How to Generate Google Shopping Feed Data with AI in conbase.ai

Introduction

• The Goal: Fill Google Shopping feed attributes for thousands of products in minutes, so you can launch or update campaigns fast.
• The Problem: Entering titles, descriptions, categories, and GTINs by hand is slow, produces errors, and stalls your marketing plans.
• The Solution: Conbase.ai lets you build a prompt-driven pipeline that writes every feed field for you, tuned to your catalog and brand voice.
• Prerequisites:
  • A Conbase.ai account
  • A CSV with product data (columns like sku, product_name, description, price, brand, image_url)
  • Basic knowledge of Google Shopping feed rules

Are you here for the first time ? Then watch a full conbase.ai workflow in action

This video walks you through a complete conbase.ai workflow, from input to final output, so you can see exactly how the platform works and how to get the most out of its features.

Step 1: Set Up Your Project and Import Data

Create a New Project

Log in, click “New Project,” and name it “Google Shopping Feed – 2025-05-xx.”

Upload Your Product CSV

Drag in your file. Conbase.ai shows a preview so you can confirm column headers.

• Keep column names short and clear (title, price, image_link).
• Remove stray characters before upload.
• Add extra context columns like material or color if they help the AI write stronger copy.

Step 2: Build Your AI Generation Pipeline

Add an Action Step

Select “Create New,” name the step “Generate Feed Data,” and add it to the pipeline.

Create Your Three-Part Prompt

Part 1: Instruction

You are an e-commerce data specialist. Write Google Shopping feed fields that meet Google policies and use friendly, concise language.
Format:
 title: <70 characters
 description: 2 sentences, 300-500 characters
 google_product_category: numeric ID
 condition: new
 availability: in stock
Do not add promotional text or excessive punctuation.

Part 2: Context

Product Name: {product_name}
Description: {description}
Brand: {brand}
Price: {price}
Material: {material}
Color: {color}

Part 3: Output

Output Column: title
Output Example: Wireless Noise-Canceling Headphones – Black

Output Column: description
Output Example: Enjoy clear sound on the go with these black wireless headphones featuring active noise canceling, 30-hour battery life, and quick-charge support.

Output Column: google_product_category
Output Example: 190

Output Column: condition
Output Example: new

Output Column: availability
Output Example: in stock

Save the step.

Step 3: Configure and Run the Pipeline

Choose AI Settings

• Model: "GPT-4o"
• Processing Mode: "Instant" for testing and for fast execution, "Scheduled" for large batches and 50% less API costs
• Integration: "Fill if empty" leaves available data untouched

Test, Review, and Launch

1. Select 10 rows and click Run.
2. Open the results, spot-check titles and categories.
3. Tweak the prompt if you see odd phrasing or wrong IDs.
4. When happy, select all rows and run the full job.

Step 4: Review and Export

Quality Check

• Scan random samples for policy compliance.
• Confirm titles under 70 characters.
• Verify category IDs match Google’s taxonomy.

Export as CSV

Click Export, download the enriched file, and load it into Merchant Center or your feed management tool.

Conclusion and Next Steps

You just automated Google Shopping feed creation. Repeat the same pipeline whenever you add new products, or clone it to handle feeds for other channels like Facebook or TikTok.

Want to explore more? Book a free Conbase.ai demo and see how AI can tidy up the rest of your catalog work.