Ballpark supports several research methods, each suited to different goals. This guide helps you pick the right approach for your project.
Quick comparison
Method | Best for | Time to complete |
Survey | Quantitative feedback, quick validation | 2-5 minutes |
Task-based test | Usability testing on live sites | 5-15 minutes |
Prototype test | Testing Figma designs before building | 5-15 minutes |
Live session | Deep qualitative research, follow-up questions | 30-60 minutes |
AI interview | Scalable qualitative research, discovery | 10-20 minutes |
Surveys and questionnaires
Use surveys when you need structured, quantitative data from many people.
Good for:
Validating assumptions with data
Measuring satisfaction or NPS scores
Gathering demographic information
Quick pulse checks on features or ideas
Example: You want to know which of three pricing options users prefer. A multiple choice question with 100 responses gives you clear quantitative data.
Task-based testing
Use task-based tests when you want to observe how people use an existing website or product.
Good for:
Finding usability issues on live sites
Testing specific user flows (checkout, signup, search)
Benchmarking task completion rates
Comparing your site to competitors
Example: You ask participants to find and purchase a specific product on your e-commerce site. Ballpark tracks where they click and captures their spoken thoughts.
Prototype testing
Use prototype tests when you want feedback on designs before building them.
Good for:
Testing Figma prototypes with real users
Validating navigation and information architecture
Getting early feedback on visual design
Iterating quickly before development
Example: You upload your Figma prototype for a new onboarding flow. Participants click through while Ballpark records their path and confusion points.
Live moderated sessions
Use live interviews when you need real-time conversation and the ability to ask follow-up questions.
Good for:
Deep discovery research
Complex topics that need clarification
Building rapport with customers
Stakeholder observation
Example: You're exploring why enterprise customers struggle with a feature. A 45-minute live interview lets you dig into their specific context and workflow.
AI-moderated interviews
Use AI interviews when you want conversational depth at scale, without moderating every session yourself.
Good for:
Scaling qualitative research
Consistent interview protocols
Research across time zones
Discovery when you can't predict follow-ups
Example: You want to understand how 100 users approach meal planning. AI interviews ask your core questions, then follow up based on each person's unique answers.
Combining methods
Many research projects benefit from multiple methods:
Start with a survey to identify patterns in a large group.
Follow up with AI interviews to understand the "why" behind the numbers.
Use prototype testing to validate your proposed solutions.
Run live sessions for final feedback before launch.
Still not sure?
Think about these questions:
Do you need numbers or stories? Numbers point to surveys; stories point to interviews.
Is your product built yet? Unbuilt designs need prototype testing.
Do you need follow-up questions? Live sessions and AI interviews adapt in real-time.
How many participants do you need? Surveys and AI interviews scale easily; live sessions take more time.