Guide your AI analysis with research goals

A few weeks back, we introduced AI Thematic Analysis. You can now use AI to uncover themes and informative and actionable insights that emerge in your study data. To help you tailor this analysis further, we’re today introducing research goals, which allows you to provide extra context around what research questions you are trying to answer with your research.

How to guide your AI analysis with research goals

Set or update your research goals on the results page: Once your study is complete, head to the results page for your study. There, you’ll find a new research goals button where you can define your research questions (or revise them if you already set them when generating your study):

‍

Generate your AI analysis: Wondering’s AI will now consider your research questions when generating your AI analysis, giving you insights that are more relevant to your research goals:

Some tips for writing good research questions

As you write your research questions, here are some best practices to keep in mind:

Think of research questions differently from interview questions
Research questions are your overall aims—what you want to learn. Interview questions are the actual prompts participants will respond to in your studies. Use your research questions, not your interview questions, to guide Wondering’s AI analysis for best results.
Be clear, concise, and direct
When writing your research questions, imagine you are describing them to a colleage with no prior context. Formulate each research question so that anyone on your team can understand precisely what you hope to find out—no ambiguity or jargon.
Before finalising each research question, ask yourself: would a teammate understand exactly what we’re trying to learn from this? If not, refine it. That clarity will translate to sharper AI findings.
Don’t be afraid to experiment and iterate
If your initial research questions don’t yield the depth or focus you need, tweak them and let the AI re-analyse the data. Each iteration can uncover nuances you might have missed at first.

Here are a few examples of research questions written with the above principles in mind:

Research questions for understanding user satisfaction

Less effective ❌: “Do people like our product?”
Better ✅: “How satisfied are our current users with our product, and which specific features contribute most to their overall satisfaction or dissatisfaction?”

Research questions for assessing product usability

Less effective ❌: “Is our app easy to use?”
Better ✅: “What usability barriers, if any, do users experience in our mobile app, and how do these issues affect their overall engagement and likelihood to continue using it?”

Research questions for testing a new feature concept

Less effective ❌: “Would users like a new messaging feature?”
Better ✅: “How do frequent travellers view a built-in messaging service for coordinating trip plans, and how does their current method of communication influence their willingness to adopt this new feature?”

Research goals are available today, and you can now add your research questions to the study results page with just a couple of clicks. As always, let us know what you think and how we can continue to improve Wondering’s AI analysis to help you get the insights you’re after!

Author

Axel Thomson

Founder

Author

Continue reading

Build better user experiences with AI-driven user research