Audience feedback on online conference platforms: a speaker's view

We are in year 2 of the Covid-19 pandemic, and open-source conferences are still, for the most part, online-only events. (And I find myself questioning the judgment of those that put on large in-person conferences.) I have spoken at several of such conferences, and I’d like to zero in on one aspect of my personal experience, as a speaker, on some of the conference platforms I worked with.

Now, I should explain one thing up front. As a speaker, the thing I care the most about is:

Is this talk useful to you?

That’s it. That’s the paramount question. I want you to take something away from the talk that you find useful. Whether that’s a technical insight, or a new angle on a problem, or even just entertainment, I want there to be something in the talk for you.

And it’s incredibly important for me to get a sense of that, as I am delivering the talk.

Now in an in-person event, that’s easy: all I need is a look across the room, and I permanently look across the room. I can tell if you’re making eye contact, or listening intently, or nodding your head, or even putting on a face that makes it clear that you’re violently disagreeing with me. Sure, if you raise your hand and ask a question, or heckle me, or laugh at a joke, that drives the point of your engagement home — but I don’t need you to become so explicitly engaged, to know that you are engaged.

And it is this kind of feedback (that you might not even realize you’re giving me!) that makes the difference between delivering a talk, and just speaking into the void. It’s also the difference between delivering a conference talk, even if it’s a pre-recorded one, and just uploading the video on YouTube. If an conference platform doesn’t give me that kind of feedback channel, the work that I put into writing, rehearsing, and recording/streaming the talk is better spent with a view toward upload to a video hosting platform, and engaging with viewers there.

So, I as a speaker am foremost interested in a single thing about the conference platform:

Is it easy for you to tell me if my talk is useful to you?

And I’m not talking about you getting into a chat and saying “this is useful.” That’s much too high of a threshold. How often do you sit in a talk and then tell the speaker, “hey, that’s useful”? Quite rarely, and only if you find the content especially actionable or insightful. Because you normally don’t have to tell me explicitly: if you’re actively listening to me, I can tell. And I know you wouldn’t be listening to me if I talked useless nonsense.

And this is one of the reasons why I like Venueless so much. Venueless has implemented an extremely low-threshold feature of showing audience engagement. You get a handful of emoji like ❤️👏🤣👍🤔 that you, as a viewer, can click on, and then they appear in the event chat stream like “emoji rain” from the top of the screen. Just one click, which also gets anonymized and lost in the crowd — this last bit is important for people who are shy or reserved, and don’t like to stick their head out. And this makes it so much easier for you to engage, than having to actually type a sentence (or even a word, or even typing an emoji) into a chat channel.

I can not tell you enough how much of a difference this makes to the speaker experience. Add to this that the event chat is not shackled to Slack or any other horribly overgrown not-even-really-chat platform anymore, and you’ve got a simple, easy-to-use, no-unnecessary-frills experience that puts you directly in touch with your audience. I loved this at PyCon AU last year.

For comparison, I also saw LoudSwarm at DjangoCon Europe. The way it was used in that conference it seemed very tightly tied to Slack, although the audience was very nice in using emoji reactions very generously. Still, it wasn’t the same quality of feedback that Venueless provided.

Voctoconf, which I saw at FrOSCon, allegedly predates Venueless and was excellent in terms of streaming, but in terms of audience interaction it’s essentially BigBlueButton on steroids, meaning it’s about on the same level as the LoudSwarm/Slack combination I saw at DjangoCon.