Interactive marketing campaigns driven by major events like the World Cup are excellent lead generators. But manually receiving messages, validating predictions, logging data, and replying to thousands of users on WhatsApp is impossible for a small business team.

At Sawwikly, we solve operational bottlenecks by engineering automated workflows. In this article, we show you step-by-step how to build a fully automated World Cup WhatsApp Prediction Bot using n8n (an open-source automation platform), Google Sheets, and ChatGPT.

The Automation Workflow Architecture

Instead of writing complex code, we connect our components in a mechanical data pipeline:

graph LR
    A[WhatsApp Msg Received] --> B[n8n Webhook Trigger]
    B --> C[ChatGPT AI Parsing]
    C --> D[Google Sheets Logging]
    D --> E[Automatic WhatsApp Reply]

Here is how each node in the n8n pipeline functions:

Step 1: Receiving the Message (Webhook Node)

When a customer sends a message to your WhatsApp Business number (e.g., "I predict Argentina 2 - Brazil 1"), it triggers a Webhook.

  • In n8n, configure a webhook node attached to your WhatsApp API provider (such as Meta's Cloud API or local gateways). This instantly captures the incoming JSON payload containing the user's phone number and message text.

Step 2: Parsing the Prediction with ChatGPT (OpenAI Node)

Users write predictions in various ways: "Argentina 2-1", "2 to 1 for Brazil", or "تعادل 1-1". A standard program struggles to parse this inconsistent text.

  • We route the message content through an OpenAI node in n8n. Using a simple prompt, we instruct ChatGPT:

    "Extract the team names and predicted scores from the following message. Output the result strictly in JSON format: {team1: '...', score1: X, team2: '...', score2: Y}."

  • ChatGPT normalizes the data instantly, extracting the structured values.

Step 3: Logging to Google Sheets (Google Sheets Node)

Once the prediction is parsed, n8n executes a write operation to a secure Google Sheet.

  • The workflow searches for the user's phone number. If it is a new user, it appends a new row with their phone number, name, and prediction. If the user already submitted, it can update their row or prevent duplicate entries.

Step 4: The Automated Instant Response (WhatsApp Send Node)

Finally, the pipeline sends a WhatsApp message back to the customer:

  • "Thanks, [Name]! We have successfully logged your prediction: Argentina {{score1}} - Brazil {{score2}}. Good luck!"
  • You can automatically attach a custom tracking URL with a minor discount code to encourage them to browse your services while they wait for the match.

Why Automate?

By building this workflow on n8n, you eliminate human data-entry errors, manage thousands of users simultaneously, and capture high-intent leads automatically.

Marketing should not be a manual burden. With a clean, automated engine running in the background, you can sit back, watch the match, and let your systems build your customer database for you.