Data sources

Train the AI with data about your store, products, services to provide better and more accurate responses

Who can use this feature?

• This feature is available for all users

Our AI assistant is still in beta version. If you have any concerns or feedbacks, please contact us to share and help us improve Chatty.

Why add data sources?

We've trained AI on your own data. By adding your own data sources, you can train AI assistant about your specific products, services, and business policies.

This helps the AI give more accurate and personalized responses to your customers.

How to train AI

How to add data sources?

1

Go to AI assistant tab

2

Add data sources

◾ FAQs: By turning on FAQs, all questions and answers in your FAQs will be imported to data sources and used as AI training data

◾ Products: By turning on Products, all product information on your store will be imported to data sources and used as AI training data.

You can manage which products are used to train AI. In Products, click Manage → Change status of each product Enabled/Disabled to show or hide it in the training data.

Currently, Chatty only works with published products that are in stock. Product subscriptions are not supported at this time.

◾ Question: Add more questions and answers about your store to train AI assistant. You can add a single question or multiple questions at once.

3

After building your data source, you can go to Test to test your AI assistant.

If you don't enable AI, you can still test AI assistant with data source you added.

4

You can edit or delete it from data source.

Available data sources

1. FAQs

Your existing FAQ content is used to AI training, helping the assistant answer common customer questions accurately.

2. Products

Product details from your store catalog are used to train the AI to answer specific product-related questions.

Your product information will be updated daily at 12:00 AM PST.

Available product information that is synced to Chatty:

• Specific product details (name, descriptions, images, vendor, type)

• Product variants (colors, sizes, styles)

• Price ranges and comparisons

• Product features and specifications

• Collection

• Product inventory

• Product & category metafields (coming soon)

Best practices for product information

1. Product description

   • Include detailed specifications

   • Mention compatible or complementary products

   • List key features and benefits

2. Variant organization

   • Use consistent naming for variants

   • Keep variant options updated

   • Provide clear variant descriptions

3. Price information

   • Keep pricing updated

   • Include any volume discounts

   • Maintain consistent price formats

3. Questions (additional information)

This is additional information you want the AI to learn, beyond FAQs and product details.

Best practices for questions

Add more questions when you need AI to answer frequently asked questions that aren't covered in your FAQs section. Focus on adding unique questions rather than multiple variations of the same topic.

For example:

✅ Good practice:

• Add one clear question about your return policy timeframe

• Add one question about gift wrapping service

❌ Avoid:

• Adding multiple similar questions about returns

• Creating variations of the same question

4. URLs (coming soon)

5. Files (coming soon)

Learn more about how to train & optimize AI with data sources