry volume is high enough to justify automation.
Financial Advisors & IFAs
Facebook and LinkedIn content drives a steady stream of inbound WhatsApp enquiries. Advisors using inbound automation handle 3–5x more enquiries per week without increasing admin time. AI qualification filters for investable assets, timeline, and existing coverage before any advisor time is spent.
Property Agents
PropertyGuru, 99.co, and Facebook property groups push enquiries directly to WhatsApp. High-volume agents managing 20–40 listings cannot manually respond within the 5-minute window. Automation handles the first response, qualifies budget and timeline, and routes urgent buyers to the agent via Telegram.
Medical & Aesthetic Clinics
Patients increasingly book via WhatsApp rather than phone. An inbound pipeline handles enquiries, sends the consent and intake form, confirms the appointment, and sends reminders — all without front-desk involvement, and without touching clinical content.
Insurance Agents
Inbound qualification filters for coverage type, existing policies, and decision-making timeline before any advisor call. Agents using inbound automation report spending 80% less time on unqualified conversations.
E-Commerce Brands
High-volume enquiries from Instagram, TikTok, and Facebook ads are automatically classified (order query, product question, returns request) and routed — either to an AI response or a human team member.
B2B Service Businesses
LinkedIn content and Google Ads drive inbound WhatsApp from decision-makers. AI qualification filters immediately for company size, budget, and authority level — so sales reps spend time only on genuinely qualified conversations.
What Results to Expect
Businesses that switch from manual WhatsApp handling to an automated inbound pipeline typically see the following changes within the first 30–60 days:
| Metric | Before Automation | After Automation |
|---|---|---|
| First response time | 3–4 hours | Under 90 seconds |
| Lead handling capacity | 30–50 leads/week | 150–200 leads/week (same team) |
| CRM completeness | 30–40% logged manually | 100% logged automatically |
| Qualification rate | ~40% | 70%+ (structured data capture) |
| Team inbox time | 8–15 hours/week | Under 2 hours/week |
| After-hours response | 0% (missed) | 100% (24/7 coverage) |
Common Mistakes to Avoid
✕ Running automation on a personal WhatsApp number
Fix: Meta's Terms of Service prohibit using the consumer app or standard Business app for automated messaging at scale. Always use a dedicated corporate number registered through the official WhatsApp Business API. Violations result in permanent number bans.
✕ Over-automating the conversation
Fix: The best inbound pipelines qualify and route — they do not try to close. The moment a prospect says 'Can I speak to someone?' or shows frustration, a human must take over. Build an escalation trigger that pauses automation and fires a Telegram alert immediately.
✕ No fallback for unexpected messages
Fix: Not every message fits your qualification flow. Build a catch-all Router path for messages the AI cannot classify confidently (below a threshold), routing these directly to a Telegram alert for manual handling rather than sending a confusing automated response.
✕ Skipping PDPA consent
Fix: Running follow-up sequences without explicit opt-in consent exposes your business to PDPC complaints. Build the consent request into the first automated message — it takes 30 seconds to implement and protects you permanently.
✕ Sending outside business hours
Fix: Automated messages received at 2am feel intrusive and damage trust. Add a time filter in Make.com — send auto-replies immediately but queue follow-up sequence messages for 8am–9pm SGT.
Inbound vs Outbound: Which Should You Build First?
Most Singapore businesses need both — but if you are starting from zero, build inbound first. Here is why:
Build inbound first if:
- →You already receive 10+ WhatsApp enquiries per week
- →Your team is spending more than 2 hours/day on inbox management
- →You are running paid ads that drive WhatsApp clicks
- →Leads are going cold before you get to them
- →You have no consistent first-response message
Add outbound once:
- →Inbound is stable and running reliably
- →You have a backlog of leads who enquired but never converted
- →You have enough historical data to know your best lead sources
- →Your team has capacity to handle the replies outbound generates
- →You have content (case studies, testimonials) to share in sequences
For a deep dive into outbound sequences, read our guide on WhatsApp sales follow-up automation for Singapore businesses.
Get Started
Ready to stop losing leads
to slow responses?
Lumora builds end-to-end WhatsApp inbound automation systems for Singapore businesses. Three builds per month. Apply now or book a call.