How to Analyze an Image from WhatsApp with Google Cloud Vision & WOZTELL (Make)
How to Analyze an Image from WhatsApp with Google Cloud Vision & WOZTELL (Make)
This tutorial explains how to automatically analyze images received on WhatsApp using Google Cloud Vision via Make.
Perfect for extracting text (OCR), objects, or other metadata from WhatsApp images for support, automation, or reporting.
Overview
Your scenario uses two main modules and a filter:
1. WOZTELL | Unleash WhatsApp – Watch New WhatsApp Message: Detects new WhatsApp messages (including images).
2. Filter: Only continue if the message contains an image.
3. Google Cloud Vision – Run Text Detection (OCR) within an Image: Analyzes the image to extract text or other features.
Visual flow:
Step 1: Watch New WhatsApp Message (WOZTELL)
1. In Make, add the module: WOZTELL | Unleash WhatsApp – Watch New WhatsApp Message.
2. Connect your WOZTELL account.
3. Select the WhatsApp channel/inbox to monitor.
4. This module triggers for every new incoming message (text, image, etc.).
Step 2: Filter for Image Messages
1. Add a Filter module after the WhatsApp watcher.
2. Configure the filter to only pass messages where the content type is "image" (or where an image URL/attachment is present).
Step 3: Analyze Image with Google Cloud Vision
1. Add the module: Google Cloud Vision – Run Text Detection (OCR) within an Image.
2. Connect your Google Cloud account.
3. Map the image URL/content from the WhatsApp message to the Vision module input.
4. Choose the analysis type (e.g., OCR for text extraction, object detection, etc.).
Best Practices
- Make sure your WhatsApp channel is configured to receive media messages.
- Test the scenario with different image types to ensure robust extraction.
- Use error handling in Make for cases where images are not supported or Vision API fails.
- Store or log the extracted data for further processing (optional).
Result
Once activated, every image received on WhatsApp will be automatically analyzed by Google Cloud Vision, extracting text or metadata for downstream automation or reporting.
Next Steps
- Forward the extracted text to Slack, email, or a database.
- Expand the flow to handle multiple image types or advanced Vision features (labels, objects, etc.).