HomeWhatsApp APIAI Conversational Forms

AI-Powered Conversational Forms via WhatsApp

No rigid forms. Your staff describe what they need in plain language — our AI extracts the structured data automatically. Supports English, Bahasa Malaysia, and Manglish.

💬 See the AI in Action View All WABA Solutions
💬
WhatsApp Business API — Billing Overview
All WABA messages are charged per message by Meta, except for Service messages (user-initiated conversations) which are charged per 24-hour session — meaning unlimited messages within that window at one session fee.
Message categories: Marketing (per message) · Utility (per message) · Authentication/OTP (per message) · Service/Customer Service (per 24-hour session, user-initiated only). Pricing varies by destination country. Contact us for a quote.
Live Example

One Message. 12 Structured Fields. Automatically.

See how a single natural-language WhatsApp message is transformed into perfectly structured database records — no forms, no manual entry.

👤 Staff sends this WhatsApp message:
"Hi, I'm Ahmad bin Ali, staff ID P001, IT dept, 35 years old, male, Malay. I stay at No 12 Jalan Maju, Puchong. My car is WKL1234, Perodua Myvi, 3 seats. Route from Puchong to the office."
⬇ AI Extraction Engine
🤖 AI extracts 12 structured fields automatically:
full_name→ Ahmad bin Ali
staff_id→ P001
department→ IT
age→ 35
gender→ Male
race→ Malay
home_address→ No 12 Jalan Maju, Puchong
plate_number→ WKL1234
vehicle_model→ Perodua Myvi
seat_capacity→ 3
route_origin→ Puchong
route_destination→ Office
Under the Hood

How the AI Extraction Works

Six steps from natural WhatsApp message to validated, stored database record.

1

Staff Sends a Free-Text Message

Staff types naturally — exactly as they would tell a colleague. No form fields, no dropdowns, no character limits to worry about.

2

AI Reads and Identifies Data

The AI model reads the message and identifies what data is being provided — recognising names, numbers, addresses, vehicle details, dates, and intent.

3

AI Extracts Structured Fields

Every identified piece of data is extracted into named structured fields — full_name, staff_id, age, plate_number, route_origin, and more.

4

Validation Against Business Rules

The PHP backend validates every extracted field — age between 18–70, plate number minimum 3 characters, all mandatory fields present. Clear rules, automatic enforcement.

5

System Asks Only for Missing Pieces

If something is missing or fails validation, the system asks only for that specific field: "Thanks Ahmad! Just one more — could you share your home address?" — not restart the whole process.

6

Data Stored, Confirmation Sent

Valid data is written to the structured MySQL database and an instant confirmation is sent to the staff member via WhatsApp — all in real time.

AI Capabilities

What the AI Can Detect & Extract

Covering all common staff data fields — personal, vehicle, operational, and intent detection.

👤 Personal Identity

  • ✦ Full name & staff ID
  • ✦ Department / division
  • ✦ Age, gender, race
  • ✦ Nationality
  • ✦ Home address
  • ✦ Email address

🚗 Vehicle & Transport

  • ✦ Vehicle plate number
  • ✦ Vehicle make & model
  • ✦ Engine capacity
  • ✦ Seat count
  • ✦ Road tax expiry
  • ✦ Insurance expiry

📅 Operational & Scheduling

  • ✦ Dates (incl. "tomorrow", "next Monday")
  • ✦ Times (incl. "8 in the morning")
  • ✦ Route origin & destination
  • ✦ Pickup point
  • ✦ Number of passengers
  • ✦ Intent detection (register / update / cancel / query / complain)
🌐 Multi-Language Support: The AI understands inputs in English, Bahasa Malaysia, and Manglish (mixed Malay-English) — the most common form of communication among Malaysian staff.
Real Examples

Real-World Extraction Examples

How staff actually communicate — and how the AI handles it.

🔄 Simple Update
"update my age to 43"
age → 43
✅ Updates age field in database, confirms via WhatsApp
🚗 Vehicle Registration
"my plate is WGT 3214, Toyota Vios, can fit 3 passengers"
plate → WGT3214
model → Toyota Vios
seats → 3
✅ Creates or updates vehicle record instantly
🚘 Trip Posting
"post trip tomorrow 7:30am, from Subang to KL Sentral, got 2 empty seats"
date → tomorrow
time → 07:30
origin → Subang
destination → KL Sentral
seats → 2
✅ Creates ride record and notifies matching riders
⚠️ Complaint Detection
"the driver didn't show up at all this morning"
intent → complaint
indicator → no-show
✅ Routes to complaints inbox, flags for admin review
Technical Architecture

What Powers the AI Extraction

Enterprise-grade AI integrated into a production WhatsApp chatbot environment.

💬WhatsApp Business Platform (Meta Official API)
🤖AI LLM Integration (Qwen, GPT, Claude compatible)
⚙️PHP Backend Validation Layer
🗄️MySQL Structured Database Storage
🔒HTTPS REST API — All Layers Encrypted

See the AI in Action — Request a Live Demo

We'll demonstrate real AI extraction from natural language WhatsApp messages — in English, Bahasa Malaysia, and Manglish.

💬 Request a Live Demo View All WABA Solutions

📞 1800 87 7061  ·  +603 2780 3880  ·  [email protected]

📱 SMS Coverage — All Countries Worldwide

A B C D E F G H I J K L M N O P Q R S T U V Y Z