← All Docs

Data & Backend

Sample Data Generation

Empty apps are hard to evaluate. Webase's sample data generation uses AI to populate your data models with realistic test records, so you can see how your app looks and behaves with real content.

How Sample Data Generation Works

When your app has data models defined, you can ask the AI to generate sample data that matches your model's field types and structure. The AI creates records that feel realistic — not random gibberish, but contextually appropriate names, descriptions, dates, and values.

1

Open Your Data Model

Navigate to the data model you want to populate. You can do this from the Data Models section in the Application Editor.

2

Generate Sample Data

Click the Generate Sample Data button or ask the AI through chat: "Generate 10 sample records for the Contacts model." The AI will create records that match your field definitions.

3

Review and Refine

The generated records appear in your data model immediately. Review them to make sure they look right. You can edit, delete, or regenerate individual records as needed.

Custom Prompts for Better Data

You can guide the AI to generate data that fits your specific scenario by providing a custom prompt:

  • "Generate 15 contacts for a real estate agency in New York." — The AI will create contacts with names, phone numbers, emails, and notes relevant to real estate.
  • "Create 20 product listings for a vintage clothing store." — The AI will generate product names, descriptions, prices, and categories that fit a vintage clothing theme.
  • "Add 10 sample tasks with varying priorities and due dates spread across the next month." — The AI will create tasks with realistic titles, descriptions, and date ranges.

Tip: The more context you give, the better the sample data. Mention the industry, the type of users, the geographic region, or any other details that would make the data feel realistic for your use case.

What Gets Generated

The AI respects your data model's field types when generating records:

  • String/Text fields — Contextually appropriate names, titles, descriptions, and notes.
  • Number fields — Realistic values within sensible ranges (prices, quantities, scores).
  • Boolean fields — A mix of true and false values for realistic distribution.
  • Date fields — Dates that make sense in context (past dates for created_at, future dates for due_at).
  • Single Select fields — Values chosen from the defined options with natural distribution.
  • URL fields — Placeholder URLs that follow realistic patterns.

When to Use Sample Data

Sample data is useful in several situations:

  • Design testing — See how your app looks with real content instead of empty states.
  • Demo preparation — Fill your app with realistic data before showing it to stakeholders or investors.
  • Development — Test sorting, filtering, search, and pagination with multiple records.