Skip to main content

Feature Request: AI-Powered Q&A Categorization - Would This Help Your Sessions?


  • Participant
  • 1 reply

Hi Slido Community! 👋

I'm an IT consultant who frequently attends webinars and training sessions that use Slido for Q&A. I've been in several sessions with 100+ participants where I noticed something interesting...

What I'm observing as an attendee:
- Questions come in faster than presenters can read them
- Similar questions get asked multiple times in different ways
- Presenters seem overwhelmed trying to address everything
- Great questions get buried in the stream and never answered
- Sessions end with lots of unanswered questions

Recent example: 
A session with 125 questions within an hour. I watched the presenter struggle to keep up, and I could see clear themes emerging that weren't being addressed systematically.

My question for presenters in this community:
- Is this a real problem you face?
- Do you feel like you're missing important questions during live sessions?
- Would having questions automatically grouped by theme be helpful?

As an IT person, I'm wondering if there's a technical solution here, but I don't want to build something nobody actually needs. 

Have others noticed this pattern? What are your experiences from the presenter side?

Looking forward to hearing your thoughts! 🤔

-Sri

3 replies

Michal from Slido
Forum|alt.badge.img

Hello ​@SriB,

Thanks for the thoughtful post! 🙌 Your observations align with common challenges in large Q&A sessions. As a presenter, I’d say yes, this is a real issue—questions flood in quickly, duplicates pile up, and great ones get buried. It’s tough to prioritize and address key themes in real time, often leaving valuable input unanswered.

We have two features that could help you with this.

  • Similar questions detection
  • Q&A Labels

You can turn Similar question detection on in Slido labs. This is a participant facing feature, which notifies participants if a similar question was asked, while they type their question.

 

Q&A Labels

As a moderator facing feature, you, or a cohost in the background can go trough the questions and assign labels to them, the labels are a free text format tag. So the moderator can categorize questions any way they want. They can add labels based on topic, or even, a label called - high priority, to make sure the question isn't missed.

Learn more about that feature here.

I hope this helps :)

Best regards,


  • Author
  • Participant
  • 1 reply
  • June 9, 2025

Hi @Michal,

Thanks so much for the response and for confirming this is a real pain point! The similar questions detection feature is definitely a step in the right direction for reducing duplicates👍.

However, the bigger challenge I've observed is that even with duplicate reduction, presenters still struggle with the cognitive load of processing and prioritizing large volumes of questions in real-time. The current features help with input management, but don't address the question organization and session intelligence challenge.

What I've been experimenting with:

I actually took the questions from one of my recent sessions and ran them through an AI categorization tool to see what would happen. The results were eye-opening:

Before: Long chronological stream of questions
After: Organized into logical categories like "Technical Setup," "Course Content," "Scheduling," etc. with themes and insights

The transformation took 30 seconds and suddenly made it possible to:

  • Address question themes rather than individual questions
  • Spot knowledge gaps in the content
  • Create a professional summary for post-session distribution
  • Ensure no important topics were missed

Potential integration idea:

What if Slido is to have an "AI Assistant" button in the presenter dashboard that could:

  • Auto-categorize questions by theme
  • Show real-time session pulse - what's the audience actually confused about?
  • Help presenters read the room and adapt in real-time
  • Gauge engagement trends - are people asking basic questions or advanced ones?
  • Help presenters pivot content based on emerging question patterns
  • Identify knowledge gaps to address before moving forward
  • Generate a structured Q&A document for post-session sharing

The real value during live sessions - AI Assistant feedback:

  • 🚨 "Alert: 60% of recent questions focus on tool access issues - consider addressing setup now"
  • 💡 "Insight: Question complexity suggests audience is ready for advanced content"
  • 🔍 "Pattern detected: Multiple calendar sync questions - recommend dedicated troubleshooting segment"
  • 📈 "Trend shift: Questions moving from technical to conceptual - audience may need slower pace"
  • 📊 "Engagement pulse: High activity in Q&A but low poll participation - consider interactive element"

This would give presenters situational awareness and session intelligence - like having a real-time sentiment analysis that helps the presenter make tactical decisions about content flow and pacing.

This would complement your existing features perfectly - similar question detection reduces input chaos, while AI Assistant provides presenter coaching intelligence and helps with meeting organization.

I'd be happy to share the actual before/after examples if that would be helpful for your product team to see the concept in action. The AI processing could leverage existing services (like OpenAI's API) so wouldn't require building ML infrastructure from scratch.

I think it could be a compelling addition to Slido's presenter toolkit.

Thanks again for the great work that you all do!

Best,
Sri


Forum|alt.badge.img

Hi ​@SriB.

Thank you for your generous feedback and for your kind words - we will be sure to pass on your experience to our team.

I can see how those ideas would be a great assistance with making the most out of a large participant size, and we certainly like to see high engagement.  Your points would definitely make it easier to conduct these type of sessions, time after time, and synthesize some powerful data for you.

Our current AI tools are centered more around helping you to develop your content, and I’m afraid the similar question detection feature is the most help we can offer at this time.

  


Reply


cookie symbol.

Your privacy matters.

We use cookies to improve your experience, analyze traffic, and serve personalized ads. .