Skip to main content

How to use PickFu's built-in AI tools

An overview of our AI features to help you build better tests, understand your results, and take action on your insights faster.

Updated today

PickFu's AI-powered features help you analyze results, extract key insights, plan next steps, and get expert guidance throughout the testing process. Here's an overview of each built-in AI tool and how it works.

PickFu AI chatbot

PickFu AI is your intelligent, context-aware research assistant – available in the sidebar on your dashboard, project pages, survey builder, and poll results pages. The PickFu AI chatbot understands the context of the page you're on and can help you:

  • Analyze results – ask questions like "What did respondents like most about Option A?"

  • Build better tests – get suggestions for writing clear questions, choosing the best poll type, and targeting the right audience

  • Understand team activity – check recent polls from your dashboard or project pages

  • Learn best practices – ask anything about PickFu polls and research methods!

Example questions to ask:

  • What elements of Option C did respondents dislike most?

  • What were the 3 most recent polls my team published?

  • How should I phrase my question for a packaging design comparison test?

  • What test should I run next based on the audience feedback?

The chatbot streams responses in real-time, remembers your conversation context, and includes links to relevant resources and help articles. You can start a new conversation at any time by clicking the pen and paper icon at the top of the sidebar.

Give helpful responses a thumbs up and unhelpful ones a thumbs down – your feedback helps us improve PickFu AI!

AI Insights

AI Insights appears on your poll results page (directly below the top section with respondent votes). It provides a brief overview of your results, plus three or more callouts of key insights, trends, and actionable recommendations based on your audience's feedback.

Instead of just summarizing what respondents said, AI Insights tells you what to do next – whether that's refining your messaging, testing a new variation, or doubling down on a winning concept.

AI Insights also suggests a relevant follow-up poll based on your results, complete with a direct link to create it. For example, if you tested product image variations, the AI might recommend testing your winning image against a competitor or exploring why respondents preferred it with an open-ended follow-up.

AI Insights is available to all PickFu users. Find it on your poll results pages.

AI Highlights

AI Highlights is a dedicated tab on your poll results page that provides in-depth analysis of respondent feedback. It includes two main sections:

  • AI Analysis – a comprehensive summary of patterns and trends in your results, plus hypotheses for future tests to run.

  • AI Themes – an organized breakdown of what respondents liked and disliked about each option being tested.

Use AI Highlights to get a thorough overview of your insights and ideas for what to do next – without having to read every response (although we still find that helpful too!)

Frequently asked questions

Are these AI features included in my plan?

Yes – AI Insights, AI Highlights, and the PickFu AI chatbot are available to all users for free, no subscription required.

Where can I access the PickFu AI chatbot?

Look for the "PickFu AI" tab in the sidebar on your dashboard, project pages, the survey builder, and your poll results pages.

What's the difference between AI Insights and AI Highlights?

AI Insights gives you a concise overview of results, calls out the most important insights, and links a suggested follow-up poll to run. AI Highlights provides deeper thematic analysis of what respondents liked and disliked about each option, plus hypotheses for further tests to run. Both appear on the poll results page but serve different purposes – AI Insights gives you actionable recommendations at a glance, while AI Highlights helps you dig deeper into audience preferences.

Did this answer your question?