How to Build a Multi-Channel Messaging Hub

A multi-channel messaging hub unified inbox system routes messages from all channels into one unified inbox. I build these for teams that need their communication tools to do more than just send messages.

What You Need

• Node.js 18+
• API credentials for the messaging platform
• A server or VPS for hosting the bot

Step 1: Set Up the Project

npm init -y && npm install dotenv

import 'dotenv/config';

const API_TOKEN = process.env.BOT_TOKEN;
console.log('Bot starting...');

Step 2: Build the Message Handler

async function handleMessage(message, sender) {
  const text = message.toLowerCase().trim();

  if (text === 'report') {
    const metrics = await fetchMetrics();
    return formatReport(metrics);
  }

  if (text === 'status') {
    return 'All systems operational.';
  }

  if (text === 'help') {
    return 'Commands: report, status, help';
  }

  return null; // No auto-response
}

async function fetchMetrics() {
  return { leads: 23, spend: 800, cpl: 34.78 };
}

function formatReport(metrics) {
  return `Daily Report\nLeads: ${metrics.leads}\nSpend: $${metrics.spend}\nCPL: $${metrics.cpl}`;
}

Step 3: Add Automated Notifications

async function sendNotification(channel, message) {
  // Platform-specific send logic
  console.log(`Sending to ${channel}: ${message}`);
}

async function dailyReport() {
  const metrics = await fetchMetrics();
  const report = formatReport(metrics);
  await sendNotification(process.env.REPORT_CHANNEL, report);
}

Step 4: Schedule Regular Messages

import cron from 'node-cron';

// Daily report at 9am weekdays
cron.schedule('0 9 * * 1-5', dailyReport);

// Hourly health check
cron.schedule('0 * * * *', async () => {
  const healthy = await checkHealth();
  if (!healthy) {
    await sendNotification(process.env.ALERT_CHANNEL, 'System health check failed');
  }
});

async function checkHealth() {
  // Your health check logic
  return true;
}

Step 5: Add Interactive Actions

async function handleAction(actionId, userId, data) {
  switch (actionId) {
    case 'approve':
      await processApproval(data.requestId);
      return `Approved by user ${userId}`;
    case 'reject':
      return `Rejected by user ${userId}`;
    default:
      return 'Unknown action';
  }
}

async function processApproval(requestId) {
  console.log(`Processing approval: ${requestId}`);
  // Trigger downstream workflow
}

What to Build Next

Add an AI layer that answers questions about your business data in natural language. Team members ask the bot "what was our CPL last week?" and it queries the database and responds with the answer.

Related Reading