Why I Love Lightning Talks

I have a love/hate relationship with conferences, program committees, and speaker selections. I've been on both sides of the program committee, so I've seen the end result, but I've also been privy to the factors of selection. I won't rant too much about this love/hate relationship because I haven't coalesced my thoughts completely on the topic, but one think I can say is this: I love lightning talks.

I love lightning talks, but unfortunately I think they are often treated as second class citizens by event organizers.

Let me explain.

Lightning talks are a good way to test out new material and theories. For example, I've been researching conversation analysis as a pillar of chatbots and conversational software. I've not finished my research, but I have enough material to show some examples and expand upon some talking points. Recently, I gave a lightning talk on the subject at Rasa's L3-AI conference. I specifically applied for the lightning talk because I knew that the pitch was experimental, and there was no way I could fill 30 or 45 minutes without stretching myself too thin. Really, I just wanted to try the material out, and see if there was interest as I continued forward with my research. Lightning talks are a great format for this.

Lightning talks aren't easy. Many people think that lightning talks are just condensed versions of longer talks, but that's far from the truth. Have you ever tried explaining a concept effectively to somebody in 5 minutes? Lightning talks require you to be concise, but also informative. You need to have a clear understanding of the material, and know how to best deliver that message in a shorter format. Simply condensing a longer talk into a shorter version will not do. There is nuance required to keep the pertinent information intact.

There is a second part to this as well. Being concise in your explanation isn't enough. You also have to know what to tell the audience and what to hold back. In addition to using lightning talks to try out new material or offer elevator pitches, you have to leave the audience in a state of wanting more. There should be people coming up to you after the talk, trying to connect with you, asking questions because you've left them curious. If the talk doesn't do this, then you need to decide whether the issue was the talk or the material. If it's the latter, and this is material you're trying out, it gives you the ability to evaluate the material to see if it's still worth pursuing. If you think it's the former, then you know that you have to work on your delivery a little more.

Lightning talks are often misunderstood and handled poorly by conference organizers. They are seen as a consolation prize. You didn't get selected for a real talk, so here, we'll give you 5 minutes instead. That's the sign of poor conference organization. As explained above, lightning talks are not just condensed versions of longer talks, and if that's how a conference is utilizing them, then the conference will end up with a handful of poorly formatted talks with people speaking too fast.

As a result, lightning talks get a bad reputation, and are somehow seen as less than a "real" talk. Honestly, I don't accept lightning talks unless I specifically apply for them, or if I already have material that can be effectively communicated in that format. Too many speakers are okay with a consolation prize because they can put it on their speaker resume, but I don't believe that offers any real long-term value.

The moral of this rant is that if you're a conference organizer, don't sleep on the value of lightning talks: Stop treating them as second class citizens, because when you do that, you also tend to treat the speaker as a second class citizen, and that reflects poorly on your conference. If you're a speaker, don't settle for a lightning talk if your original proposal was rejected. Lightning talks are more than just condensed versions of other talks. Put effort into understanding what you can and should do in that shortened period of time.