Conference bingo

README.adoc

Conference bingo

Background

In many tech conferences, attendees are invited to rate the talk and/or the speaker from 1 to 5 stars. This type of ratings is interesting but has a few drawbacks.

The discussion started as a twitter thread with this french proposition.

• As a speaker, if you get a 1/5, you don’t always know why (and what to improve).

• As a speaker, if you get 5/5, you may think the audience liked the talk for the demos while they actually liked the diagrams explanations.

• As an attendee, you’re looking for talks to watch on YouTube. How do you chose between 10 talks that have a 4.5/5 rate?

• As a conf organizer, how do you know why this speaker/talk was liked or not?

• …​

Proposition

A different approach would be to provide (as an app/website) a bingo grid with ready to use feedbacks. Examples in english would be:

• I learn something

• Too fast

• Very interesting

• FUN!

• I loved the demos

• Hard to understand

• A bit boring

• I understood absolutely nothing

• Not deep enough

• Not enough demos/examples

• Too complicated

• Best talk ever

• …​

And the list goes on.

Open questions

I think each conference could come up with it’s own grid (variation on language obviously but also according to types of talks).

• Would a 3x3 grid or 4x4 be enough?

• Could we allow the speaker to provide a complimentary row of feedbacks like:

  • I liked the first part about X

  • The demo about Y was not very production ready

  • The slide deck was ugly

• How can we create some kind of hall of fame? (most interesting talk, best demos…​)

• Should we try to elaborate a rate from the grid?

What do you think?

"I learn something" -> "I learn something useful for me" :-)

Not sure if zooming on particular times of the talk wouldn't clutter the easing (from an UI standpoint I mean)
Or maybe allow to precise this as free comment field

Personally, not in favour of attributing rates on items. Allowing to show if an item is negative or positive is although something which should be implicitely tracked (negative feedback is generally a good area of improvement for speakers)

negative feedback is generally a good area of improvement for speakers

So true. Even when you get good ratings, there's something to improve. A talk is never done.

I don't like the idea (if I understood it) to let the speaker tune available choices. IMO they should be the same for everyone. I like the idea of a trend based on what the user has chosen. And if the trend is negative, I'd love a strong reminder that a comment is worthwhile. Too often we get 2/5 with zero explanation as the why...

Perhaps you could split in 2 grids :

• one for the money content
• two for the show

You could help attendees to focus on what they like/dislike on a talk.

I think the idea is to fix a limited set of choices (not making it tuneable by the speadker), at least on a per-event basis (in order to allow to compare talks from each other during the event)

Off topic : not sure if gist is a good place to talk about stuff ... I may be dumb, but I'm looking for a way to "subscribe to notifications" on gist for a few minutes and I don't see anything available except this

Yep, I haven't though about that :-(

We may also have one 3x3 grid for what the attendee liked, and one 3x3 grid for bad things or things to improve.

As a speaker, it looks like a great and powerful tool.
We must keep it simple. Too few attendees give one single rating, so one or more 4x4 bingos... (conference organizers should gamify the rating in order to increase participation)

conference organizers should gamify the rating in order to increase participation

Interesting ;-)

I think the #1 problem to fix is the feedback rate. Whatever the system is, conclusion can't be relevant when only 10% of the audience participate. Therefore, 👍 for gamification and/or rewarding feedback.

I already try the gamification and it doesn't work for the moment... Even if i offer something cool, people don't take the time...
The moment where we get the most feedback was when i ask to attendees after the event. But definitly your idea is good and there is something that we have to try.

Yes, I've noticed if you don't remind the audience to vote, they usually are less inclined to do so. +1 for an entertaining experience.

What I have seen working for long conferences (>45min) is a sli.do shared by QR code. It helps people interact because they find it funny to ask questions, especially in an anonymous way. Speaker would make 2~3 sli.do breaks to answer top 3 best graded questions.

And also, sli.do has a survey mode.
Thing is, sli.do has to be created conference by conference, speaker by speaker, not at a whole global conference scale. (I thought about that because it can be used as a source of inspiration)

@ncomet I've seen someone using sli.do this year at devoxx and it's definitely something I'd like to try. But as you said, I don't think it's really suited for talks below 1 hour.

The Mix-IT conference used the idea with a paper poster format and color stickers 😍

https://twitter.com/guillaumeehret/status/855021145191534592