In this episode of Event Experience by Bizzabo, host Rachel Moore is joined by Chris Duke, Head of Brand Experiences at Glean, for an AI-powered masterclass designed specifically for event professionals. Based on a popular webinar, this conversation dives into practical ways AI can transform the event planning process, from writing campaign briefs and automating reports to visualizing booth designs and analyzing post-event feedback.
Chris shares real-world workflows using tools like ChatGPT, Claude, and Gamma, demonstrating how event marketers can streamline tasks, generate professional assets from simple inputs, and scale their impact. Learn how to build AI agents, prompt effectively, and safeguard sensitive data, all while embracing AI as a creative and strategic partner.
What you'll learn:
Enjoyed this content? Subscribe to our newsletter and get insights like these delivered straight to your inbox.
Welcome to Event Experience by Bizzabo, the podcast where we bring the best and brightest event experience leaders together to share stories, tips, and lessons learned from creating some of the world's biggest events.
I'm Rachel Moore, your podcast host. If you've chosen audio only for this particular episode, take my advice and watch this one on Bizzabo's YouTube channel because we're about to embark on an AI masterclass full of show and tell.
This special episode features the contents of our recent webinar with Chris Duke. Head of brand experiences at Glean and a hands-on approach to skill building. In using ai, Chris demonstrates various tools like ChatGPT, Claude, and Gamma for tasks such as generating event briefs, visualizing booth designs, and creating post-event analytics.
Ultimately highlighting AI's potential to save time and improve efficiency. Emphasis on experience in this episode of Event Experience.
[00:01:24] Rachel Moore: Hi everyone. We are so glad you're all here. I am Rachel Moore. I'm the host of Bizzabo Event Experience Podcast, and I'm also thrilled to be your guide through our webinars and including today's. If this is your first BBO webinar, you're in for something special. We're gonna go off the beaten path here a little bit.
This isn't just another talk about AI. Today is a guided workshop designed to help you build real skills right here in real time.
So, uh, please let me be the first to welcome you to the Event Pros AI Masterclass. We're all gonna be masters at the end of this. This webinar will be available on demand for your convenience.
You're gonna get much more of it if you're going live with us right now. We are gonna be sharing the resources that, uh, our guest is gonna walk us through today, so you'll have everything you need at your disposal to be an AI expert.
As we know, AI is reshaping event marketing at every level from planning and content creation to post event analysis, but we all know that tools alone don't solve problems, systems can solve them.
Today's session is all about turning AI into a strategic partner that helps you and your team do more with less. Can I get an Amen to that? If that's what everybody wants. And a quick disclaimer today is going to be a bit different in how we approach this session. I'm going to play the, explain it to me like I'm five buddy, as we witness the ways AI can and should be working for us in event planning.
So, just know I'm your avatar for being like, "Oh gosh, please make me understand this because I need it." With that in mind, our collective goal today is to equip you with knowledge and practical tips that are gonna help you first learn how to integrate AI across your entire event workflow. From ideation to post-event analysis to create lasting impact instead of just one-off wins.
Also, we're gonna discover how to write clear and effective prompts that produce consistent on-brand results while saving time and reducing revisions. This is gonna be my Achilles heel that I'm so excited about.
And we're gonna explore how to build and apply simple AI tools, templates and workflows that automate everyday tasks and help your team scale its output.
Now, without further ado, let's begin our journey to walk us through all of this. I am thrilled to introduce someone who's been putting this into practice with some of the most innovative event teams out there.
It is my pleasure to welcome Chris Duke head of brand experiences at Glean. Chris is the Head of Brand Experience leading global events, live productions, brand advertising campaigns, and executive and customer engagements.
He's previously driven event and brand initiatives at Greylock Venture Partners, Robinhood Rubrik, GoPro and Red Bull. Creating experiences that connect audiences through creativity and storytelling. His expertise spanned experiential marketing, sports marketing, and large scale brand activations.
On long event days, he's wearing Nike's with inserts.
Welcome Chris. Thanks for joining us today for this awesome workshop.
[00:04:35] Chris Duke: Yeah. Thanks for having me. I appreciate it. The inserts thing I just realized makes me sound like I'm getting old.
[00:04:40] Rachel Moore: Oh, please. I want everybody, if you wear, have inserts in your shoes that you wear. Uh, I, half of our family does. So, no those are lifesavers. So it's all good.
[00:04:48] Chris Duke: Back problem are real, so.
[00:04:50] Rachel Moore: That's right.
Well, let's get started. Um, I'm, I'm really excited about this. I'm sure we're all just chomping at the bit. Now, uh, before we do though, uh, we wanna dive into our key topics today. But let's start off by checking in with you, our dear audience.
We know. AI is a big topic. It's literally the topic, right? And everyone's journey with using it looks a little bit different. So we'd love to ask you all which areas would you like the most help with today?
So we've got a poll up there, creating event content with AI tools. Is that where you need your help? Maybe it's using AI for data insights and post-event analysis. Perhaps you need help with automating repetitive planning tasks. Those things you're like, oh my God, I do this all the time and spend so much time doing it. Can it take it off my plate?
Or maybe you wanna train teams to adopt AI effectively. You might have a team where you're like, y'all. We gotta be using this. Chris, do you have any guesses maybe for what might be the, the weak, the weak link for people as far as like what they're gonna vote for the most to get help with today?
[00:05:48] Chris Duke: I think the automating repetitive planning tasks, I think that's a part of our job in events that everyone like despises, but it's like necessary evil we have in our role.
I think the beauty of events though, is they are relatively like new and fresh a lot of time, but there are the same like year over year events. So I think the repetitive tasks is probably one of those things. But, personally, as someone who's like, leans more onto the creative side, I think like using making event content. The new creative tools that are coming out, like some of those things as well are really interesting.
So that's,
[00:06:17] Rachel Moore: let's not hold ourselves in, uh, you know, anticipation anymore. Uh, let's see the poll. Oh, you were right, you were that first one nailed it. The automating repetitive planning tasks, that seems to make the most sense. And if at least it's a great place to start, right?
Let me start freeing up more of that time so I can figure it out. Right?
A hundred percent.
Yeah. Okay. Well thanks everyone for answering the poll. The great part is we're literally gonna gonna tackle all of this today, so this is gonna be so great. So let us get into the discussion and I promise y'all. I'm talking the most right now, 'cause from here on out it's mostly gonna be Chris and he's gonna be, uh, taking over.
But first, let's, let's do, get something out of the way. The myth nay the threat that AI is going to take jobs away from the people who design and execute events. And yes, I'm talking to you dear audience, as Chris is going to demonstrate today.
AI. Is not your jobs dealer, it's your assistant. And quite frankly, it can be your first go-to resource when you're about to tackle crucial aspects of your work. Uh, and that my friends probably requires a bit of a mind shift. We need to do together. Be like, okay, really can it really do this? And with that, I'm gonna say it's time for masterclass.
Chris, take it away. I think you have stuff to share with us today and I'm gonna let you take it. Go for it.
[00:07:29] Chris Duke: Definitely. Yeah, we're gonna get through all of that. Um, we're gonna talk about, um, those mindless tasks. We're gonna talk about agents, we're gonna talk about pretty much everything across the board and even in getting into some of the creative suite stuff that I talked about a little bit.
So excited to, uh, get into it
[00:07:43] Rachel Moore: and, and I think you're gonna show and tell a bit. One thing to, to show us right off the bat, like, I mean, it's, it's morning for me. I don't know what time it's for people based on your time zones and stuff, but let's all like think about like, okay, when I first start my day, do I think about AI right out the gate?
I mean, do, do you do that? Do you start your day with AI?
[00:08:00] Chris Duke: I have been, you know, it's, this is a, a shift that we all need to take, right? We've used AI tools like ChatGPT and Claude, and some of these other tools to like work. We try to like speed up our work, but the mind shift set like that we need to like move to is how do I become maybe AI first? And so I need to start rethinking the way that I'm gonna start my day even
As well as getting through the rest of my day. So it's, it takes a little time, it takes practice to kind of get into the mindset that I, "Hey, I need to shift the way that I work."
So for me, the one thing that I do now is I start my day. Using the agent which we'll talk about agents in a little bit, but the start my day, I'm actually not gonna run this because it has all my, like, private information on it, but essentially what I do is I go here, Hunter is on our sales team here. He made this agent, he created it.
And basically what it does, it takes all my information that I have throughout the day and it sends me a Slack DM or an email. With everything that I have going on, the objectives of those meetings, it'll look at my Slack messages., It'll look at, uh, my emails as well and it will curate basically everything that it thinks is high priority for the day. And that's how I start my day.
I also use things like, the Glean, daily Digest. Again, I work at Glean, so the tools that I have available to me are Glean. I also have access to ChatGPT and Claude and some of the creative tools as well. But I use Glean a lot 'cause that's the tool that I have internally 'cause. That's where I work.
You may have different tools at your companies. You may have different access to different things. So again, take advantage of what you have. But I use the Glean Daily Digest and Slack, which gives me a, uh, while I'm driving. I have a, like almost a two hour commute every day to get to work. And so I listen to my Daily Digest and Slack and it gives me a voice, uh, a voice note of all the slacks and things and channels that I have going off that I missed maybe throughout the night or the day before.
So I look at that and then I run this agent to start my day.
[00:09:43] Rachel Moore: I, I would, yeah. So, so as far as that goes too, so like you said, it's integrating into all these different tools, and I know we're gonna get into that a little bit. 'cause that just sounds, I mean, that sounds super complex to me. And remember, I'm, I'm the, like, I'm five person, but like, I'm assuming then it, I, I, I wanna ask a question.
I mean, it's like, okay, well I don't know how to maybe integrate all those things. Does AI like help you integrate things together so it actually talks to all those systems together?
[00:10:10] Chris Duke: Definitely these like, um, enterprise AI tools especially do a lot of the connecting by themselves. You use API connections through different, you know, Salesforce, Slack, Gmail, Google Docs, whatever that is.
And so your IT team, if you're in an enterprise, you need to have an enterprise account for some of this stuff. Yeah, things like ChatGPT and Claude. I think the word context is being used a lot right now, and context is really important for AI Whether you're prompting it correctly. Um, you need to give AI context and sometimes it's built in natively.
So like a tool like Glean or any of these other enterprise AI tools will have context to your company data. But if you're using just like a, you know, personal account for ChatGPT or Claude, you definitely need to give it context because it doesn't have access to those systems natively.
[00:10:53] Rachel Moore: Got it. I have a feeling you might, this might come into play too, as you're gonna show us some stuff. Because I, I think what we wanted to see was if you could show us a use case in real time.
This is a great example though, because I think this, this, uh, does kind of tackle, you know, the, the daily, uh, Hey, tell me what my day's gonna look like. Who doesn't want that? It feels like, oh my gosh, I've, I've got like my own personal assistant that's helping me prioritize and stuff, but I also wanna ask for that use case in real time.
Think something that a lot of our, um, our audience probably deals with are campaign briefs. Like, let me give you a scenario like, uh, if, if a, a company CMO asks a person in this, in this audience, Hey, create an event campaign brief, can, um, I need to simulate attendee journeys. I need to craft some tailored messaging for the specific campaign, and I even need to visualize booth designs.
Uh, can you show us how to be AI first in approach? Like if we get asked for that event campaign brief, how to proceed.
[00:11:48] Chris Duke: Sure. Yeah. When we're talking about those like repeatable tasks that we do every day, making briefs and, and creating messaging docs is all part of our role. In events, you're not just logistics. You're not just in the ERCs checking boxes and putting in, you know, uploading files. Like you own the event, you own the project. And so we also need to do a better job at taking the time to step outside of our comfort zone a little bit and learn how to write briefs, work on messaging. Like, be take the first pass at all of this, and AI really helps us do that.
So, again, I'm using Glean here, but you could use ChatGPT or Claude or anything. The benefit of using an enterprise tool as well is there is context as here, so I don't have to give it as much because it has access to all of my planning docs, my slides and everything that I've already worked on.
But if you were in ChatGPT for example, you might wanna add more information. You may wanna, uh. Attach a file or something like that, that may have like some, uh, some planning slides or any other information that you can give it. So the context, again, really important.
But the one thing that I'm gonna do, because I could spend two hours writing a first pass at a brief, just kind of staring at my screen, wondering what words are gonna come out. Basically what I would do is I would prompt this, uh, again, prompting is really important and we'll talk about it as, you know, throughout the day.
But essentially, I need to give the AI tool a job, which sounds kind of weird, but I want this to start thinking about it as I, you know, you're a, a member of the strategic events team at Glean, that's the lens in which I need this to operate through.
I'm gonna write a sample event brief for AWS Reinvent. That's an event that's coming up for a lot of event marketers that work in B2B. I need to map attendee journeys, craft tailored messaging and visualize booth designs. Not every tool in large language model has access to creative tools.
So again, what you'll see is there's actually, and we will talk about this, that humans need to, uh, intervene with these tools a lot because what I'll show you is that we don't have access to image generation right now. But, um, given that these things take time to run and we are limited on time here, I ran this one already ahead of time.
So, basically what it does is it writes it a whole brief for me about AWS reinvent, who the audience framework is. Tailored messaging, role-based value messages, attendee journeys, content and offerings, booth designs. So booth design concept. This is just a random screenshot it pulled, it's called a hallucination. It's like this is not a boot design. This is a screenshot from my user conference that I run that's running on demand right now, so that doesn't make any sense. So that's completely unnecessary. But I would rather have this first pass with some kind of junk in there where I can then hop over, I can ask it to build a Google Doc with headings, and then I click share, export the docs, and it would generate this Google Doc for me, now that I have a full pass at a brief, it took two minutes.
There's a lot of stuff in here I need to fix, but again, I'd rather be in edit mode than starting from scratch. Even just like formatting it with all the tabs, like it's all documented nicely over here. It's organized and it's well thought out and um, but there's some stuff in here that I need to fix.
So again, editing is better than creating. In the day and age where we need to work fast and we have less head count and the days go by quick and we don't wanna work till 10:00 PM every night. So again, this is a good example of that. So again, good example of a, a brief in, in five minutes.
[00:15:01] Rachel Moore: No, that's, that's really good.
I wanted to ask you too, so like you, you mentioned prompts already, um mm-hmm. And context. I know you've mentioned that a few times. Can you guide us through, because this is all great. I mean, it's, it's great to get this, but I think we're all then worried about like, well, great.
Should, can I turn around even if I Yeah. I'm gonna tweak it and edit it. How much more tweaking do I have to do to make it sure it sounds like it's a My voice or something, or, or our brand's voice. How does context work and, and how do you know what to provide to the chat?
[00:15:30] Chris Duke: Yeah, prompting is like the most critical part of working with an AI tool.
Like it doesn't work unless you prompt it correctly. I think everyone here can probably resonate with putting something into one of these tools and then just being like, that's horrible. Like, that's just wasted two minutes of my life, I'll never get back.
And so learning how to prompt really well. Giving an AI tool context is really, really helpful. So there's five things when I think about a prompting that is really important. So one is who is the AI acting as? So again, earlier you saw, I wrote, you know, you're a member of the strategic events team at Glean. So that's the important to give the AI tool a role so that again, that has the lens and the framework of what it's going to do. When it comes out with its response.
Yeah, you wanna give it a task. So that would be like me saying, um, I need an event brief with X, Y, and Z. How should deliver the result? You know, one of the things I failed to do in that one was I forgot to say that I want it in a Google Doc format. So I actually had to go back and say, okay, put this into a Google Doc format.
So if I would've just done that from the beginning, I would've had a response from the beginning. Yeah. So that would've saved me another minute. So that's on me. I'm still learning myself. The tone and the audience. Another thing that you can always add, this is. You know, context and tone and audience is a little bit, it's not necessary. It's not required.
But what you can, the thing that you can do is you can say like, who is this for? Because Then that might change the language. I may say this brief is actually for my CMO versus maybe the product marketing team. So if the product marketing team. Is the audience for it. It may be more technical in nature, whereas if it's for the CMO, it may be more outcomes driven in nature. So that's really important as well to give it, um, to know who the audience that it's gonna write the answer for.
And then what's the vibe too is just like, I, like I'm a vibes guy, so I always like to say, "Hey, make this brief or, you know, make this really punchy." I think that's also important because, um, when you talk about personalization and context personalization for me, I'm not very, it may not seem like it right now, but I'm actually not very long-winded. I like to keep things pretty short. I like 15 minute meetings instead of 30 minute meetings.
And so for me, I like to make everything really tight. And then for context, again, this is optional. But you know, it could be your brand guide where, you know, you have certain attributes or tones or colors that you wanna include in things or product information. 'cause again, not everyone has an enterprise AI tool.
So it's really important that you're uploading non-confidential information to ChatGPT think that's critical to talk about too is, definitely don't wanna put any critical or confidential information into these public AI tools. It's just we don't know what the world looks like right now. And you gotta be careful with stuff like that. So.
[00:17:52] Rachel Moore: That's a great, that's, that's a great tip too. Yeah. I, it really does come down to like, and for anyone who's watching this too today and be like, okay, I've been thinking about making sure my team is set up with like an enterprise tool internally. yeah.
That's probably, that's a good consideration have be like, well, how, how free do we wanna be with like, some of the more proprietary or confidential stuff versus Oh yeah, just go use your free, free go set up a free ChatGPT accounts. Well, don't go put company data into that then, right?
So you, you definitely wanna keep that separation.
[00:18:20] Chris Duke: And there's a way you can balance that too, right?
There's a difference between like non-confidential information that would help you do your job better versus like, you know, maybe you don't wanna upload your product, product roadmap for the next two years into it, or, I don't know anything that's. You don't, you definitely don't wanna do that.
But it's also a call to action for everyone here. Like if you don't have an access to an enterprise AI tool and you work at a company that's over 50 employees, you should be talking to your CTO, CEO, CMO saying we need to take advantage of these tools. So there's a lot out there and you should be an advocate for that.
[00:18:52] Rachel Moore: Good, good call. Yeah. Well, and I, I, I know we've got a lot to show today, so we're gonna move on to what weve already teased out.
We mentioned the word agent. Okay. I know personally when I hear the word agent, maybe we've got specific characters or avatars that come to mind. I think of Agent Smith from the Matrix, which probably is not the theme we're aiming for here. He's a bad dude.
But agent. As in an AI agent is what we're aiming for and what we can use to catapult us into untold automated processes that have a lifting of work that can happen in the background while we get to focus elsewhere.
So, Chris, I know you're gonna talk to us about agents and I think a whole reason we wanted you to do this work workshop with us is, uh, you just presented at CMA and, uh, our CMO was blown away, uh, and the entire audience was as well with workflow examples of AI agents acting as virtual teammates.
We'd love to get some wow factor from you now if we can, with one or two of those examples. So we can see how does this AI agent work for us?
[00:19:49] Chris Duke: Sure, no problem. I'll do my best to. Wow. Um, so on the.
So agents are basically, they help you with like repeatable tasks and we talked that was in the poll, which is really important. Agents work autonomously. They also work on demand. Basically what agents do is you can set up a workflow.
So instead of using like a prompt and gonna chatGPT and typing a one-off thing into some type of an assistant tool, agents are something that helps create repeatable things over and over again. And so you build a workflow once and then you can use it over and over again.
So I'll give a couple of examples. One on like a very practical example of a session builder and speaker, speaker, bio agent. So, I need to write, uh, session names, right? Usually you have to go to product marketing and say, "Hey, we've had a session at Reinvent coming up in December. I need a name for it." So this is actually gonna help us do that.
And it's also gonna give us a speaker bio. 'cause sometimes when you need a speaker bio, I don't know, you guys ask me for one, it took me like a week to get back to you. So maybe use this and it would've saved some time. So, basically what an agent is, um, it has that input, uh, input fields that you need to put in.
So I'm gonna give an example of, uh, we're going to AWS Reinvent with their weird naming, the audience persona, a lot of developers at Reinvent. And there's, C-level and like, maybe like director plus, um, tech leader technology leaders.
The speaker name from, at my company, I'm just gonna use my CEO. So Arvin, Jane, um, because again, some of these things take time to run.
I have it prebuilt already, but this is where I would put in vin's, LinkedIn, any additional context as well.
I built this one. This is something that I built, so I can actually show you the backend real quick before we run it just to give you a little bit more information how it works.
Again, I have access to a Glean agent builder. There's Google Agent Space, there's ServiceNow. You guys all may have different tools. So this is the one that I have access to, but they work relatively similarly. So the key things that we need an agent to do is we need a trigger, which is the input fields to put all that information. So this is where I put the event name that I need, the persona, things like that.
And then I need it to think. The think step is really important because this doesn't work unless it's thinking. So, we need to analyze the event theme, the audience persona, take any consideration, additional context for the specific event, speaker name and LinkedIn profile, determine key topics, audience interest, speaker expertise, your compliment style.
So, basically it uses, web data, real life information as well as your company data to create this agent output, which you'll see in a second. And then the respond feature, by the way, there's like 25 more features you can put in here, this is a very simple one. But on feature is where you basically put the information in on how it's going to output the information that you just put into it from the trigger. So I want a very specific use case. I want five session titles. I need them under 60 characters each.
Um, the, for the speaker bio, I want it concise, but I want it to highlight certain parts of the background, right? What's cool about this is that if I was the speaker, the topics and the session names would be different than Arvin. Our CEO is very technical in nature, executive level person, he's gonna have different outputs as the speaker.
Um, and then response style, you know me factual, straight to the point, but you could do balance, creative, um, and then you also, there's some technical information here if you need to get into that.
So, the output of that would be something like this. So we'd have, you know, securing AI agents Next Gen Cloud collaboration from Search to Insight, Transforming IT with AI scaling innovation. Again, all these are like leadership heavy because Arvin is doing it. Our CEO. Yeah, and here's a speaker bio right here.
So I would probably like, you know, run this by the comms team or whatever, but at least it took a first pass. So that's an example of running a really simple, easy, um, our. Uh, these agents, I mean, there's hundreds of them.
So again, I, it's really important that we, uh, start to build these things to repeat some of the tasks that we have. But I can give you another example real quick of something that's just like very simple and, and mindless that I do every week as well, if that works.
[00:23:46] Rachel Moore: Yeah, do it.
[00:23:47] Chris Duke: Every Thursday my boss, Katie. Sends me a Google Doc and tags me in it and it's like, give your update. So I have to give an update to the team every week on what I worked on. So like my accomplishments, my blockers, and the key things that I have for next week.
And that would take me. An hour, probably. Probably, yeah. Yes. This very mindless, simple ask takes an hour of my day on a Thursday that I need to share out with the rest of the organization about what are the things that I worked on.
So again, I would have to go through my Gmail, my slacks. I had to look at my old meeting, what happened on Monday. I don't even remember what happened on Monday. You know, what am I working on right now? And so one of these things that I set up again, if I look at my agent set up here, there's a trigger. It's gonna read my personal activity and it's gonna respond and it's going to show what I accomplished this week, what blockers I have going on, and my upcoming task.
Um, so again, I can now copy and paste this. I can edit it, put it in Google Doc and I'm done. So i'm gonna take this off screen because that's actually what I accomplished this week, which wasn't much. Um, so because it's Tuesday.
Um, but again, uh, that's an example of an easy copy and paste those repeatable tasks that you're, you know, we talked about in the poll earlier that we can solve for and we can move on some, um, more interesting and creative things throughout.
[00:25:02] Rachel Moore: Everybody would love to get an hour back from, these are necessary things. We know we, we are required to do through the course of our work and our job, but it's still, it's like, yeah, but if it's all, it all lives in the thing tools we're using, if an assistant can go in, if an agent can go in and say, yeah, let me go round that up for you. Well, you go do some other stuff and now it's done. That, that just seems like a no brainer. I wanna ask a quick question too, 'cause I know. I thank you too, so much too. I know you're emphasizing too, like this is based, a lot of our ability here can be based on what tools we currently have access to in our own environment.
I just recently heard about Google Opal, I work in Google workspace in like, Gemini is, is available to me, but apparently, so is Google Opal. And this very much looked like it was just like a traditional workflow being set up. That's kind of how the agent works, but yeah, yeah, could you, um, speak to that maybe a little bit?
[00:25:50] Chris Duke: Yeah, definitely. I mean, I think, um, again, this is where we all need to be advocates internally of the tools or that we have or don't have. All of these agent builders and tools, some of them are really good and some are fine. I think it depends on, you know, who you ask or what you have or what the tool is itself.
But this stuff is evolving so rapidly that the differentiator between these is gonna start to get like smaller and smaller. Mm-hmm. And when one company does it really well, the other companies will follow suit.
And so again, whether it's Google Agent Space or Opal, Gemini, ChatGPT, Enterprise ChatGpt. Just normal one Claude. Um, all of these things are start to evolve and, and, um, there's just, I think certain companies have focuses right now, like I can speak for Glean agents are a huge focus of us because we're trying to, you know, improve the life of the human worker. That's our goal.
And I think a lot of these other companies that have Gemini, which is a consumer facing tool or ChatGPT, or Claude are also seeing that as a, an opportunity for them.
And so you'll start to see if you already have access to those tools, that you'll start to have an agent builder like, in your suite pretty soon. Nice. Because all, you know, these companies are building them essentially.
[00:26:57] Rachel Moore: Does this require a lot of coding ability? Like how much of a coder are you? I mean, you know, in your team?
[00:27:03] Chris Duke: No, I built like MySpace pages when I was in high school, like with html. That was about it. That's about all I've ever done. This is all, um. It's the beauty of AI tools right now. And you'll notice there's terms called vibe coding and all this stuff.
Uh, yeah. But essentially, uh, you can write in na like just natural English language. Um, I, this is not something that I had to code. This is something I typed. There's even an enhanced feature here that makes it sound better than I even wrote it. So nice. We don't need to worry about code, we don't need to worry about Java or Python or anything.
You just write and normal human speak language. And that's the beauty of large language models, is they're trained on human English and other tools. Languages across the world. Yes. Um, and so yeah, it's, it's all just normal talk. You don't have to worry about code at all.
[00:27:45] Rachel Moore: You're getting so many reactions on the screen right now.
I think people are like, oh, thank God
[00:27:50] Chris Duke: guys. All building MySpace pages. Yeah, same.
[00:27:52] Rachel Moore: That's right. That's right. Well, okay, so speaking of, I love that you brought MySpace because a lot of that came down to how, how our MySpace page looked.
So We will be right back with more Event Experience after the break,
Ever wished your event badge did more than just display your name? Whether you're an event organizer, exhibitor, or attendee, Bizzabo's Klik Smart Badge transforms the event experience. From interactive networking to real time attendee insights and lead capture, the Smart Badge is the ultimate game changer. Visit bizzabo.com/click, that's B I Z Z A B O dot com forward slash K L I K to learn more about the next generation of wearable event technology. Bizzabo, your partner for building conferences that captivate.
We're back with Chris Duke, who is about to show us how AI can create visualizations for what's cooking in your brain.
[00:28:50] Rachel Moore: So y'all we're gonna get into, uh, visualization now. So, um. Yeah, I'm sure we've all, any, if we've even scrolled literally any social feed out there, AI generated images and videos are there. They're in our feeds, and now it's almost like, Ooh, can I detect it or not? Right? Um.
We can probably all remember the last one we saw. Uh, if it's safe work one, feel free to describe it in the chat. Um, but, uh, visualization tools are crucial for event planners. Everyone on this call, on this webinar, uh, we need to plan and present what we have in mind.
It's going to manifest in a three dimensional, you know, display for somebody to walk past, walk in, and et cetera.
Um. Chris, I'm gonna ask you this question 'cause just like we just talked about everyone's ability with coding, uh, I'd like to ask, do you consider yourself an artist?
[00:29:33] Chris Duke: Um, I think I'd like to consider myself an artist, but I mean, as you'll see in a second here, like, I'm terrible.
I got like chicken scratch and I can't draw. And, um, I, a big thing about being an event marketer as you bring. You know, you do bring creative ideas to life and that's been a challenge, I think, forever, since I've been working. Yeah. At least I have these ideas in my head, but getting them down on paper is one thing, and then into like a CAD file and then into like, if you work in trade shows, like there's no better feeling when you're working in a trade show booth, when like you're about to be show ready and like, everything's like squeaky Glean and the booth looks good and the lights look good and everything.
And so it's, uh, it's cool to see your ideas come to life. And that's the beauty of working at events. We're not here to just like, you know, crank out staffing sheets and stuff like that. There's the, the cool part about the job too is bringing, you know, it's a physical representation of the brand that you work for and, uh, whether that's a conference, a trade show, whatever that is, you know, being show ready is a, is a good feeling.
And so, um, yeah, I think AI tools have given us the ability to like get things out of our head now and onto. And basically on paper and into email and into approvals. And we're moving way faster now, um, as event marketers than we ever have. And a lot of that has to do with working with, um, these AI tools, which we'll share in a second.
So, well,
[00:30:48] Rachel Moore: Let's, yeah. The second is here. Let's, let's see it, can you demonstrate how it can help with. Take what's outta your brain and make it into a visual concept.
[00:30:56] Chris Duke: Yeah. So you asked me if I was an artist. I'm gonna share, uh, behind the curtain here a little bit. Thank you. Um, a booth design that we were talking about internally.
We needed a new 10 by 20 booth. We built this great 10 by, uh, 20 by 20 for a trade show, but we needed a smaller one. We wanted it to look similar, and it was just trying to like, get ideas outta my head and kind of, you know, through meetings and stuff.
So here's my example of what I drew. Okay. So not, um, you get the point. This is something that I would normally like, draw, and then like send to my agency. To help me turn into like an actual like CAD file or a three dimensional thing to like get, again, my ideas. I did get it outta my head and onto this piece of paper, but that's not gonna cut it. Like, I would never send this to my CMO to like approve.
So, um, with good prompting, with some revisions and back and forth, I can turn this drawing into a booth that I actually had in my brain with the right colors, the right, um, you know, materials. So I got some wood grain in there. I've got the logo correct, you know, I see this and I look at it now.
That's too many logos. So again, there's things, but like I would rather send this off than I would the, the chicken scratch one. It's embarrassing. Two, there's like a lot of room for, um. You know, interpretation and this. Now I can give this without being, you know, I wanna give creatives the ability to be creative, but I also know in the interest of time that if we can cut out some of the, you know, the weeks of back and forth, this would probably have taken, this turnaround, probably would've taken five business days to get something back from our, our agency.
And now I'm turning it around in V1. They can take this, put it in the actual CAD file, you know, make sure the specs are right and everything. Yeah. But we've got a, a good starting point.
So again, that's a, a good example of getting ideas outta your head. And I just did ChatGPT. This is just ChatGPT, I just uploaded this image. I explained, I gave it a good prompt. I explained down to like, Hey, I want the cabinets to be wood grain. I want neon light bars around the edges. I want hard angles. I need a check-in desk, a small area for meetings, and then demos for, um.
[00:33:00] Rachel Moore: Thank you so much for bringing that up, that this was actually a ChatGPT, which pretty much everybody has access to.
Are there other tools out there, uh, that are, that you would call like the best AI visualization tools that for say, an event planner trying to do what you just demonstrated?
[00:33:13] Chris Duke: For sure. ChatGPT is now in five. I had four on here and now it's in five. So these things are evolving rapidly. Um, yeah, Sora, now you've probably seen a lot of these like AI videos coming out, which are really just like stupid stuff with like your face on it. So it's not really like, perspective.
But the OpenAI suite is very, very impressive. Um, it works really fast. It's just really, really impressive. So I would definitely highly recommend the OpenAI suite for that across the Natively and ChatGPT but also using Sora.
ChatGPT also had a tool called DALL·E, which is now just integrated into their platform, so you don't have to treat it as a standalone thing. But for, naming narratives, as you know, images, videos, that's all taken care of there.
Runway and Midjourney are also some interesting tools that you can mess around with. They always give you free credits to go and try things, so I definitely would recommend just going and messing around.
But Runway, you could do an image and video generation, Midjourney, you can take a static image and bring it to life. So one thing I could do with that booth design is I could put it into Midjourney and I could say like, add the experience of a trade show around it, and it would kind of like insert people like moving through it.
So I can think about the flow of the booth or maybe where they're coming from. Are they gonna see a logo from the left or right side? Like you can get a little creative on how you think about are you making the right decisions? And Midjourney is really helpful with bringing static images to life in, in short videos.
And then Gamma. There's a lot of slide generation tools that are out right now. There will be more I'm sure in the next few months as well, but I personally really enjoyed using Gamma.
Gamma is a tool that you can, and I'll actually show you here in a bit. But Gamma's basically take, um, it can take your natural language, your prompt, or your output from a prompt. From an AI tool, and then turn it into a, a storytelling deck, a PowerPoint slide, a Google deck, whatever that is that you're using, uh, internally for your slides.
Those are the four that I would definitely recommend for all of med marketers to be using right now, but mainly the top two.
[00:35:02] Rachel Moore: That's great. No, this is so helpful. Thank you. I was even thinking too, like, as, you know, like, okay, I know our booth space and I know what's gonna be around like, maybe, uh, different vendors or even like, "Hey, maybe the coffee or the, the snack, um, booth or kiosk is gonna be near us."
And you can actually like, okay, great. Let's visualize how we can capitalize on that flow of people to make sure that the booth is open to that and receptive. So, it's just a natural. So it just seems like such a great way to not just have to all, you know, from your own brain, say, okay, what am I not thinking of?
But to really use these to say, no, let me see it and see it in action. And ooh, you know, make some adjustments based on, oh this, this was great. It highlighted this opportunity for us.
[00:35:43] Chris Duke: Another example that I thought I would share with everyone was, um, we got a new office space in San Francisco.
I got asked to help build a stage for it, uh, internally, so for all hand spaces, live productions, things like that. And I wanted to just kind of help I wanted to visualize it. I kind of wanted to get some approval for it. I don't have budget for an agency to help me with this, besides the fabrication of it, but all the creative stuff, it just, we strapped internally, so we needed to do it ourselves.
And so, uh, I took some photos of the office space. I gave it some example of Anthropic event that I really liked. They recently did a launch event and I just kinda liked the setup. I thought it was good and simple.
And so I gave it this prompt. And again, you'll see things like, um. Who it's acting as, you're a 3D designer. So it's thinking about the space as a three dimensional designer. I am giving it some context about this information of what this is. Again, this is all in ChatGPT, so this doesn't have access to my company data, so I have to provide as much information as possible to get a good response.
If you don't do this, you're just gonna get junk, and then you're just gonna spend the next 45 minutes trying to figure it out. So it's better. Just do it right from the beginning. Give it a really good prompt. I talked about, um, the layout should feel Glean, modern, integrated, and existing office design. So, we should be able to see desks in it. The ceiling pipes, there's muted colors. It's very simple. It's a brand new office. There's no decor or anything like that. And then again, use the attached image from the anthropic event as a visual reference. So that's a really good prompt and.
What I was able to do is help me generate this right away immediately.
It's not perfect, but, um, it's a really good starting point for me to think about, okay, do I wanna keep this pretty simple? Do I wanna add more color to the stage? Do I want to, uh, keep it like this? Do we just wanna focus on content and have a large screen there? Do I want a small stage or, you know, and how is it, how does the wood tone look in the more industrial office space that we have?
And what's cool about this was I sent this to our CMO. So I did this in 10 minutes. I said, I was thinking something like this, you know, some decor, maybe a little wider, but you get the point.
He says, "Nice, I like it. Can we get it built by our offsite, which is in like three weeks." Um, and that getting from this prompt with these photos to this, and he responded to me in like about two hours. Like that used to be a thing where I'd need to send it off to somebody. Yep. I'd get a rendering back and then like I look at my CMOs calendar on like Thursday, we'd sit down and talk about it. He'd be like, this is terrible. Can we redo it again?
So now we're into week two. Like we just went from like, we just saved like a week and a half.
And again, we had to talk about it. This is not what it ended up looking like. It ended up looking way better than this, but it did share like, Hey, we can do this. Like we can build this space and it's gonna look fine.
So that's another example I just wanted to give. Yeah. Um, of some creative tools that we're able to get to from, you know, not only just get like consensus, but even we can even start to get approvals. Yeah. Without meetings, without waiting a week, without waiting two weeks to, you know, revisions back and forth.
It's just, uh. Yeah, it's a really, really good asset that we have access to now.
[00:38:37] Rachel Moore: I love that use case and so many we can all relate, right? Where it's like, oh my God, just, just get this, get us to yes, faster. And it helped you do that. And thank you again too. That, that prompt is super key. I really appreciate that.
I know we teed up the beginning of this whole demonstration and everybody said, I want to get the repeatable tasks off my plate. That was like the, the biggest poll answer, right? And so, so far we've covered up stuff. We've covered stuff that represents the actual event related work but let's come full circle.
We know there are those parts of our jobs that are complete snooze fests that nobody really wants to do. Data analysis, budget calculations, preparing weekly progress reports like you've demonstrated that, that maybe someone will glance at and maybe they won't. But this hopefully will be the most exciting part of today's session because we all might get back that thing that we never have enough of, which is time.
Sure. So Chris, what's a more boring task or se series of tasks that you are more than thrilled to hand off to AI to do for you?
Please share it with us because we all want to do the same thing.
[00:39:35] Chris Duke: Yeah, totally. When the event ends, like event marketers know events don't end.
They, it's just kind of the beginning sometimes what it feels like with all the, like, you know, what's ROI? What were the results, the feedback, you know, you're dead tired after working a week and then you gotta like turn around a presentation deck in 48 hours. 'cause people want results and things.
So some of the stuff that I like to automate and in a world in which events are always kind of moving. Each project can be different. It's hard to find those repeatable tasks. The, like backend of events are something that can be very repeatable. Yeah.
So I'm gonna give you an example of using Claude. This is maybe something that I would actually build an agent for in the future, where it can be repeatable. So I'm gonna build that, you know, one time, and then I can use it over and over again.
But for the sake of this conversation, I'm gonna use Claude Claude's really good for, um, it's one of the better ones for writing and for working on like data sets, I think personally compared to ChatGPT or anything. So I like to use Claude for stuff like this.
One of the things that I am gonna just show you is how to get from an Excel sheet, uh, or a CSV file of some, uh, your survey data. So you sent a survey out, you got some, uh, quantitative and qualitative feedback from your attendees. And basically what I'm gonna do is I uploaded a, a fake CSV here of, um, fake attendee data that I had.
I used ChatGPT to give me a fake CSV of fake attendee data from a fake event.
So this is, we're in like, you know, we're in layers.
[00:40:52] Rachel Moore: It's very meta.
[00:40:53] Chris Duke: Yeah. It's for sure.
So, um, I'm gonna give it a prompt. I'm gonna say you're a Customer Insights Analyst. I have no idea what that is, if that's even a job, but it sounded good to me.
So we're gonna read this feedback, we're gonna cluster the comments into clear themes. We will flag anything urgent or negative. I think that's important. We all love to hear how great our events were. We want all the praise, but we also need to know what did not work. That says like equally, if not more important, so that we could fix for the future.
Pull any strong quotes, we could use the testimonials. That's good for slide decks and for ROI decks is, you know, some qualitative feedback from some, you know, director who said this was the best thing, you know. This session was great. So we wanna make sure we have some of those.
And then summarize what C-level attendees versus ICS cared about the most. That's another thing that's important. Mm-hmm. So depending on your event, you know, I'm thinking more of this as kinda like a user conference use case maybe where you have a lot of different job titles there.
So essentially I would run this, and again, this probably takes a minute or two to do, but for the sake of the demo here, I'm just gonna show you basically what came back from it.
So what it took from this theme in feedback here. And one thing that I would like to share, um, really with everyone is not necessarily what came in here, and like actual, like the hard data and everything, but just some of the qualitative things that came back.
So again, this has no context. For anything. Right? I just uploaded a CSV file and I said, give me information essentially. Yeah. And so it doesn't know that like the main event organization was the standout winner.
Like, if your CMO doesn't care that your event was organized, like that's the expectation you have. So again, like this is really helpful and this is a good starting point, but, uh, it's one of the most organized events I've ever attended a grant. Like that's just for me as like a pat on the back, right? It's my fake event.
But it's not. That's not helpful or anything I'm ever gonna put into feedback. So there may be some back and forth with the AI tool to kind of focus on different things. Um, I also didn't go through all like 50 line items in this fake CSV file, so I probably should have done that too.
But, um, there were some urgent issues that were flagged though, which was important, which is session, uh, session timing problems for people specifically mentioned sessions running long and overlapping.
[00:42:53] Rachel Moore: Hmm.
[00:42:54] Chris Duke: Um, there were some insights, um, 9 out of the 50 that were C-level focused on strategic value of the event speaker relevance and overall experience quality.
Um, they care about big picture and business impact. The ICS gave more granular feedback about networking opportunities, operational details, etc. Uh, the good news, both groups, uh, rated the event highly 4.3 out of 5.
So now I have a, a summary of all that data instead of going line by line and typing it out.
So I gave it another prompt. I went back to it and I said things like, you know. You are still a customer insight, uh, insights analyst that reports to the CMO. So now I'm giving it the mindset that you actually need to share this information with our Chief Marketing Officer.
[00:43:34] Rachel Moore: Yeah.
[00:43:35] Chris Duke: I wanna get this slide ready 'cause this is important information.
I think if this wasn't a demo, I would probably go back and forth a few times and really get some of the hard numbers and hard data out here so that we could get it a little better. But essentially what it did was that Claude can't generate slides. It's just not something that I can do right now, but it can put everything into slide format.
So I was able to take, here's an executive summary, like you'll notice this is how a slide would look. This is how a slide would look. There's critical issues that we need to fix. That'd be a slide, that one we skip over real fast in our meeting. And then testimonials are also important.
So again, it's all broken out by slides. So what I would do is I would head over to a tool called Gamma, and I would basically highlight all of this. I'd copy it. I would paste it into here and, um, we'll see actually how this goes. 'cause I didn't do this ahead of time, but, um, Gamma is the slide generation tool I talked about earlier in the creative suite.
And this is gonna take all of this raw data and put it into slide format. So I'm gonna generate it. Did a, it says there's about 10 slides in here. Okay. So I'm gonna click generate.
[00:44:38] Rachel Moore: But I think it's key too. I, I love that you mentioned you're, you're kind of daisy chaining over here and look, it's already kind of doing it, but that's amazing. Yeah.
[00:44:45] Chris Duke: Yeah. So I'm building slides right now, which is great. Wow. Yeah. And this will go on for a while. What I think that I love about this stuff is it also helps with just formatting, like, how do I wanna represent like this slide?
And again, you'll notice things like. I, I haven't built this out properly, but like, this is not a slide size.
[00:45:04] Rachel Moore: Sure, sure. But you can go back and tell it. Right?
[00:45:07] Chris Duke: Exactly. You can do that. Or what you can do is once we get through here, you know, we've got the nice, like the way that this is built out, you've got this, you know, the quotes and everything here, got feedback.
Little things like this is just a nice looking slide. Right. You know what I mean? Like the information in there, I definitely wanna go in and edit. But what you can do here too is you can click, um, Share, Export, Export to Google Slides, I'd approve it.
It's exporting to Google Slides. And now I can just move this whole thing over. I gotta change the colors. Um, if you have a Gamma Pro tool, um, an enterprise tool, you can actually upload your corporate brand template into it.
So it can just
[00:45:47] Rachel Moore: Oh,
I'm sweating.
[00:45:49] Chris Duke: Yeah. Um,
[00:45:51] Rachel Moore: that's amazing. So
[00:45:52] Chris Duke: Now I, now I've got. See how it auto formatted that slide that was a little short into a slide deck.
So now I've got a slide deck here. Geez.
[00:46:00] Rachel Moore: Wow.
[00:46:01] Chris Duke: Um, you know, 10 slides, if you have a pro tool, it can do more than 10 slides. If you don't, you can only 10. Uh, but again, this just an example of going from this with this. This took 30 seconds to turn this into. Yeah. Again, I'd probably spend five more minutes editing it into this, into this, and then into here.
Wow.
[00:46:21] Rachel Moore: Um,
[00:46:22] Chris Duke: and again, I would go back and, you know, purple's not in our color palette, um, despite the color behind me and, um. Yeah, there's some changes that need to be made. But look, I mean we're in a slide deck here in five minutes from a survey, so
[00:46:33] Rachel Moore: This is what we're talking about too, and you've done such a great job with like really bringing this home for us too, because it's like the things that could take you at least an hour or two to do, you can actually consolidate and just.
Again, using these tools and maybe just not one tool, but realizing what all works. And I, I know we are, we're gonna get to questions in a bit because it's, we're gonna Yeah. Make sure people know what tools are out there and stuff, but this has been fabulous.
Thanks again to Chris Duke for joining us on event experience, and thank you for listening. If you're enjoying the show, we'd love to hear it.
Connect with us on social and Subscribe, Rate and Review us wherever you're listening. Also, don't forget to share the show with your colleagues and friends. You can find transcripts of each episode and key takeaways on Bizzabo.com/podcast. You can also access this webinar on demand by visiting Bizzabo.com.
On behalf of the team, thank you. We'll gather again soon for a new episode of Event Experience.
