Uploaded your survey responses, NPS data, or customer reviews? Great, now it’s time to turn them into clear, actionable themes. That’s where Caplena’s Prompt-Based Topic Generation comes in.
Whether you’re building topics from scratch or enhancing what you already have, this guide shows how to use natural language prompts to steer the AI, with you firmly in control.
How Topic Generation Works (Without Prompts)
When you open a project, Caplena automatically suggests:
-
Categories (e.g., Price, Quality, Performance)
-
Topics within each category (e.g., Affordability, Picture Quality)
A solid starting point, but for more control, that’s where prompt-based generation comes in.
What Is Prompt-Based Topic Generation?
This feature lets you talk to the AI using everyday language to generate topics based on your input.
You’ll find the Prompt-Based Topic Generation toggle at the top of the Topic Editor.
Examples of What You Can Ask
-
“Give me only topics related to contracts”
-
“Group feedback into: Price, Features, Service, Other”
-
“Find all brand names mentioned”
-
“Suggest new themes based on smart TV features”
Playground vs. Entire Project — Choose Your Mode
Playground Mode:
Apply your prompt to a small sample, ideal for safe testing.
Use this if you want to:
-
Try out ideas
-
Manually review and add topics
💡 Example:
“Return all topics about features.”
Entire Project Mode
Applies your prompt across the full dataset and rebuilds the topic structure.
Use this if you want to:
-
Start fresh
-
Apply a new topic framework
💡 Example:
“Group feedback under: Pricing, Usability, Performance, Design.”
“Only” vs. “Also” — Choose Your Prompt Scope
✅ Only
Topics are based only on your prompt — useful for focused exploration.
“Only return contract-related issues.”
➕ Also
Includes your prompt plus commonly mentioned topics.
“Explore installation issues and also show top-mentioned topics.”
Summary
