Integrating AI in Your Marketing Efforts with Kirsty Grant

Erin Pearson: Hey, everyone. Welcome to today’s episode of our podcast, Decisions Now. I’m the host, Erin Pearson, and I’m very excited to have Kirsty Grant with me today. Kirsty is the International Marketing Director of Public Safety Solutions at Everbridge. Kirsty, welcome. 

Kirsty Grant: Morning! Thank you, Erin. 

EP: Could you just start by giving us a little bit of background on what your role is at Everbridge, and what Everbridge does before we get into some of the AI topics that we have for today? 

KG: Yeah, absolutely. So Everbridge is a company that probably most people won’t have heard of, but we are essentially a resilience organization that helps our clients respond to, well, prepare and respond to, major incidents. So, we work with large enterprise, we work with government authorities at state, county, and national level as well. And we help to really automate the processes and the communications that take place to respond and recover from a major incident. And my role within the company, as you indicated, is an international role where I support the marketing of our public safety solutions. So very much targeted at government entities. 

EP: That’s such a critical thing. And I think as we do bring in the AI topics, there’s a lot that AI is doing in this space to adapt very quickly. And especially when you’re doing broad-scale communications like that, AI has a huge level of applicability in it and what it’s able to do and the speed at which it’s able to communicate that. So, I guess that’s a great segue to just jump right in for today. 

So, one of the things that we were talking about previously is there’s obviously a huge surge in AI with ChatGPT and OpenAI, starting at the beginning of 2023. So, from your view when you’re actually looking at the B2B perspective, where did this come from and why do you think there was such a big need for it? 

KG: Well, I think it’s really interesting if you look sort of at the macro level about innovation in marketing and go back, sort of, I suppose, 30 years, really, because I’ve been around long enough to recall the days when marketing was a very analog environment. And, you know, we would literally be sending out mail shots to tens of thousands of individuals that would then fill out coupons and send them back to us. And we would do data entry. And then we came into the personalization era where everything was about the customer’s needs and personalizing communications. Then we had real sophistication around email programs, nurture, predictive analytics, et cetera. We’ve kind of always, as a function of business, been at the forefront of adopting technologies. And I think if you went into most enterprise organizations today, whether they’re on the consumer side or the B2B, and had a look at their tech stack, it would involve potentially hundreds of different solutions that are all working either together or independently to automate some of the facets of what marketing does. So, I think in the era of AI, which has really sort of come about in the last couple of years and is now accelerating at a real pace, it’s not surprising that the marketing function has probably got the most to gain from AI. And it’s probably a testing ground for much of the technologies that are out there now, including ChatGPT, as you mentioned, but that’s not the only one, clearly. Interestingly, I read a report recently, I can’t remember who it was, unfortunately, but these things come into your inbox, and you read them, but there’s a prediction that around 45% of total spend on AI will be attributed to marketing initiatives by the end of this decade. So, there’s almost half of that spend will be going into marketing activity, which is really revealing. 

EP: Yeah, well, I think marketing has always been in a cool position also, that a lot of marketing is rapid testing, whether it’s messaging or different go-to-market strategies. But it’s really a fail-fast type of mentality. So, I think when you do have something like all of these different AI-based tools that are coming out, right now, what we’re seeing is a lot of stuff that is likely going to be kind of table stakes use cases within a year or two that everyone’s going to be using. And we’re still yet to see some of the bigger advancements that are going to be coming from it. But I think marketing is in a really unique position to be able to test out these tools quite quickly because we just have that general mentality of testing out different SaaS-based platforms and just the level of experimentation that we tend to do to just figure out what works and what doesn’t as we are trying to figure out, you know, what’s the right messaging, what’s the right market, the audience, et cetera. 

KG: Yeah, totally. And I think that’s probably the most exciting — well, the way I see it, there are two areas where it will help the most and I think where I expect to see it being used the most. And the first is what you’re just describing. The other one is, of course, around content. But talking about how you design a campaign or how you run a campaign, you know, through test and learn, I think where AI will make a huge difference is speeding up that sort of closed-loop cycle of understanding what campaigns have worked, you know, for example, testing out more effective subject lines on emails. We know that AI is being used already for that. In fact, Everbridge is doing some initial pilot work on using AI chatbot-type solutions to formulate email messaging, and then test the effectiveness of those in an A-B environment to see which one is more effective. So, I think in the past, as you say, we all know that most of our spend is wasted, we just don’t know which bit works necessarily always. So it is test and learn. So hopefully, over time, as this matures, we’ll see the emergence of tools that will help us aggregate all of this data from consumer behavior, from email performance and give us real, valuable – possibly in real-time as well – feedback on what’s working so that you can adjust campaigns on the fly, which would be amazing. So, there’s a lot that we haven’t seen yet or tried, certainly in our business, but I think in the wider sort of marketing community, there’s still a lot of sort of testing going on out there. But it’s definitely an exciting time, isn’t it? I mean, it comes with a lot of caveats as well, which I’m sure we can talk about as well. 

EP: Yeah, no, absolutely. So, I think you also nailed it. Generally speaking – this is true in marketing, but it is true in every other department as well – expectations are generally rising, but costs tend to be cut. And so, AI is really there to help fill some of those voids as well. They are essentially asked, to steal Microsoft’s term, they are acting as a co-pilot to people individually, so you can be much more productive, but they’re not really there to replace you, either. But as you mentioned, AI is not a silver bullet, and there’s still a lot of concerns. So, I think there’s some that you had been talking about previously, like the level of like accuracy and reliability. Do you want to talk through a little bit what you guys are thinking through there? 

KG: Yeah, I think as with any new approach to doing things, there’s always concerns about, particularly in the environment where I’m operating, which is public safety communications, our clients are governments, you know, we’re dealing with very important, critical messaging at an important time when lives might be at stake. So, you can’t afford for any ambiguity or misinterpretation of the information, it has to be trustworthy. One of the initiatives that we’re running with an EU project called Horizon is to look at how AI can be used in a crisis situation to provide information back to the public. So, you effectively avoid a flood of calls coming into your 911 or 112 services (as they are in Europe) by providing that sort of knowledgeable, trustworthy interaction with the public with reliable information. So, machine learning, aggregating what are the official statements that are out there, and presenting them back in a human form to people who are looking for guidance on what to do in an emergency. But that’s really early stages, and that’s something that we’re piloting, we’re not sort of actively providing a solution like that. But I think all of this is about saving time as well as money, and if AI can speed up the communication that’s either going out because you’re trying to sell something to somebody, or you’re trying to save a life, then I think that’s a really valid use of the technology. I think where it becomes ambiguous and worrisome is validating the source of the information. And we’re now seeing regulatory frameworks emerging. The EU is working on an AI framework for member states in Europe. Because it’s needed, you know, with any new technology that has the potential for greatness, also has the potential for, you know, unfortunately, dark methods, which we’ve seen in other things over the lifetime of humanity, I think. So, yeah, I think, you know, it’s not for me to say, you know, how that will be regulated, but there’s certainly a lot of discussion now about the need for regulation. 

EP: I think the example that you gave is actually quite pertinent as well with the emergency services, but certainly – I’m based in the US, and one of the big things that comes out, in particular around any political election, is what’s true and what’s false in terms of the news that’s coming out. It certainly goes beyond politics as well. So being able to, as you put it, also being able to find what are those reliable sources and to really validate where this information is coming from, is even more essential. But I think the, as they call it, the age of AI has really started accelerating substantially this past year, which could lead to a lot more potential misinformation, but there’s also a greater ability to potentially find and correct that type of information faster as well, so long as it is used correctly. 

KG: Yeah, I agree. And I know that we’ve seen companies, large companies, already implement restricted use of AI tools, specifically ChatGPT, but others and country-level, some of the more conservative countries have actually banned ChatGPT. No surprises, obviously, countries like China, Russia, but Italy, you know, Italy and Europe coming out and saying that they don’t want that type of technology in their country. So, there’s a lot of fear out there, but the genie’s out of the bottle, which is one of the phrases my colleague used. We can’t put it back in. It’s like the internet. The internet started with very humble intentions and has now become part of daily life and has multiple applications for good and bad. It’ll be similar. So, I think the lessons we’ve learned from the internet should be applied to AI. So that, yeah, we move forward safely and for good and not for ill. But that’s easy to do – sorry, easy to say, not so easy to do. But I think within a marketing community, we have a responsibility for that as well, to, you know, pilot test responsibly with good intentions and show that, you know, the goodness and the potential of the tools rather than, you know, focusing too much on the bad stuff, I guess. 

EP: Yeah. 

KG: But I think the other thing is, we haven’t touched on, but privacy is the other issue, isn’t it? Because if we’re releasing a technology that’s able to basically crawl the internet and search for all kinds of information and very quickly turn it up into something else, then, you know, rather like Facebook using your imagery without your permission or you’re giving your rights away to your imagery, where will that end and how do you regulate for that? So, there are many, many discussions to be had about that, I’m sure. 

EP: Certainly. Yeah, and I think that as you do have the generative AI tools that are really the consumer-facing versions of them, which is why there’s been such a huge acceleration, but AI has been very prevalent in everybody’s lives. If you order anything from Amazon or any shops or your phones listen to you when you’re talking and then you get personalized ads. So, it really is everywhere. And it’s already been impacting you in like, forming your decisions, whether or not you know about it, for the past few years. So, I think that this conversation is almost overdue as well, but the consumers weren’t ready to have this conversation because it wasn’t as consumer-facing and accessible. So, because we’ve actually created now this consumer-facing product that is so easy to gravitate towards, it’s so easy to use, and people are like… ‘Oh, wait a second.’ Like, there’s a lot of other applications around this and now you start to realize how much it’s impacting you in other areas. So, I think there’s actually a lot of good that can come from having all of this there as well. 

KG: I agree with that. And I was thinking about it in a different way as well, because exactly what you say – in the last few years, and maybe even longer, actually, we’ve all been in our consumer lives subject to retargeting ads, you know, popping up all over our desktops on our phones, trying to sell us things that we might have been Googling for not so long ago. And if like me, I’m sure there’s thousands of people listening if they are that have got a speaker, an Alexa or equivalent, you turn that facility off because you don’t want them listening to your conversations and then presenting you with ads for things when you’re on the computer. So I think we’re becoming a little bit more savvy about the techniques that are out there for AI in terms of retargeting for promotional ads and trying to sell you or upsell you. But I think Amazon obviously we’re at the forefront of this with the, you know, people who bought this also bought that. And guess what? Yes, I’ll put that in my basket as well. Why not? What a great idea. So yeah, we have been, it’s happened by stealth, hasn’t it? But as you say, I think ChatGPT has now opened everybody’s eyes to actually, oh, this is what’s all going on in the background. And maybe it will become harder to sell because consumers will become more aware. and perhaps more resistant to being – not duped, that’s the wrong word, but you get what I mean. They’ll maybe start to question the content they’re being served and the intention behind it as being not human-related, but just being manipulated through AI. I think that would be quite interesting because certainly in the consumer space, there might be a degree of digital switch-off from people. And that kind of brings me back to some of the discussions we’ve had in the past about how we arrived at Evalueserve. Because as a B2B company and a B2G company selling to government organizations, we took a good look at how we were approaching our marketing campaigns. And actually, there is so much noise and there’s so much email and there’s so much automation around nurturing, targeting people through social media, that we chose to take a very direct approach, which is how the relationship with Evalueserve began. So, I’m not gonna go into too much details because we don’t wanna share our secrets too much, but it has worked. But I think it’s a balance, isn’t it? It’s about bringing lots of different tools to play in your marketing strategy and not just relying on one tactic. So, there is definitely a place for AI in what we do. And as a company, as I’ve mentioned with our product side, we’re investing in piloting things like a chatbot, but also, we’re using it within Everbridge Marketing. So, we’re working with tools that are helping us to craft email drafts, for example, because that’s saving time. We’re using them to do social listening so that we can figure out, you know, what are the most effective keywords that we might be wanting to use or base our discussions around on social media. So, we’re just dipping our toe in. Our chief executive even sent a company-wide communication, which he used ChatGPT to write. So, to kind of involve everybody in the company in this new era if you like. So, it’s happening within Everbridge, but we are dipping our toe, we’re not jumping in. And I think that aligns to what the advice seems to be from other organizations like Harvard Business Review and others saying, you know, crawl, walk, run approach, not try and do everything. So, yeah, I’m optimistic, but I think I’m also cautious, from both the consumer side and the professional side. 

EP: I mean, that makes perfect sense. I think it’s quite interesting and quite smart to do the crawl, walk, run approach, certainly. Within the marketing team and how you guys are using generative AI, I think you’ve also talked a lot about, and certainly one of the big conversations, is how easy it is to create content, for example, especially within the marketing side. So, whether it’s, I mean, content’s really the foundation to everything, right? You need content to create emails, social posts, eBooks, white papers, demo presentations, like anything that you share out, the root of it is content. Generative AI, it’s not really creating things from scratch, but instead, it is aggregating a lot of thoughts that are already out there. So, one of the things that I actually think that could be quite cool about generative AI is it helps to eliminate some of these buzzwords and it actually puts people with more original thoughts and content actually in a much stronger position than it did before, because if everyone can have that standard content, then it’s no longer that useful. Or you can just ask it yourself instead of having to search it online. But why theoretically somebody might start researching online is to get that more original content. So, what are you seeing with Everbridge and the approach that you guys are taking to the multi-channel approach and the more custom content creation that you’re doing? 

KG: Yeah, I think I can answer that in two ways. I think from the Everbridge perspective, we’re really at the start of the journey and we’re not deeply into it yet, using AI to create content for marketing material beyond what I’ve described, which is the crafting a draft email for a nurture program and just testing it out that way. But I think… If you ask for my opinion, I think there’s a couple of things I’ve seen around the topic which really struck a chord with me. And I will always advocate for plain speak and not tech speak. And working in a tech company, it’s very easy for us to get drawn into non-human-like language in the way that we produce content, whether it’s print or audio. And ironically, I was looking at a blog post. There’s a really great website, if you’re interested, called AIauthority.com. And their premise was that humans have begun to write like machines and use language which is very non-human because we’re just churning out what we think will resonate with search engines and Google Analytics and get us rankings and, you know, send people to our website. But we’ve lost that human way of communicating and, ironically, ChatGPT is actually talking more like a person. So, it’s a bit of a wake-up call for those of us that are in content creation to remember that what we need to do in our content is convey ideas and thoughts and value and not just churn out, you know, buzz phrases like you say. So, I think that’s a really positive outcome from the adoption of AI in marketing. And the other thing, the second point is that I think, there’s a great Gartner Report as well, which is kind of a guide to CMOs to sort of how to approach AI and what to look for both from the campaign side, from recruitment, what sort of talent to look for in the future. We can come on to that as well in the discussion. But one of the points they make is that it’s a great tool for ideation. So, I love the word ideation. It’s like imagineering. I love that word as well. But, you know, using ChatGPT to create original thought, to kind of challenge your own thinking, either individually as a group, to see what else might be out there. So, if you ask for something, you know, on a topic, and what you come back with from ChatGPT could be a totally different thought that you hadn’t had as a group that might change the way that you go. So, I think there’s benefits of using it in practice but also as a kind of staging tool to create ideas or to sanity check things that you might want to do. So yeah, we’re not really using it deeply in Everbridge, but I think it has got massive potential. 

EP: Yeah, well, one of the fun things that I found to do with it is you ask it to, you know, tell you a little bit about a new topic that you don’t necessarily know as much about, or you’re just trying to learn more. And you can essentially ask it to teach it to you like the 80-20 principle and tell it to focus on the 20% that goes a lot deeper, but most people don’t actually know very much about. So you can effectively say, what are things within this topic that are more in like the white space or emerging area? You can use it as that initial research tool to really, whether it’s, you know, you’re starting in your career and you just are trying to use it to get up to speed about, you know, what on earth my company actually does and what is the space and category all about, versus you’ve been in it for 20 years and you’re just trying to say, you know, what have I forgotten sometimes about the fundamentals? Sometimes when you’re doing something for such a long time, you build up a lot of complexity in your head and it’s hard to distill it down into simple words. And sometimes this is a lot easier to say, here’s my brain dump of thoughts. Can you highlight what my key points are? And it’ll just help you kind of refine your notes and put your thoughts together. 

KG: Completely. Yeah, I totally agree. And those of us who’ve been around a while probably can count the hundreds of hours or Mondays that we’ve spent revising content drafts and circulating for changes and redrafting and exactly that, you know. And then we become so close to it that we lose sight of it. So yeah, it’s another great example of how it could be used. I think whilst we’re talking about fun things, and I mentioned this to you before, but… I have a great colleague in Everbridge who’s a real pioneer on the, he loves tech, he uses all the apps going and he challenged ChatGPT to describe what Everbridge does in pirate language, which was absolutely hilarious, and he shared it on his social media feed. And it was very accurate, it was good, but it just, it was as if it was being spoken by Captain Jack Sparrow so it was fantastic. So, you can have some fun with it as well. 

EP: Yes, that’s so much fun. I think it also, it’s a good take on, is the message that you want to be putting out there, is it being interpreted? If ChatGPT can spit it back, whether it’s in pirate language or a different one, maybe Pig Latin or something, is it actually coming back in an accurate way and how you want to be perceived so you can get a sense of, is what you’re attempting to do actually working? 

KG: Yeah, exactly. I know you touched on recruitment as well. I don’t know if we want to talk about that now. But I think there’s a certain amount of trepidation amongst marketing that this will put us out of work or, you know, what kind of skills would we need to be training the next generation for in marketing. But I don’t feel that, actually. I think, as you describe, these tools are really only reusing information that’s already out there. So, the responsibility of marketing or content owners is to ensure that good information goes out so that good information comes back. And the skillset will be different for sure. We’re going to start to look for people to work in companies that understand AI, that are familiar with all of the ways that we could use the technology and represent that within the company. So very different types of roles I think we’ll be recruiting for in the future, probably not in the near future. And that will ripple down into the kinds of educational courses that are offered in higher ed. So, you know, there’s some big changes coming. I mean, you know, these are all at the heart of a new technology. It kind of starts as the spear, doesn’t it? And everything else comes out underneath. 

EP: Yeah, for sure. I think it’ll be quite interesting. I mean, you already see job postings for, like, prompt engineers emerging as well. So, imagine what’s going to happen in like, you know, 10 to 15 years when the people who are using this in middle and high school, that is just going to be so well integrated into like how they just work in general, how different it’s going to be. It’s a very similar… 

KG: Well, I know every parent in the world is worried about how their child is going to use ChatGPT to pass exams, so we won’t get into that rabbit hole, because that’s a huge discussion. 

EP: Yeah, it’ll be quite interesting to see how it emerges, the difference in jobs and responsibilities, but I think there’s actually a lot of really great potential and a lot of ability to actually take this and really improve the content. As you said, generative AI is really good at sounding like an algorithm, or creating an algorithm based on essentially what the algorithm that humans have created just by trying to appeal to search engines. So, there’s a lot of potential to actually have more creativity in what people say and to have that skillset be even higher value than it would have been a few years ago. 

KG: Yeah, I agree. It’s interesting, we’re in 2023. And I think certainly when I started out in my career, I kind of assumed that automation would make us less busy at work. But it hasn’t happened, has it? I think we’re all still really, really busy. So, I’m busier, and we’re probably working longer hours. And yeah, so I don’t see it as a time-saving technology or a people-replacing technology, but it can definitely improve the quality of what goes out and the results of what our campaigns are delivering. But yeah, it’s not going to replace people. I think where it will be important is it just aligns to what we’re already doing and supports what we’re already doing. And ultimately, humans will still need to interpret what’s coming back and make decisions based on that. So, I don’t see that it will be ever used to completely run a marketing function. I don’t think that would be possible. 

EP: Yeah. 

KG: I know that’s at the extreme, but I don’t think it would be possible. 

EP: Yeah, well, I think that’s actually a great ending thought and a pretty clear summary, actually, of what we talked about for today. So, Kirsty, I’ll just say thank you for joining us today. And it was great having you, and I think you had some great perspectives. And I think what Everbridge is doing and how you guys are doing the crawl, walk, run, approach is quite cool with generative AI and how you’re getting your different teams involved in it and how different people are experimenting in some fun and clever ways. 

KG: Thank you, Erin, and thanks for the opportunity. I mean, as I say, I’m by no means an expert, but everyone’s involved now, so we all have an opinion. And I hope that was interesting. Thank you. 

EP: Awesome. Thank you.