In this week's episode, Paul Roetzer and Mike Kaput sit down to talk about AI and how we use it at the Marketing AI Institute.
With our annual Marketing AI Conference (MAICON) coming up in August, we're currently marketing the event and pushing registrations. But with so many folks in our database, how can we laser focus our efforts?
We're using AI and technology to help us predict high-priority contacts based on their probability to purchase (more intelligent)...so we can reach out to them with customized, personalized communications (more human).
We get transparent about our internal processes and hope you can learn something from this as well.
[00:07:00] MAICON 2019 is launched
[00:13:27] Living a more fulfilled life and have more fulfilled career
[00:15:34] How do you stop blasting your list with your multiple promotions?
[00:16:49] Who are the 100 people we need to be having phone calls with?
[00:19:46] Creating a movement vs. selling event tickets
[00:21:47] This isn’t a decades in the future thing, this is a now thing
[00:22:04] Building a lead score in 2021
[00:23:44] What intent and loyalty signals should we be looking for?
[00:31:01] Scoring for individual outreach and personalized learning journeys
[00:35:07] Selling 300 MAICON 2022 tickets in a world with travel restrictions, health and safety concerns, and an uncertain economy
[00:39:27] Shameless Cleveland plug
[00:41:57] The greatest indicator of probability to purchase is intent followed by loyalty, engagement, personal connections, and strangers.
[00:45:01] Everything we do as marketers should get smarter with the help of machines that should continually learn and augment what we're building
[00:55:02] Personalization is not “insert name, insert company.”
[00:59:37] Using LinkedIn Sales Navigator as part of this process
[01:02:46] How can our data help us reimagine our expo hall?
[01:05:25] What value-based can we give our customers based on their data?
The Land block letters image source: Nathan Migal via Destination Cleveland
Disclaimer: This transcription was written by AI, thanks to Descript and has not been edited for content.
[00:00:00] Paul Roetzer: Welcome to The Marketing AI Show. The podcast that helps your business grow smarter by making artificial intelligence approachable and actionable. You'll hear from top authors, entrepreneurs, researchers, and executives, as they share case studies, strategies, and technologies that have the power to transform your business and your career.
[00:00:20] My name is Paul Roetzer. I'm the founder of Marketing AI Institute. And I'm your host.
[00:00:27] Thanks for joining us for episode 15 of The Marketing AI Show. Before we get started. I want to take a moment to tell you about our show sponsor MarketMuse. MarketMuse Suite, an AI powered content intelligence and strategy platform analyzes millions of articles on demand, uncovering gaps and opportunities for better content.
[00:00:47] Imagine an on-demand content audit that automatically identifies your best and worst pages. Content with high ROI, potential quick wins at risk pages and more. MarketMuse uses AI to accelerate content planning, creation, and optimization. So you can build authority on your topics. Get started for free today at marketmuse.com.
[00:01:13] Now onto the show.
[00:01:16] Alright. We are back for is our second weekly Mike, our new show format, where Mike and I sit around and talk about trends, big ideas, topics, timely, relevant stuff that we find interesting. And hopefully you find interesting, um, And this one is really interesting to me.
[00:01:37] Mike Kaput: Yes. I'm excited.
[00:01:39] Paul Roetzer: Yeah. So basically the way, so ...Mike... If you've again, new listeners. So Mike and I work at Marketing AI Institute together, I'm the, uh, founder and CEO. Mike is the chief content officer. Uh, we are also co-authors of the new book, Marketing Artificial Intelligence, which if you're watching on YouTube, we have this lovely new oversized cover, a framed cover of the book that just arrived today from Ben Bella.
[00:02:03] So shout out to Matt Holt and the Ben Bella team. Really cool backdrop for today's podcast. Um, book comes out June 28th. We'll talk a little bit more about that in a minute, but so today's topic based. What am I going to do is we keep a running list of like interesting things to talk about. And then we have a weekly meeting on Fridays.
[00:02:21] I think it is like, is that right? Let we do Friday mornings. This is new. I'm more in the second week. So I'm talking about this as though we've been doing it for a year. So every Friday we get together and go through some potential topics and either pick a few or just pick one thing. And there was some really interesting, timely stuff we were working on last week for our Marketing AI Conference. And so just like, I think Mike throws, like we should do something related to events or to the conference. And then it just sort of came to like, let's just share what we're doing. We're trying to solve for selling tickets and sponsorships to a conference in real time, trying to think about how do we do this smarter?
[00:03:00] Like how, how do we use AI ourselves? How do we, uh, get into better personalization, better prediction? The things that we teach people, AI does events are like the perfect place for this stuff. And we do events. We have friends who run events. Like, I don't know if the innovation happening in the event, marketing space, um, where AI is really being talked about as a driver.
[00:03:25] And I know Mike, like you get, and it's like, as soon as it came up, you and I were kind of on the same page with it. I don't know if you're seeing anything. I'm not seeing.
[00:03:32] No, that was that kind of a key takeaway from what we're about to talk about is that, um, you know, we test a ton of different AI tools and technologies, uh, across a number of different marketing use cases.
[00:03:45] And a lot of them power parts of the business in content marketing in advertising, um, in general marketing campaigns, email marketing and event marketing. When it came time for us to sit down and really, uh, get smarter about our event, uh, marketing strategy. There's not from what I can see much innovation in AI at all.
[00:04:11] So far in the event, marketing space like Paul, you had mentioned, you know, our virtual event last year, we found exactly one platform that did anything with AI and it was cool, but like, it probably doesn't even scratch the surface of what's possible. So what I'm really excited about for this conversation is we're going to.
[00:04:31] Theorize some ways that if the technology existed, we could start using AI tomorrow to do event sales. But yeah, I've seen, it seems like a really, uh, not mature space when it comes to solutions
[00:04:44] that
[00:04:44] Paul Roetzer: use AI. And I mean, so I'm relatively new to the event business. I ran an agency for 16 years and sold the agency and the Institute was growing in parallel to the agency.
[00:04:56] And then once I sold the agency, the Institute became the core focus and the Institute is a media company and we publish a lot of content, tons of content, really valuable content for different audiences to educate them about AI and make it understandable and approachable. We run a conference, an annual conference, which we're obviously going to focus on today.
[00:05:12] Then we have an online education platform, um, AI academy for marketers with on-demand courses, but you know, to over-simplify the event world, if you are not in the event world, if you have a extremely. A loyal, engaged audience, 10 20, 30, 40, 50,000 readers, subscribers, whatever it may be. So you may be a media company that launches events.
[00:05:34] You may be a tech company that has tens of thousands of loyal customers. You launch an event like you're, you're basically just convincing people who already love you and know you to come and be together to be part of a community together. That's not an easy sell by any means. Don't get me wrong, but to start a community from nothing to, to start a movement from nothing, which is what we tried to do.
[00:05:58] So. In retrospect, it was crazier than I thought it was at the time. Like I, at the time knew it was kind of crazy. Now I look back to 2000 16, 17, 18. I'm like what in the world were we thinking that we could, we could pull this off. So basic premise will kind of hit the rewind button real quick. Start the blog marketing Institute blog 2016 ramp up got five to 7,000 subscribers by late 2018.
[00:06:24] So my original plan was once we got to 10,000 subscribers and I'm talking like people who've subscribed to the newsletter, attended webinars, like given us con opt-in contacts, I'll generate generically call subscribers for our purposes today. Once we got to 10,000, I felt like that that's a reasonable number.
[00:06:42] You get 3% of those people to come to an event like you got a decent starting point, but we, we decided to go earlier. Uh, we were about 7,000, but it was a really interesting group. It was lots of major enterprises, VCs, influential investors, um, some high profile people. And it was like, I think we're onto something here.
[00:07:00] Let's go. Let's give it a go. Like what the heck? So I put a bunch of my own money in to like launch this conference. So 2019 rolls around, we split the Institute office, its own company and the conference conferences, its core thing. So Mike's plugging away right? In a bunch of content, organic traffic's going from 5,000 to 10,000 to 15,000 users a month.
[00:07:19] Things are going great. And then the goal for the first conference was 300 attendees. So we, we do that. We run it in Cleveland. It was amazing at the convention center in Cleveland. If you've never been there, overlooks the rock and roll hall of fame and the, the, the brown stadium and lake Erie, it's a beautiful venue.
[00:07:36] Um, and from an operational perspective, makes a ton of sense because the costs obviously are nothing close to what it would do. New York, LA, Chicago, Austin. You know, if you, if you're focusing on the AI hubs, like there was just no way we could afford to put an event in those major markets right away. So 2019.
[00:07:54] Goes. And it's like just through the roof. Amazing, incredible speakers. Like Douglas Rushkoff did a closing keynote from a team. Human was amazing. The Karen Hao from MIT tech review, Mitch Joel, David Meerman Scott, like just phenomenal. The experience was incredible because the team that ran at the ops team, uh, they ran content marketing world for years.
[00:08:14] So Kelly Whetsell and her team, um, came in as like outside contractors ran the event, just like world-class. So just like you're coming off of this high, like that was amazing. It didn't hit the revenue goals we wanted. I lost quite a bit of my own money, honestly, in the first year. Um, but that was just part of the deal.
[00:08:31] Like you knew it was kinda, it was going to happen. So. We had 300. So we had 300 attendees from 12 countries. And if I miss anything like in the, in this historical here, like, you know, throw in so 310 days, 12 countries, 28 states, predominantly north America. But you know, there was a spread across, across the world.
[00:08:49] Um, so we go into 2020 thinking, all right, five, 600 attendees, run it back. And Cleveland, uh, this is gonna be huge. And the audience keeps growing user base is growing. I'm starting to think about raising money. Like things are heading in a good direction. Then obviously March of 2020 hits and the world changes for everybody.
[00:09:07] And we all the sudden don't have an event. And, um, you know, COVID sort of wipes out the event business for, for net, for us and everybody else. So we scrapped 2020. We launched, uh, AI academy for marketers instead in July of 20, with 40 on-demand classes and five certifications. And, you know, thinking that that's where we'll go just online education in the near term.
[00:09:27] Um, that goes fine. 2021 rolls around. You know, we, we think we might be able to pull an event off and then it becomes real clear early in the year that that's not happening. So 2021 is virtual, about 300 attendees to the virtual conference, and then going into 2022, um, now our audiences, you know, what are five times the size?
[00:09:47] It was in 2018, 2019. So you get more and more confidence in your ability to, to build a conference business. Cause now we do have a growing community, a much more to pull from. What we faced like every other event. And this is why we thought this might be an interesting conversation to have, um, is there's a lot of obstacles.
[00:10:06] Like, let's be real about where we're at in the world. So when we look at the challenges we face right now, the headwinds to, to running a conference and the why we need to be as smart as possible and use as much tech and every, every tool we have, uh, at our disposal, we obviously still have COVID, you know, there's still lingering concerns.
[00:10:23] There are travel concerns, uh, depending on other countries are almost like rule out other countries coming in because you just, you have no idea. It's almost like if someone does come in international, it's almost an outlier at this point. Um, but you, you can't control that. Uh, I, I personally, as of January 2nd, when I go back and look at the investments I was making on January 1st and second of this year did not foresee the NASDAQ crumbling 50% or more.
[00:10:49] Um, so, uh, you know, it is what it is, but the tech industry, which again, is our major advertisers and sponsors. Now, again, it's not all publicly traded companies that would sponsor us, but th th it's all connected. So the world where you draw from, from a revenue perspective there, everyone not only individually, but as a company is getting affected by the current economy and by inflation.
[00:11:12] And you have to not only look at that from advertising perspective and sponsorship, but. Which companies have the budgets to travel like this discretionary income for pro dev or whatever you bucket you put it under. Is that getting sucked out and, you know, where's the data on that? Um, obviously there there's geopolitical issues going on there's war happening.
[00:11:33] Um, as some of the tech companies we work with, some of our sponsors are in Kiev, like th th this is the real world and these people have way other concerns, much more real concerns that like, I don't even want to be talking to them about anything related to this. I'm just checking in on personally, like, where are you guys at?
[00:11:49] How are the employees like, and you have to just consciously be aware that like us selling sponsorship is the furthest from our concerns when, when it comes to that kind of stuff. Um, and then the other is. Uh, people just don't want to travel as much. Like we got used to being at home. So any headwinds I'm missing Mike?
[00:12:06] Cause I feel like,
[00:12:07] Mike Kaput: well, I haven't checked the news in the last two hours, so maybe there's new ones now. That's yeah, it's just a lot has come together that have, it's like, it's really the uncertainty, right? We can't necessarily sit here and say, okay, the war in Ukraine versus the economy is going to be an obstacle to events.
[00:12:31] We're not sure based on who comes, but the point is they both could be like, we have to get so much smarter about how to actually. Navigate a very crazy world. And you know, there's always been crazy stuff going on in the world, but it just really emphasizes to me, like we've been preaching that people need to be smarter about their marketing and this was pre COVID when things were fine.
[00:12:55] Now it just seems like an absolute critical necessity to me. Like the playbook of the past just straight up
[00:13:04] Paul Roetzer: is not going to work anymore. Yeah. And, and kind of along those lines, I'm going to, I'm going to read to you all, something that I wrote last year. So this again is like COVID was happening, but these other things weren't yet, and this was at the front end of our ticket sales strategy, our internal brief for our team on the ticket sales strategy and anybody who's familiar with marketing AI conference, the tagline is more intelligent, more human.
[00:13:27] So our whole thing is teaching you about AI and what it's capable of, but so that you live a more fulfilled life and have more fulfilled career. And that we build better businesses as a result. That we save a bunch of money, cut staff and maximize profits. Like that is not what we're here to teach. So this is an excerpt.
[00:13:46] And what I started doing is every, uh, marketing strategy, not just for the event, but anything, I would start off with these two elements. So the first is more intelligence. And so for our ticket sales strategy, this was last year. But again, it applies to this year. Uh, the first bullet is predict high priority contacts based on probability to purchase.
[00:14:04] So we need to know, we need to make predictions, which is what AI excels at. If again, if you follow us at all, like we talk all the time about marketing is basically just a series of predictions about everything. So predicting high priority contacts based on probability to purchase predict 2022 sales. So when we're forecasting and I'm talking to my event team, and I'm trying to tell them how many people are going to be there and how much spacing we need in each room and things like that.
[00:14:29] They want accurate numbers. Well, how like that, it's a really hard thing with all this uncertainty to get to predictions. So it's like, can we get better at making predictions? The third predict sales of future Institute initiatives such as online education. So you're trying to go beyond just this event and say, okay, well, if we, if we launch something new for online education plate at the event, are we able to predict sales of those?
[00:14:55] And I'm trying to look at revenue now and I'm again, I'm the CEO. Like I gotta think a lot of things there's tactical things, but I got big picture things and I got investors and things like that. And then the last is driving efficiency in marketing resource allocation, like time and money through more targeted campaigns, get rid of the waste.
[00:15:09] Like it's just, there is so much waste in the th the bullet I don't have here. Mike can attest to this cause he runs a lot of our email and the newsletter and the. The thing I worry about all the time is the noise. So we have 29,000 subscribers. They do not all care about Maicon, there's a big chunk of them probably are never going to come to a conference, but they may care about the content marketing stuff we put out, like, how do we stop blasting?
[00:15:34] And this is, again, this is a universal, like demand gen Legion, whatever you're in. How do you stop blasting a bunch of stuff to the all contacts list or to just like the big segments, just because you'd need to hit a number. You got to get the open rates or click rates or whatever it is like enough. Like, I hate it.
[00:15:52] I hate doing it. I hate getting those emails and it's like, we have the tech to be better than that. So that's okay. So the more intelligent predicting probability of purchase, predicting actual sales, predicting sales of other initiatives, and then driving efficiency and resource allocation. And then my other one of like getting rid of the noise.
[00:16:11] More human. So this is like, okay, if we, if we do that now, ideally, we'll talk about using AI to do those things. But if we can do those things more intelligent marketing, the more human side is understand individual learning paths, journeys, and needs. So make it about the people about the actual end customer learner, the leader, like what do they actually need?
[00:16:31] Don't just blast them a bunch of stuff that they don't need. Like what are they here for? And getting to know that information, then prioritizing context for one-to-one engagement, nurturing, and community building. So again, rather than spending a bunch of time sending emails to like 29,000 people is that we don't know who they are.
[00:16:49] Who are the a hundred people. We should be having conversations with them in the next 30 days and getting to know personally, and being able to use that to enrich the way we communicate with these people that then leads to personalization of messages and experiences and marketing at scale. Like once we have a better understanding these people, now we can create these incredible experiences for them online and offline at our events.
[00:17:09] And then the final piece. And again, like can appreciate this as the guy who has to create a lot of this content, focus, the staff on strategy and creativity while investing significant resources and interpersonal communications with our audiences. So rather than doing a bunch of the mundane stuff, like building all the lists and analyzing the data and running all the pivot tables and do all this stuff.
[00:17:27] Can we free Mike up from having to do that. So Mike can actually just like sit back for 10 minutes and think about creatively, how to solve these problems. So that's our more intelligent, more human framework. I've kind of stopped there, Mike, and see if you got anything to add to it.
[00:17:41] Mike Kaput: Uh it's absolutely.
[00:17:41] Yeah, I think, uh, it's a good way to frame it because I honestly think, and I get it. Everyone is everyone is having a hard time. Doing relevant marketing, cutting through the noise, but too often people think that the more intelligent part as let's see you, some automation or let's look at an analytics report.
[00:18:00] Yeah. That's all good, but that's really not what we're talking about anymore. When we're talking intelligence, that's like the price of admission in my mind. And then the more human part, I think a lot of people end up just in this kind of, uh, you know, half-baked definition of it where we say, oh, okay, that email is personalized.
[00:18:19] We put a personalization token and, or we switched out a little smart content and like a line or two to three different segments. And it's like, okay, like, again, it's better than nothing, but that is no longer. Even what we're talking about. Things are light years beyond that in what's possible. And what's also necessary.
[00:18:39] I think all the points you mentioned, just all the headwinds. I mean, everyone knows it, even if you don't say it, like everyone is overwhelmed, stressed out. Uh, if they get free time, they're taking all the free time they got from working from home now and reinvesting like what they find important and chances are, it's like not your marketing emails.
[00:19:00] So it's probably gotten dramatically harder in the last two years alone. And so I just, I think it's critical to look at it through those
[00:19:08] Paul Roetzer: lenses. So again, we're just kind of running through why we're sharing what we're sharing while we're going to talk to you about like a specific our event and ways we could be smarter with what we're doing and helps you learn something.
[00:19:20] So another one that has created this for me personally, like, there's this great sense of urgency? Like we've been talking about AI. I mean on and off since 2014, as an Institute, as an entity, since 2016, we have invested thousands of my personal time. Thousands of hours of Mike's personal time, hundreds of thousands of dollars of my own money, millions of other people's mind, like this is no small thing we're trying to do.
[00:19:46] And we're trying to create a movement, like get people to care. And here's where I arrived at recently, when we think about the kinds of companies that will exist in, like, I would put it like five to seven years where it becomes essential, probably 10 to 12, where. Yeah, it's just it's game over. And what I mean by that is there's three types of companies that like, let's say five years from now that will exist.
[00:20:10] There will be AI native companies. So these are organizations that wouldn't have been created without artificial intelligence, that the predictive power of AI, the language envisioned technology, that they just literally wouldn't be able to do what they do as a company, as a core offering, if AI didn't enable it.
[00:20:27] So those are like the newer companies that have been created since probably 2011, 2012, that were based on what AI enabled. Then there's the AI emergent companies. These are the organizations that figure out they can build smarter business models. So Netflix would be a prime example here. They used to mail DVDs.
[00:20:44] There was no AI in that. Now, you know, how many versions of the thumbnail image do you see on the thing? Like they switched over toward their business model today, not possible without AI. So AI native emergent in my belief is every industry, every company. We'll we'll have to be AI emergent. Like you won't have a choice because someone else will be build an AI native company better than yours.
[00:21:07] So if you're a marketing agency or graphic design agency, or if you're a freelance writer, whatever it is, you have to solve for AI, you have to be AI emergent or else someone's going to build an AI native company that's better than yours, smarter than yours, more efficient than yours, better performance than yours.
[00:21:21] So AI native AI, emergent obsolete does it. Those are the three options for the types of companies that exist five to 10 years from now. And so like, I believe that with all of my soul, that, that is where we are going as an economy. And so like this urgency for us that like, we have to make people care, the marketers, the business leaders, the investors, the educators at universities, like they have to care because we don't have this.
[00:21:47] Isn't a decades in the future thing. This is a now thing. So. That's the, that, and then the other thing that that happened and what led to us saying, let's do this one is last week, we were working on the lead scoring model for make on ticket sales. And so this is really the jumping off point to today's conversation.
[00:22:04] So what we started doing is last year when I built the lead score, um, it was all manual. We were not using it. I know HubSpot, which is our CRM does have a native machine learning model in it. Prioritizes leads based on some factors. It seems as though the main factor in it, again, machine learning, you're not supposed to be, I'll just like open up the doors and see what's in there, but it would appear based on a lot of testing that it was heavily reliant on whether or not someone was in a deal pipeline.
[00:22:34] You know, if you had an, a lead and what stage they were in. So it's not overly complicated stuff. Number of page views, number of email clicks, things like that. You could pretty much build your own version of it and know exactly went into it and have it probably be better representative. So that was a very early effort by HubSpot though, like probably 2016, 17.
[00:22:53] I don't know if they've improved that or not. I have not used it in two years. So this is not a ding on HubSpot. I don't know. There are companies out there like mad, kudos and example. People swear by Matt kudu. Like I know Jeff Coyle, I don't know if he's on the case study. I understand that. But like Jeff loves Matt Cruz.
[00:23:08] Like he was the one that put us onto it. Jeff called market muse. So like we know there's products that are really good at this, but they need like 2000 conversions a month for them to justify using machine learning. So what we looked at and said, okay, how do we just build a smarter human based model?
[00:23:24] Like just smarter rules in this thing. And so last year we went this model where we looked at intent signals, loyalty signals and engagement signals, but I didn't, it was, it was a evolution, but it wasn't an end product. And so. We started there two weeks ago and we're looking at may con ticket sales for this year.
[00:23:44] We said, okay, how do we make this smarter? And so what we ended up doing, and we're going to kind of talk through today is this idea of intent signals are things that someone behaviorly told us, they're interested in may con they put on a register button, a buy now button. They looked at event pages. They, um, they downloaded a promo code, like whatever.
[00:24:02] Like they did something that specifically tells us they know this. So I don't need to like waste my time, emailing them and saying, Hey, may con is an a annual conference that happens in Cleveland. And we have these amazing speakers and all this amazing content. And you're going to get this advantage in your critic.
[00:24:16] They know that like, w we know that because they're not signals, tell us they know that then there's loyalty. These are people who have previously bought from us for a conference or online education are the two main components they've spent money, which means they've they've debt. They've shown some dedication to the organization or some interest and demand for the, uh, the content.
[00:24:35] And then there's engagement, which is just like all the other stuff. So number of overall pages on the site, email opens email clicks, um, uh, are they a member of the community? Like just all of these things, you mash it up. There's like 50 signals in that one. The intent when at like 16, the loyalty one had, I I'm almost done building it.
[00:24:53] I built it this morning, honestly. Um, I think it was like 12, like not a ton of signals. And then the engagement, when I, I think I was like 46 or something and I'm not, but I'm not done with that one yet. And then we have an ICP, one that we're building, that's like another squirt. It says, okay, is this someone that's our ideal fit?
[00:25:08] Which for us is like director level and above probably 200 plus employees. And then in one of our seven tier one industries, and I'm not gonna, I'm probably gonna forget what off the top of my head. Let's say it's financial services, CPG retail, healthcare, uh, marketing agency slash professional services software.
[00:25:29] I'm forgetting one or two, but that's about it. Yeah. Basically they have a ton of data. They need to personalize content. And they've probably been using AI for 10 years in other parts of their business. So they've proven it works. They don't need convincing. So like our core audience, the people who are likely to come to the conference don't need to be convinced that AI matters.
[00:25:47] We spend our entire year on the audience that like making them care that it matters the intro day for marketers, which gets like 500 people. Every time we do that live class, all the content might create. Like we know, I mean, we have like 25,000 visitors a month to the website. Like we cater to the intro audience.
[00:26:05] That's just trying to like, see if they should care. We need the people who, who are ready care. Cause they're the ones that are likely going to show up. Problem is who are those people who actually understands and cares? Because our data is, it tells us like there aren't that many. So a couple of weeks ago, a month or two ago, I came up with this theory of this AI adoption model.
[00:26:25] This. Basically, I think I've told this story on webinars on it, but like the Fermi paradox of AI. So for me, paradox is you look up at the universe and like all these stars, all these exoplanets, like all this stuff, and yet no intelligent life. Like where, where is everybody is the Fermi paradox. I feel that way about AI.
[00:26:40] We know it's critical, it's affecting all these things like trillions of dollars in annual impact. And yet you look around marketing, you look at the top CMOs, you know, according to all the lists of like, these are the 50 best CMOs in the world. Nothing. Like they're not talking about AI. They don't have public points of view on AI.
[00:26:57] So it's like, well, who, who does, like who in the marketing organization is actually already using it or knows they should. And so we came up with this theory that like, they may have digital in their title. So maybe it's chief digital and marketing officers, not just the chief marketing officers. Um, they probably a personal, if they, if they have personalization in their title, they don't have a job if they don't understand AI in two years, maybe, maybe two months, um, data and analytics is another one.
[00:27:22] Um, digital transformation was one. Let me, you might think of some other ones. Um, MarTech, uh, pricing was another one. Like if your job is pricing and you're not using AI, I, I don't even know how you have a job right now. Like predicting pricing models at scale is like, it's the perfect example where AI could be applied.
[00:27:43] Um, digital marketer mentioned e-commerce revenue, maybe. So we basically created this high, conversational, this hierarchy of based on your job, title and keywords within the job title, we kind of built a predictive model around the likelihood you would be to be adopting AI over the next 12 months. And then the impact AI would have on your career over the same given period of time.
[00:28:07] And then you project that over one year, two years, five years, and you start to actually like see the movement in the industry of the necessity that people have AI. And what we know is like these general marketers. They don't, there's 11 million of those. And someone may get by for the next five years without knowing about AI or caring about AI, it's just not in their function.
[00:28:25] Like maybe it's like, not that I'm a, I'm a PR major. Maybe it's a PR person. Like thank you goodbye for five years about AI. Like it's not going to not going to ground swell where every PR agency, all of a sudden figures out how to build smarter media databases and target journalists based on actual interest in using natural language processing to read everything they write and have it predict what pitch you should make.
[00:28:46] And when like that should happen, it's not like, so you got time if you're in PR, but if you're in personalization, automation, data pricing, digital, you don't have time. Like it's here. Like it's changing your career and you're their move or you, or you don't. Yeah,
[00:29:09] Mike Kaput: I couldn't agree more with that. I was actually just on a, like Twitter live chat with content marketing Institute.
[00:29:17] Um, since I am speaking at their content tech conference, um, I think later in may, May 31st to June 2nd. Um, and they asked, what are the top mistakes marketers make when it comes to AI? And, you know, in the past, I probably would've given them like five to seven things that people do wrong when they're piloting a.
[00:29:41] Uh, people do wrong when they're thinking about what AI can do. But I gave him one answer this time because it's just really readily apparent to me. The biggest mistake is not thinking they need to care about it right now. And it's either out of fear or it's either out of ignorance thinking this doesn't apply to me and it's like, it's going to apply to you like it or not.
[00:30:05] And I. I think the biggest mistake is sitting around thinking that you don't have to pay attention to it now, or like thinking, okay, I'll get to that later, next year, whatever. No, that's a, that's a critical error I would argue in someone's marketing career.
[00:30:22] Paul Roetzer: Yeah. And that kind of goes back to what we talked about.
[00:30:24] Episode 14 was like, what is lost and what is gained. And when specific to every career you just think about what is AI going to do to your career path? And what's the opportunity for you in that when you accept that it's it's here and it's going to change stuff. Um, so to, to get to the, the core of what we wanted to cover today, which is like AI for event marketing.
[00:30:43] And we use, I, I don't know, like a dozen different AI tools every week, like different functions and Mike can get into some of that. Um, but for event marketing in particular, We had to stop and think so when we go back to that whole idea of the more intelligent, more human it's like, well, how are we making this more intelligent?
[00:31:01] I'm building this intent score, loyalty score, engagement score, which is going to dramatically improve our ability to do personalized outreach and customize learning journeys for people. Like it's a really important step forward for us to help people figure out AI and apply it to their careers, but it's not using AI yet.
[00:31:16] Like we're not, there's nothing it's me crunching data and thinking about which signals might matter. And looking at past buying data and putting conversion rates related to pass lists, but it's basically just going on HubSpot building dozens of lists, and then predicting which signals will have high, medium, and low intent, and then assigning a score to them that that's the process.
[00:31:34] And then taking those scores and then prioritizing who we're doing personal outreach to. So like, if you you're listening to this and me, Cathy, someone from our team personally reaches out to you in a week, two weeks, whatever it's big because. We, we see intense signals. Like we see that like, okay, this is someone who cares, how can we help them?
[00:31:55] Like our approach right now is like, what do you need in your learning journey? If it's not the conference, that's cool. We have 900 articles. We have access to 70 like instructors and past speakers. Like we have experts in this area, like, what do you need? And our feeling is like, if we build this Institute around, just help people along their learning journey and are on their career paths.
[00:32:16] Yes, we have to make money as a company like that. That obviously we won't be around, but if we focus on the, the learner, we focus on the professional and we, we ask that question enough times, what do you need? Where are you at? What are you trying to do in your career? And we then build. Sessions and online courses and webinars like we, if we create the content around answers to those questions, then everybody wins.
[00:32:39] Like, and so that's the theory here. So the question becomes like, well, how, how do we make our event marketing smarter as we're going through this beyond just some obvious human-based things. So Mike and I thought it was like, well, well, the two ways we teach people to do this is one the use case model.
[00:32:58] So if you've taken our intro to AI for marketers class, or if you've even taken the online course about how to get started with AI, um, the use case model is make a list of all the things you do. So we can just take our event marketing, like personalized emails, picking segments, creating lists, building lead scores, writing email copy, picking CTA is like, just create a list of all these things we do to do our marketing targeting companies in certain geographic regions, whatever it is.
[00:33:22] Then just go through and assign what would be the value. If there was an AI tool that could help us do this, could augment our ability to do this better, or even automate in some cases, this process like performance reporting. What if we just use AI to automate all the performance reporting? So we didn't have to spend time doing it.
[00:33:37] So that's a use case model. And then you go through and say, okay, is there tech to actually do this? Like, do we have the ability to actually automate this? So that is one way we could do this. We could just me, Mike, Cathy, McPhillips our chief growth officer. We could just sit down, make a list of the 50 things we got to do over the next 11 weeks or whatever it is.
[00:33:54] And then just like find the right tools for them. But we actually thought the more productive thing to teach you all is the problem-based model. So in a problem-based model, you identify a problem statement. You come up with the issues and drivers that are, that are causing that problem. You assign a value to solving the problem, and then you come with a hypothesis of how to solve it.
[00:34:13] And then you go through a matrix like, okay, what are all the ways we could do this? And you all along the way, you're kind of validated the issues and drivers were correct and all this stuff. So. We chose to go down the problem based model for you today. And actually like again, behind the curtain, here's what's going on.
[00:34:27] Here's what we're thinking about in real time, justifying potential uses of AI to ourselves in real time that we may ourselves go act on tomorrow. So that's kind of where we're at. So what I want to do is when we teach the problem based model will tell you like, oh, sometimes it can take three months. We did this in three minutes, like no lie, like Mike and I got on the call five minutes before we started this.
[00:34:50] And we're like, all right, let's go problem based model. Okay. Pull it up. So in our chapter on, you know, the chapter like chapter four of our book, getting start with AI, this, this framework is in there. So go by the book, every everything we're about to tell you is written out and page with the example of email marketing.
[00:35:07] So we can with the problem statements, the first step. So we said, we need to sell 300 tickets to an AI conference in a world in which travel restrictions, health and safety concerns and the economy. Present unprecedented challenges to achieve the goal. So I was taught that like eight weeks, I think it was eight weeks out from your conference.
[00:35:25] You should be 50% sold. So that gives you a benchmark, like at eight, at eight weeks, it's like, okay, cool. If we're at 150, we should be on target to hit 300 tickets like that. That would be the pre pandemic world formula. Now again, if you're an event, marketers spend 30 years in this and I'm wrong on that formula, please reach out to me on LinkedIn.
[00:35:42] I would love to hear from you. I hope this starts a conversation with event marketers. Um, we are not here saying we figured this out. We were telling you, we're doing this in real time, trying to figure this out and hopefully it helps you, so the value to solving this problem to us. Cause again, problem statement should also come with a value statement is 300 times X, whatever it is we think we can sell a ticket for.
[00:36:02] Um, and then there's the future of it. So I think like right now on our site, the tickets. 1399. I want to say. So let's just theoretically say it's 300 times 1399 would be the value. That's the direct value. The indirect value is we also rely on sponsorship, revenue and support to run an event like this.
[00:36:22] And if we don't deliver 300 attendees, why would they sponsor next year? So like, it is, it is on us as an event that we have a responsibility to not only the attendees to connect them with the right technologies, but to the technology companies to connect them with the right attendees. And so the future value of the event.
[00:36:41] So it's beyond just the 300 times 1399. It is, well, what, what is the loss opportunity next year, if we don't deliver? And so you can actually make a case like, okay, this is a much, much more valuable problem to solve than it might look like on the surface. I would all am Mike, and I've done this for different corporations as a consulting gig.
[00:37:00] Like we'll go in and run this problem based model in a consulting format to help them figure how to use AI to improve their business. So I w I would stress very strongly problem statement has to be accompanied by a value statement. You have to know the value of what you're solving, because that tells you how much you're willing to spend to solve it.
[00:37:18] Any thoughts on problem statement, value statement, Mike?
[00:37:20] Mike Kaput: Yeah, no, I, I think that's exactly right. Like the value of it is how, if you are someone that is more of an early adopter of AI and trying to pilot these projects, that is the number that gets you buy in from executives and from colleagues, because then you can say, Hey, it's worth spending.
[00:37:39] Uh, nominal amount of money to test out a certain tool over three to six months or whatever to potentially have this much gain. It becomes a business use case that I think that's a really streamlined way to describe these things, but yeah, for our purposes, it's just that it's very straightforward
[00:37:56] Paul Roetzer: sell tickets.
[00:37:58] And then in the problem-based framework, then you do what are the issues and what are the drivers? So for us, let's say issues. And again, this could apply to any event, not just ours, uh, slower than usual ticket sales. So maybe you're not trending at that eight week mark. And again, it might be the new normal, like maybe everybody waits until six weeks out to buy tickets these days.
[00:38:15] I don't know. Like I only know a few events that have actually been run in person in the last three months since we sort of came out of. The pandemic into the endemic or whatever it is. So I don't know what the model should be, but it would appear it's not what it used to be. Price sensitivity. Are people more sensitive to price due to the economy and due to inflation.
[00:38:38] And is that, is that changing by day to Mike's point to start if we wake up and the market is down 4% again, did that just increase the sensitivity to price of our buyers? Yeah, I think that's
[00:38:50] Mike Kaput: a really interesting quick point to dwell on is the speed at which things change, because obviously we always knew like inflation could be a serious issue down the road, but like how quick did the narrative go from?
[00:39:04] Like, okay, like we're getting out of the COVID COVID room, you know, post COVID recovery and then suddenly. Every single person is concerned about this in a big way. And that's because it happens so fast and so dramatically price spurred on by, you know, things like war in Ukraine, this stuff is happening over a matter of days and weeks.
[00:39:22] Like you thought you had to move fast before that's like a laughably slow pace now.
[00:39:27] Paul Roetzer: Yep. Uh, another one is location. So we, we love Cleveland. We're from Cleveland. Cleveland is hosted tons of international events. I mean, they, the NFL draft here last year in the middle of pandemic, like it was like the first coming out party of events period was like the NFL draft in April last year in Cleveland.
[00:39:44] Uh, our friend Joe Pulizzi ran content marketing world in Cleveland, still in con in Cleveland. I mean, there were 4,500 people here coming to Cleveland marketers. Like we know people will come to Cleveland for conferences. The convention center is amazing. It's amazing entertainment restaurants like the iron chefs.
[00:39:58] Like it's an amazing city, but. Is it have the same perception when people are less likely to want to travel? Like if you haven't been on a plane in two years and you're going to get on a plane for the first time, is it to go to Cleveland? Or like, we're not like we don't know. So, but you have to accept that like, oh, okay.
[00:40:18] Like, well that wasn't a variable three years ago. Like never heard a single person say I'm not coming to the event. Cause it doesn't Cleveland. People love it. Like on the, on the, the surveys after the fact it's like, love the city, like love everything. Like, so we non-issue before maybe an issue now, I don't know.
[00:40:35] So issues slower than usual ticket sales, price, sensitivity, location drivers, what's causing those issues. COVID can't get around it. It is what it is. Inflation in the economy, war people maybe don't want to travel. And then the other one I'll throw in there is, again, we're not starting with a hundred thousand.
[00:40:52] We're not a media company. That's talking with a hundred thousand readers. They just need to do. 300 of the a hundred thousand. We built a community from nothing like at a time when no one else was talking about AI in marketing a time when the platform companies had probably fewer than a handful of AI and ML employees working on marketing software, making it smarter.
[00:41:14] So in 2016, this was not a discussion outside of probably Salesforce and Adobe. Like, I don't even know Google Microsoft, obviously, but like marketing software, it, it wasn't there yet. So we got in really early and tried to create a movement and build a community around something that we thought was going to change the world and change the industry.
[00:41:36] And we believe that more now than ever, but we don't have that massive engaged community yet. So how do you get to where we need to go? Without that community right away. So, okay. So then the hypothesis. So problem statement, issues, drivers hypothesis, like how, how do we think we can solve this? So again, I just grabbed this right out of our ticket sales.
[00:41:57] This is like literally verbatim from our ticket sales strategy. Brief, the greatest indicator of probability to purchase is intent followed by loyalty engagement, personal connections, and then strangers and the strangers. I'm not going to spend a bunch of time on today, but that that's like strangers with that fit ICP kind of data.
[00:42:16] Um, we need to focus our personal outreach on the highest probability segments in order to reach our ticket sales goal, the higher, the probability of conversion, the more personalized the communications. So what gives the strategy, a greater probability of success than previous models? Stronger focus on intent signals, integration of recency into signals.
[00:42:34] So like if you're, if you're a market and you're building a lead score, Somebody has been viewed pit 50 pages or more on my website when three years ago. I haven't have you been on the site since then. So if your lead score is in factoring in recency. So for us, I usually run it back. Like, let's say an example would be 50 plus page views on the site and has been on the site in the last six months or been on the site in the last 12 months.
[00:42:56] So like, I know that they are still them, that they are still active. Um, we give preference like for our online academy, if you've been a member ever, like if you've ever previously purchased the annual or the monthly plan, you would get point values. But if you're an active paid member, you're going to get more.
[00:43:14] So like recency is always thought of within these signals and then, um, using low to high scoring system for each segment and then prioritization of context based on segments. So our hypothesis is that if we can get. More predictive on probability to purchase and we can get more personalized in our communications.
[00:43:33] So again, help them along their learning journey, like know what their learning journey is and help them. Then we should be able to achieve it. Then the questions becomes, okay, great. If we believe that hypothesis to be true, how the hell do we do it? Where, where do we go smarter? So what we have been focusing on is human powered manual build, build a better lead score, not a smarter, it's not, it's not learning on its own that lead score isn't going to improve unless I improve it.
[00:43:58] This is a really important example. So I build this intent-based score. We start running it tomorrow, personal outreach based on it. And let's say it starts working like, and we get conversions coming from that. Unless I go into that intent score, look at conversion data and tweak the weights, change the point values based on it's like, oh, wow, that was a signal.
[00:44:18] I didn't think would be an indicator of purchase, but 44% of people that had that have now. And that was seven of them. Like, is that enough to say that that's actually a real signal or is that just noise? That's where AI comes in. So if we were using AI to do lead scoring machine learning would constantly be analyzing the data and improving the score itself.
[00:44:44] So right now we are living in a world where I built a score. We will use that score to manage not only personal outreach, but targeted email communications and ad targeting all these other things, but it's not going to get smarter unless I make it. What we want to do with everything we do. And this is where AI comes in.
[00:45:01] Everything we do as marketers should get smarter with the help of machines that should continually learn and augment what we're building, the strategies we create. And so that's, I'll kind of start, there's like a jumping off point because I don't like, like you put a bunch of ideas in here, ways we could, and I'll let you kind of tee up, like, this is not, we're doing this, this isn't like, we're saying you have to do it.
[00:45:22] We are now going to share our thoughts on how event marketing can become more intelligent, more human, basically. Yeah,
[00:45:29] Mike Kaput: for sure. Um, yeah, to start like, like Paul mentioned, I mean, we are. Experimenting at any given time with dozens of potential use cases for AI across some of our core operations, email content ads, um, probably a handful of others and not thinking of a social.
[00:45:49] And today we use probably a dozen different tools to power our strategy. And I can tell you for a fact, we have saved probably hundreds of hours and gotten much better results using a lot of AI tools in email and content marketing and advertising specifically. And what we've known for years is that every area of marketing is not created equal in how mature.
[00:46:13] AI is in that segment. So content marketing is a huge, hugely advanced thanks to advancements in language models. Um, advertising is hugely advanced because there's so much data to learn from. Um, I'm not seeing the same thing, like I said, in a event marketing, I think it's actually relatively immature. So we're using a bunch of AI tools, uh, for use cases in our business and other areas.
[00:46:39] But honestly, I think a lot of AI that we could be using in event marketing potentially doesn't exist or isn't far enough along yet. So a lot of these are kind of, I know these things are possible in other marketing segments or other industries. They could all be applied to the event world. So even just starting with really basic things, like predict for me the likelihood of someone to buy a ticket.
[00:47:07] And not only that predict the likelihood of them to buy bulk tickets, presumably have some event expert out there with 30 years of experience in our space knows who is most likely to buy a ticket. What's the average income, what's the average title. What's the, um, what size organizations usually buy in bulk?
[00:47:27] I mean, there's probably some data out there or could be if it was tracked somewhere, say in an event management platform, all of that could be used for really smart
[00:47:35] Paul Roetzer: predictions over time. And if, even like, I just thought about this, like, again, there's the whole point of this, like Mike and I did not go, we haven't prepared for this section at all.
[00:47:43] Um, so this is going to be probably a lot of off the cuff things. Awesome. If it's on your own site and your own historical ticket sales data, how much better if you're an event marketing software company and you can anonymize ticket sales data for everybody, and you have hundreds of thousands or millions or billions of data points, potentially.
[00:48:04] Like this has been my argument for like HubSpot and other CRMs. Like if you all the performance data you have access to, man, could you build like smarter solutions? Like where is it? Like, why, why don't we have these predictive models yet? And I made a note, um, in your show. And I was like, if you match this predicting Lego to buy a ticket with a confidence level, and this is like how Salesforce pipelines will work confidence level.
[00:48:27] I now can get to like what the number I want, which is total events, sales, and revenue, based on the current model of the likelihood to buy across the entire contact database. And you're the one about the bulk tickets? Like what is the overall conversion rate is predicted to be 2.7% with an average purchase price of X.
[00:48:46] And here you go, like, and then each day you're just looking at the confidence level and looking at how that number moves rather than us picking a number out of the air. I'm like, I don't know. I think we can sell 3 25 at 1490. Like, we're just guessing, like we're not analysts, like maybe there's like data geeks that are really good at this stuff, but we're not like we're marketers.
[00:49:06] We can run pivot tables and we can build formulas. But like at the end of the day, that's not our world. Like that's where AI should be able to help the average marketer.
[00:49:14] Mike Kaput: That's really interesting. The point you made about platforms is like, I would love to just, if anyone's listening or afterwards, we can look up, like, what are the top three platforms we have.
[00:49:26] Event platforms for our type of event. Like one of them has presumably tens of thousands of customers. I assume, given how many events have been held, like we should be like that, whoever they are. I would love to know if they have this online
[00:49:41] Paul Roetzer: format or God, we're just going to give away like trade secrets or like some big competitors are going to go do this.
[00:49:48] So like, if you're an event company, let's say you're maybe a publicly traded event company and you own hundreds of events, you can anonymize, or you don't have to anonymize. You can aggregate sales data. Damn.
[00:50:00] Mike Kaput: Yeah, you would at least. And even if it wasn't perfect to start, you would at least. More than you have today.
[00:50:08] I
[00:50:08] Paul Roetzer: guarantee you all right. If you work at a major event company and you have hundreds, dozens, or hundreds of events, like give us a call. I think I have an idea what you need to build slash
[00:50:17] Mike Kaput: buy a, buy a ticket to the event because that's like worth it alone. What? We
[00:50:21] Paul Roetzer: just talked about, buy a ticket. I'll I'll buy you a drink.
[00:50:25] You bet. I think. Okay. I, I think I know that. Okay. I'm not going to talk more about that, but I have to do. Go ahead. So,
[00:50:34] Mike Kaput: um, another, so this actually kind of plays into some of these newer factors that, um, are affecting the event world. I mean, This would be, I think very, very difficult to start with, but it's possible is why can't someone.
[00:50:51] We have tools already that are AI tools that look at things like market and economic factors. We have tons of sentiment analysis tools that are tell you brand sentiment and commentary on social for whatever segment you're looking at. Presumably there would be a couple of signals where you could say, okay, show me like a combo of like fuel prices, airline tickets with inflation and fuel being crazy.
[00:51:18] And then show me like in this segment, you know, commentary around events or even around the pandemic, perhaps I think there, you know, that's very vague, but I think there are probably signals where you can say, okay, if I could analyze a certain groups commentary on social media over a three month period, what are they negative, positive or neutral.
[00:51:42] In-person things like, are they all super site that they never have to go into the office again? Or if they say, oh my God, I'm not traveling, looks terrible. Like that could tell us something about sentiment about like almost physical likelihood to physically attend an event.
[00:51:58] Paul Roetzer: You know what I want, I don't want for speakers where I can feed it a list.
[00:52:04] Let's theoretically say top 50 CMOs in the world, feed it names. Maybe I can give it Twitter accounts. And then I want the AI to go out using natural language processing, analyze all their public data, every talk they've given, that's transcribed. Every social media share every interview in a publicly available media.
[00:52:26] And I want to report a brief handed to me by the machine that highlights the top five to seven CMOs who have a public point of view on AI and summarizes it. Yep. And there's my speakers. That would be awesome. So if you're a tech entrepreneurial and you want to build that, give me a call. We'll just build it ourselves.
[00:52:45] But I want that, and I want the same thing PR I want the same thing for media relations. Here's the topic our client, or we want to talk about in the media. Go analyze all the journalists in that space. Find the people already talking about it, their sentiment around it, their tone around it. Like, come back to me and say, here's the four journalists you should be talking to.
[00:53:05] Here's the angles that would resonate best with them based on their personalities, based on what the parents say, this is doable. Like I know a guy who built this stock. I may call him after this. Like, this is not hard. Like what we're telling you doesn't exist, but I promise you, there are people who could build this probably in a weekend with the right access to data.
[00:53:22] Like these are tools that truly can be. In real time almost,
[00:53:28] Mike Kaput: you know, I think another interesting one that is much closer. I honestly, it should exist already. I'm sure it does. You could use an existing tool for this. If you were at the enterprise level, we're sometimes a little too small to be doing, you know, a thousand AB tests across 200,000 emails at this stage.
[00:53:45] But really if you are running a huge conference, if your way is publicly traded companies that runs all these, I would seriously be looking at some type of tool to at the very least do a thousand AB tests at once of every possible combo of general event. You know, subject line, copy angle. Focus on in one email, um, the awesomeness of getting back together in person, see if that resonates and who it resonates with.
[00:54:17] There's CF cutting prices works in another email. And you mentioned inflation and trying to do everyone a solid there's a hundred different things you can do. And we do this again
[00:54:28] Paul Roetzer: manually. What message do you think resonates? Right? You can talk about the economy. Talk about location. Like where do we go?
[00:54:35] And go ahead. I almost
[00:54:37] Mike Kaput: thank you. And it just goes down so many levels because then you have data on, oh shoot. Out of 2000 people we'd like to attend. 20 of them are really, really strongly caring about getting back in person. Let's give them a discount or give them a hotel room or something. You can personalize promotions endlessly.
[00:54:56] You can literally be selling almost like a differently structured event to different segments.
[00:55:02] Paul Roetzer: So as an industry, Let's all agree. Personalization is not insert name, insert company. What's possible with personalization is so far beyond if that's what your personalization is today. And I'm a trust me, I'm lumping us into this.
[00:55:19] That is personal. Is that like, we're going further down. We're going to intense signals and we're trying to like get to messaging that resonates at an individual level or at least within a segment. Um, but we're not there like fully. So like what I'm telling you is marketing can be smarter. Now you have to go seek it out though.
[00:55:37] Like it's not just universally embedded into the CRM system you already use with automation or the email marketing platform. They don't doing this stuff.
[00:55:44] Mike Kaput: Yeah. I mean, I think what's interesting about these ideas as well is, you know, we're saying here, like, okay, some of these solutions probably don't exist and won't for some years, yet they should tomorrow.
[00:55:54] But the strategy behind them, you could start using now in a lot of these areas, I think. And I, I, you know, I wonder how many people. Listen to the lead score and intense signals you mapped out and you kind of think it's obvious, but I don't know. Have you ever gotten, like, I don't know. I haven't gotten to too many events last year, but like, I didn't can't remember a single email or like communication.
[00:56:19] I got that. Like, had that looked, anything like
[00:56:21] Paul Roetzer: that? Yeah. It's always the
[00:56:25] Mike Kaput: same event. I know it because I read it and delete it usually like, cause I'm not in like, I haven't been engaged with the brand or the
[00:56:33] Paul Roetzer: event for a year and maybe it's important real quick. Take a step back because the core to doing AOL is the data.
[00:56:39] So like we, we create, so we have 29,000 opt-in contacts in a good month. It's like five to five to 700 new contacts come in through webinars, through downloads. Our state of the industry reports through our intro to AI for marketers class, whatever it may be not paying. This is just like demand gen basically like people opt in.
[00:56:58] If, if we had been better over the years at collecting the right information, we could already be there. So like we ask in our state of the industry, what areas of marketing are you involved in the size of the company? What industry? And like, you know, the same stuff that you're used to filling out a form.
[00:57:15] What if that filling out, like when you, um, I'm trying to think, like a social media app I went into recently, I took screenshots as I was going where it's like, what are your interest areas? Oh, Ted. But go to Ted talk. Like they do this for Ted. You go up and say, what am I even interested in? What am I trying to do?
[00:57:30] My career, what areas? And then they're going to tailor it. The videos they recommend to you. So Ted is one that actually does this based on your profile that you shared with them. So to get to this end game of deep personalization, you need the data. Like you need to know what does, and we can't get it all from publicly available sources.
[00:57:49] I'm going to read their Twitter feed using IBM Watson and like, just get everything we need, but you can get some, but you need to get the, you have to have a strategy to collect the data and then how to how's it and how to like, process that correctly and keep it clean.
[00:58:02] Mike Kaput: Yep. Yeah. And as we, we kind of, you know, move forward with the event and, you know, potentially like new website and things like that.
[00:58:08] We're already starting to think about, like, that was the first thing we started talking about is how do you standardize the data we collect moving forward? And how is that going to play into AI? Um, potentially AI models and tools in the future. Um, you know, here's another one that's, that was really interesting based on our own outreach that we are doing is I think it'd be really interesting to use AI.
[00:58:32] To tell me, like, just look at LinkedIn profiles and tell me what are like, I could sit down and put in where I went to school. What age I am, what, um, uh, I don't know what my major was, what past jobs I've had. It could look at that in my profile. Tell me who has those in common if it like at scale? I think it does a little bit.
[00:58:53] It'll show you in your network, but like finding personal touch points with people is an instant way to start a really. Uh, authentic conversation,
[00:59:04] Paul Roetzer: right? I will say one idea I had, um, that I guess you're welcome to steal this, like all these other things I'm about to tell you. So if you're not a sales navigator, user of sale navigators, awesome.
[00:59:15] To get like really drill into segments and try and find people in like market size. So again, when I was talking about people with personalization, their title, digital in a title, uh, pricing, and we have all that data. So through sales navigator, we're able to go in and say, okay, how many people in companies with 200 or more employees in these seven industries have personalization in their title, have conversational in their title and then able to prioritize outreach?
[00:59:37] The other thing I just thought of though, is a backdoor way to use AI is go into sales navigator, build a lead list of people who have already been. So LinkedIn uses AI as a recommender system. It's like a matching system and it tries to identify people who look like other people in your lead list. So if you either upload or add on the most efficient way to do is probably upload a list.
[01:00:03] And then you save that as like a may con 22 customer list. And then you say, give me matches. It's going to find other people that look like those people have similar backgrounds, similar education, but then whatever, this, I don't know what the signals are. That's LinkedIn. Doesn't tell you what their signals are.
[01:00:22] You could guess, but they're using machine learning. I can almost guarantee you to find leads like the ones you think are valuable, but rather than me guessing, like I think people with personalization would come to the. Why don't I just take our list from the last two events, upload that. So just show me people like these people, I may be completely wrong.
[01:00:39] Maybe they aren't personalization and digital, and maybe, maybe it's somebody that I'm not even thinking of. And that's where AI's magic comes in, is finding the audience segments that you don't even know exist. And it might be sitting in your data and I could build all the signals I want all day long on lead scores.
[01:00:56] And I may not be thinking about the most obvious thing.
[01:00:59] Mike Kaput: Yep. Yeah. And I think that's where the segmentation, you get these almost endless possibilities because within that. Finding people you didn't even know you had. I think also you can do that from an interesting topic perspective where you say it would be awesome.
[01:01:13] One of the things I would love to see is AI that will tell me within a given industry, what are the top five most popular? Sub-segments that industry among whatever group. Um, that would be something I would look at when creating my conference agenda, right? Because you can say, Hey, you and I are interested in something, or we think this is really popular.
[01:01:34] And we kind of do that in a manual way today we build our agenda around. What we know to be popular in our content. So that's smart, really smart way to do it, but
[01:01:43] Paul Roetzer: like it takes a long time. And the same with online courses, like as we're building out academy and making more on demand courses available, we look at the content marketing data.
[01:01:52] So like market muse is a tool that Mike uses to build content strategy. So we can use that to look at demand for specific types of information, and then say, okay, let's parlay that into sessions that may con and, um, and what Mike and I tend to do is like, we think of all content in, in this like funnel or in a journey where it's like, okay, let's start with a pillar page.
[01:02:12] That pillar page does really well. Like AI for social media, I think is our most popular page last 30 days, if that does really well, then it should have a session at Mae con. And if that, you know, there's demand there, like it should have an on demand course, and maybe it should have a certification course, something bigger than that.
[01:02:28] And so we think about like all the ways to carry this through, and it starts off with something we publish on the blog, but we're thinking long-term about it. And we are using it. To inform the initial decisions that are made.
[01:02:41] Mike Kaput: Alright, I have one more. I want to throw at you and then we can, uh, we can move on.
[01:02:46] So we had talked actually the other day about thinking more creatively about our event and sponsors space physically at the conference center convention center and saying like, oh, could we actually do this? Could we do that? I feel like tomorrow there could be some app where you go up on the second floor balcony, you scan look at the floor space and it tells you, okay, flip there's some options.
[01:03:12] Here's how many booths you can put down. Here's how many, uh, like five by five, whatever, the small, smallest, medium, largest ones. Here's all the permutations of different types of sponsorships you could have. That would be amazing. But I feel like there's apps
[01:03:27] Paul Roetzer: that already do this. Yeah. And that's kind of a parallel path and someone has to be doing like, I'm going to be shocked if we say this and someone isn't doing this.
[01:03:34] So if you go to any major retailer, like online retailers for furniture, if you go to the apple store, if you go to home Depot that whole like using, uh, augmented reality to view the thing in the space, like you could totally do. With sponsors, like where you can view their booth. So like send us a picture of your 20 by 20 booth boom.
[01:03:54] There, there it is visualized in our exhibit hall and you can actually, like,
[01:03:58] Mike Kaput: I want to get, you want to get really futuristic? What if there were, this is a wild idea, but what if you, there was actually an augmented reality app that the event attendees had that when they scan the ceiling or the walls, other sponsor logos and QR codes and information, like you could sell space
[01:04:16] Paul Roetzer: everywhere, digital space.
[01:04:19] Metaverse by say, I wouldn't be
[01:04:21] Mike Kaput: for as much,
[01:04:21] Paul Roetzer: but I read this like,
[01:04:26] oh, damn,
[01:04:28] Mike Kaput: There's like a, I feel like there's a lot of, I hope some event, uh, event marketing people and planners listen to this or see some content related to it because the world it's just endless possibilities from what little I know already of the event marketing world.
[01:04:41] Paul Roetzer: Yeah. So I I'll re and if you have any other ones you want to touch on before we go here, but I'll kind of recap some of the big ones I thought about going in.
[01:04:48] And again, Mike and I didn't complete compare lists coming into this. It was just sort of throwing some things down. So I had NLP for social media shares from context on our list. So, um, let's say I have my high intent list and there's hundreds or thousands of peoples on that high intent list right now.
[01:05:05] All, all I would know is I can set up alerts. If someone comes back to the event site, that's on that list. So it's like, oh, Mike is on the register page right now. And I can get the alert and we can choose to be more human and not be creepy and send the, Hey, I see you're on the register page email that some companies will send and then they'll call you and text you.
[01:05:25] And like, that is not more human let's like, um, but what value based thing can I give? So if I know Mike's in content marketing, I know his title, I know his role. I know his loyalty signals that he's a, you know, a member of our academy and he's watched these feet three courses. Like I can know a lot about.
[01:05:44] That can lead to a very personal touch point with him saying, Hey Mike, I know you've taken these courses where like, what is it you're looking for? It may count. How can I help you? How can I make sure that it's going to be a fit for you? If you're going to spend the money and time to be here, how can I make sure it's a fit?
[01:05:59] I can do that from all my CRM data, if it's structured rec, but what I don't have visibility into is what is Mike saying on Twitter? What is he saying on LinkedIn? What is he saying on Facebook? If it's public like Reddit, like where else is Mike talking? That might be an intense signal. So maybe Mike shared on Twitter this morning that he's given a talk at content tech.
[01:06:20] If Mike has a high intent by. I should know that like I should be getting an alert that something is watching his social stuff and pulling from me something and thinks is important and summarizing it for me. And then giving me an alert saying, Hey, Mike's could be a content tech. Now I can reach out and say, Mike, I'm going to be a content.
[01:06:37] Tech would love to talk to you about getting to you to make con like, I think, I think you'd love it basically. So if I could enrich my CRM data, my, my first party data with some AI that pulls from what's going on online and then analyze it for me and summarize it in flags. What's important. Not just like here's his last 10 tweets.
[01:06:56] No, no. What I want to know, what's important, those 10 tweets related to what I'm trying to do with him. So that was one pricing is the other pricing is a guessing game is a pure guessing game. And with all the variables we're dealing with, it's even more of a guessing game. And so like, there's lots of ways to look at it, but the reality is humans suck at prices like that is absolutely an area where I just don't feel.
[01:07:18] Humans will play much of a role in pricing in the future. And certainly like airlines and stuff like that. They're already not like it's been seven years plus they've been using dynamic pricing. So Uber, like you think about all these sites that use dynamic pricing based on demand, based on whatever. Um, and we don't do any of that stuff, but we're not even thinking about dynamic pricing, personalization.
[01:07:38] We've harped on prediction. We've talked a lot about prediction. Conversational is a big one for me. So like our site just has a chat bot. It's powerful, it's a HubSpot chat, whatever HubSpot calls, their chat bot. Um, and it's just simple. Like, it's like, Hey, if you have any questions about the event, like it's personalized in that.
[01:07:55] It's we know you're on an event page and we're showing you the employees who work with the event. But beyond that, like we're not pulling in that you're on the intent list or you're on the loyalty list or anything like that, or that we know you like content marketing or that we know you took these to on-demand court.
[01:08:10] Like none of that's factored in, but in a true conversational AI agent, like a drift might do or a live person, something like that. You can get to that. Like you can, you can know that person in their journey and you can engage with them in real time. On the event site, it's more like an e-commerce thing.
[01:08:25] Like think about all the best e-commerce sites. That's what you are. You're an event like you are an e-commerce site and think about how you're getting blown away by the tech that other e-commerce companies are running. You're thinking about yourself as an event company, not an e-commerce company. Um, also related to the sessions, conversational AI that recommend sessions and speakers.
[01:08:43] So not only like, Hey Mike, I see you're here. Here's a discount code. Why don't you buy? It's like, Hey Mike, you might be really interested in these three speakers because we know he's interested in them because it's tagged on the back end that they're related to things we know he's interested in. And here's like four sessions we would recommend.
[01:08:59] We hope you come. Like, think about the difference between that chat experience. We were like, man, those do look like really good sessions. I think I'm going to come. And they feel like that agent was personal to you. So those were a couple. And then the other ones, I throw out the sales navigator stuff. So those are like big picture ones.
[01:09:18] I was, I was kind of hot on.
[01:09:20] Mike Kaput: Yeah. I feel like we've almost said too much at this point. I know. I,
[01:09:24] Paul Roetzer: I just, we just made our list like 10 times longer. These just want to get these. Um, well, I, yeah, I mean, that's probably sufficient. Maybe we're going to need a part two in six weeks. Like, we'll tell you a day.
[01:09:38] But, uh, I mean, I, I would be remiss if I didn't say like come to our conference, like if this stuff is interesting to you, these conversations need to be happening like this, but we just did these, this isn't happening. There's not enough of this. Like what's possible, what can we do now? How can we collectively be better as event marketers as marketers and as what we're trying to do with marketing I conference.
[01:10:00] So it's just www.maicon.ai - M A I C O N dot AI. Um, we'd love you to be there. I mean, it's, it is, it is challenging to create a movement and it's challenging to make people care. Um, and in our case, again, we believe it's so fun to write. To the success of your career and your ability to control the narrative.
[01:10:21] AI can have negative impacts. It can like, again, if you go back to listen to the last episode, there is a lot of negative. It's going to gradually take away parts of your job. If you're a writer, you're a designer, you're a creator of any kind. It's going to start infringing on things that you do that you believe to be truly uniquely human.
[01:10:42] Um, but if you understand that, now you can get out ahead of that and you can figure out what moves to make, what to do with your career to take advantage of it, to let it augment you rather than control you. So I don't know, we just, we believe immensely in what we're trying to do and the importance of it to the business world, to society, to our industry.
[01:11:02] Uh, and we would love for you to be a part of it. And if, again, if you have questions, you have thoughts on this session, like hit Mike and I up on LinkedIn or Twitter. There's again, there's so many. I mean, we could probably talk another hour, just like my mind is racing right now. We need to do. So, yeah, that would be my call to action.
[01:11:19] And if, if the conference isn't for you, that's cool. Like I get it just like be a part of the community. You get active in our community. Um, let your voice be heard and, and help us kind of move this industry forward. Anything from you, Mike, when you final,
[01:11:33] Mike Kaput: I couldn't have put it better myself now. Just go out and, uh, get involved in one way or another.
[01:11:38] Get started now couldn't be more.
[01:11:41] Paul Roetzer: All right. Well, we appreciate all of you listening. And, uh, if you are, uh, you know, on that loyalty list already, you've been an academy member or bend to make on. We appreciate you. Uh, if you're a sponsor that's made these first few years possible and, uh, help support the conference and the academy, everything else we're doing, we appreciate you.
[01:11:59] And if you're on that high intent list, I hope the personalization and the messaging resonates with you and we get you there. And I hope we all find some AI to help us do this all better. So I'm an event marketers. We love you. Hopefully this gives you a bunch of inspiration. So thanks again. We'll, we'll be back next week with another, hopefully a really intriguing topic as well.
[01:12:18] So there's been the marketing I show for Mike and myself. Thanks again for being with us. Thanks.
[01:12:24] Thanks for listening to The Marketing AI Show. If you like what you heard, you can subscribe on your favorite podcast app. And if you're ready to continue your learning head over to Marketing AI Institute.com. Be sure to subscribe to our weekly newsletter. Check out our free monthly webinars and explore dozens of online courses and professional certifications until next time, stay curious and explore AI.