The AI Marketer's Playbook
The AI Marketer's Playbook is an actionable podcast focusing on AI and marketing. Each episode covers AI strategies, tools, and trends that are changing marketing. Listen to interviews with industry experts, analyze case studies, and get practical tips. This podcast is for anyone looking to leverage AI in marketing to improve results.
Tune in to stay updated on AI marketing.
The AI Marketer's Playbook
32 | Nicole Leffer on Scaling Marketing with Generative AI
Nicole Leffer, seasoned marketing leader and AI consultant, joins Audrey Chia on the AI Marketers Playbook to unpack how generative AI is revolutionizing marketing workflows. From early adoption in 2021 to advanced automations today, Leffer shares her journey, favorite tools, and how teams—from lean startups to Fortune 500s—can use AI to scale content, drive efficiency, and boost performance. She dives into real-world use cases, how to avoid common pitfalls like hallucinations, and why focusing on just one AI tool can transform your strategy. Tune in for practical tips and a peek into the future of AI agents in B2B.
Join my weekly Newsletter: https://lp.closewithcopy.co/welcome
Hello and welcome back to the AI Marketers Playbook, where we cover actionable frameworks to help you leverage AI and marketing strategies in your business. I am Audrey Chi, your host, and today I have with me Nicole er. Now Nicole is a seasoned marketing leader and an expert in deploying gen AI to enhance marketing efficiency at scale With over three plus years of experience using AI tools. Nicole has worked across multiple marketing domains, including content marketing, growth marketing, D, demand generation, conversion rate optimization, A BM, and more. I am super excited to have you on the show. Welcome to the show, Nicole.
Nicole Leffer:Thank you so much for having me. I'm super excited to be here today.
Audrey Chia:Tell us more about your journey, right? How do you venture into the world of Gen AI and how do you get started?
Nicole Leffer:So I fell into this a little bit. I was the head of marketing at a business to business SaaS company. And I needed some tools to help improve the writing that was just like the same stuff. Every marketing team and small business. It was a series a startup. every company kind of deals with like, how do we produce more high quality content, keep up with the pace of what we need to get done? And this was in 2021, so this was several years ago before anybody was talking about ai. And I kind of was just doing some. Research on the internet, trying to figure out like what are some options? And I stumbled across a generative AI writing tool and did a demo with it and I was like, this is so cool. We have to bring this on. And so brought it on, started getting my team using it. Every day for everything that they were doing. I was using it every day for so many different things. I just thought it was super, super cool. And this was of course, pre-chat DPT or any like way before that stuff came out. but even then, it was making a huge difference to what I was doing, what my team was doing, and I just fell in love with the technology. And was about a year of using it with the team, went separate ways from that company and started digging in during the time of going what's next to all this other technology in Gen AI that was out there and discovering more, getting deeper, you know, it was progressing quickly. And because I was in that world, I was on open AI's email list and got an email about chat GPT the day it came out. Got on there very, very quickly and was sharing all of this in marketing communities I'm a part of now. I had been sharing it and it was all crickets. nobody, you know, it's, what is this crazy lady talking about? Nobody had any idea. And then other, they started hearing about chat GPT elsewhere, and suddenly people were like, Hey, wait, what, what were you talking? I think she knows something about this. So, the, it was a CMO community I'm a part of, and they started asking me, would you come in and trade? My marketing team had to do this stuff. And so it just kind of took on a life of its own. And I, it's really just. I'm a little obsessed with the technology. I have fun with it. I love it. I find it really fun to figure out what you can do with it and how far you can push it. And I also tend to be good at communicating that to other people so, and how to do it to other people. And so it's created an interesting career for me.
Audrey Chia:Wow. So can I ask, right, since you've really experimented with Gene AI for a long time.
Nicole Leffer:Mm-hmm.
Audrey Chia:What has changed? I know Gene AI is evolving at such a massive. You know, speed. Right. But at the same time, how has, you know, people been receptive to it? Have they been receptive to it? Are they currently, more open to it? What have you seen?
Nicole Leffer:You know, it's a really interesting mix, right? So like, at first, obviously when, when it comes to the people, they had no idea what you were talking about. Heck, I didn't know what I was talking about at first. You know, years ago, so like at first it was just like you did this crazy magic trick for people and I think it is still that for a lot of people, like we don't realize how few people have actually, I. Really been spending time in this technology, especially if you're in a bubble where you're using it all day every day. You forget that there's tons of people who have like never opened Chad GPT and never had an experience, or they've like used it for very minimal stuff. so. There's kind of been, I think, maybe a little less, oh my gosh. Wow. Magic trick than there used to be. but it's a wide range with people. You kind of, you see some people that have expectations. They're, they're really like almost too high in their expectations of where they think the technology is. They think it's further along than it really is. It's more capable than it really is, and that it literally is just a magic wand and you don't have to do anything, and it will just. Read your mind and do what you want on command. And then on the flip side, you still have the people who think that we are where chat GPT was when it first, like the week it came out. Right. Maybe like don't even realize it's that far along. And the people who don't really realize where it is, I think are more resistant to use it. Interesting. The. Biggest thing I think I see is that the overwhelm has dramatically increased. Like before, it was not so, so complicated. It has gotten to a point of complexity that I think that what I'm seeing with marketers, founders, business people, is that it's just so much Yes. And they have no idea what they're supposed to listen to, what to look at, all of that. So there's a lot of overwhelm right now.
Audrey Chia:I have also seen that myself. Right. I think people are perhaps a little bit more open and receptive to adopting ai, but yet now there are so many different tools, use cases, where do I begin? And then as an organization, you have to figure out whether it's like a top down approach or a bottom up approach. And since you have worked with, you know, many different types of organizations, right? From the startups to your Fortune 500 companies, what do you think is the best approach for a company to take, given that AI has so many functions?
Nicole Leffer:You know, it really depends on the company. So what works in like a series a startup versus what works in a public, you know? Company that's traded on the stock market is not gonna be the same thing. Because when you think about it, you know, some of these small startups maybe have a marketing team. 1, 2, 3 people, four people. Where I've worked with companies where there are literally hundreds of people on the marketing team and what they need to get out of AI is not gonna be the same thing. And so in a smaller company, I think you've got to focus a lot more of like every single thing that every person can do with it. Yeah. Because like the main goal of that tool is gonna be, you want the quality of course, out of it. Like you want really high quality results, but you really have got to get that productivity because, and you've really got in a. Small, small environment where people wear a lot of different hats and roles, the AI can take on this ability to become a lot of different marketers than you would've already had. Now, you of course, need a human as part of that, but the AI can be leveraged to have a whole huge, massive skillset. Where in a really big company you probably have all of these skill sets. And so like the bigger thing that you need to have everybody learn is how do they within their specific existing skillset really perfect using this technology? And those really are very kind of, in certain ways, they're similar. Things that you have to do, but in other ways they're very different. And so I see a wide range and you know, in a really huge company, there does have to be a little bit more of a top down, only because you've got processes and bureaucracy by, I mean, important bureaucracy, right? But they have to do a lot of things to protect their company and their brand and checkpoints and things like that. We're in a small company, you can really have your team take the technology, learn how to adopt it, and run and grow, and figure these things out. So cultures matter, size matters, and I have worked with big companies that have. A closer where they like people can run a little bit more. But again, it just really depends on the company of how, how to bring it in and adopt it.
Audrey Chia:I think it's very interesting to also realize that every company will have its own like unique tech stack AI stack, and of course, you know, a mix of people and how they're leveraging it. Right. but perhaps you could also share with us. What are some like super actionable use cases for an enterprise level company versus a startup? Like what does it look like for them? How is it different? Yeah. A lot of people haven't even tried using AI and founders who have I have spoken to who want to use ai, they come in from such a. Macro level, but it's really hard to get things going because they don't know where to start. So what does it look like for both types of companies?
Nicole Leffer:So it's, in certain ways it's a bit different, and in other ways it's exactly the same. So I'm gonna focus in on some of the things. It's just more who's using it where. When you're at like a big company versus a small company, the same person might be doing all the things of a small company, where at a big company you have like one person focused in on very specific use cases. but some of the most popular things I see use case wise for marketing are the biggest one by far across company sizes is repurposing content. And so what I mean by that now I mostly work with business to business companies, so they kind of tend to have similar types of content, but maybe they have a ebook or a white paper, or even like a webinar transcript that they've done, an educational webinar. They have all this like long form, really intensely created content that they've had maybe for a good amount of time, and they need to take that and it's. Always been something that's very hard to take that existing piece and make lots of different content based on it, where you can use AI to help you go, okay, you have this like webinar transcript. Maybe there's the content built of our ideas within that for 10 different unique blog posts. All completely different because in that hour you were talking about lots of different topics. How do you have the AI identify those topics? Get you the outline, the first draft, all of that, that the person can work off of. And it happens really quickly. It's still your company's ideas, it's still your company's thought leadership and that's getting spun into content. I would say that repurposing use case has from the big, from, you know, the over two years I've been consulting teams on this I to today like that is. Truly been the number one requested thing that everybody wants to understand how to do. so that would be number one. The other then, then from there we kind of spread out a little bit more in what people are really into a lot of, again, with content, things like adaptation. So maybe you have, whether it's. Copy on your landing page, or it could be emails or marketing copy, whatever, different kinds of copy you have, adapting that for different geographies, different verticals, different personas, so you now have a lot more custom content depending on who you're talking to. AI can help do that very, very quickly. conversion rate optimization. So that's a bit different than like creating the content, but optimizing your content, giving the AI a screenshot of your landing page and having it provide you recommendations to improve that page so more people actually buy from you. Helping create email cadence strategies. So if you have a lead nurture or a BDR campaign, the strategies for, and then ultimately writing the emails is another really big one. There's so, there's so many different things. I mean, the, the sky is the limit at this point of what the technology is capable of, but those are some of the most popular that I'm seeing.
Audrey Chia:Definitely. And I think like what you say, right, for a B2B company, your comms is your vehicle to scale. And that is so, such an important part. I think one interesting point that you pointed out was the, ability to hyper personalize or basically speak to different target groups, right? And I think as marketers, we already know that you need to speak to that audience, but sometimes you don't have the time to create so much content for 10 different personas. And I think that's where AI comes into play, right?
Nicole Leffer:Yeah, it's so cool. And it can also help you like narrow down those personas, right? So you got kinda two different pieces to that hyper-personalization part. One is you have to actually know who you're personalizing to and a lot of companies may have like their, they've done the work, especially smaller businesses, like maybe you have that one key target persona who's the number one. But in all reality. Let's say you're a marketer and you're selling to other marketers, right? So like you're speaking and you're like my, my target persona is the chief marketing officer. And so you've created this canvas of a chief marketing officer and you've gone through, but in all reality, CMOs could come at enterprises, they could come at small businesses, they could come at mid-market. Maybe you serve all of them. Maybe you've narrowed that down. Let's say you only do. SaaS startup, you know, B2B startup, and you have your persona there. You could actually understand that persona for health tech, MarTech, like all of the different, those are gonna be a slightly different persona, and so the AI can actually help you go in and research and refine that persona. So each of these different verticals you wanna talk to or you know, specific segments of your persona. You have individualized personas for them and you can get a lot more specific and narrow. And then the other piece, so that's the research piece, the understanding who you're talking to piece. And then the other side of that is actually creating the content that is speaking to each of those specific people. And it's just so powerful to be able to do that so much faster.
Audrey Chia:I think that is like marketing right? Is one of the most powerful use cases for gen AI at this point in time. Like with, the AI tools and the effortlessness that you get to create content. Yeah. You know, it's really beautiful because now like teams are really increasing in, their productivity levels as well as saving time and money. Right. Have you seen the actionable kind of like, or tangible results from any of the clients that you've worked with so far?
Nicole Leffer:I mean, every, all the clients I work with are getting very tangible results. they, you know, the biggest of course is time savings, but also like the metrics are just better, right? So like the reality is we talk a lot about how much time and efficiency you say, what we don't talk as much about is that it's also. In the same amount of time or less time, you can create higher performing assets and like higher quality outputs. And so. One of the big things that like they are seeing is that the quality does increase, which means the performance tends to increase as well.
Audrey Chia:Yeah, definitely. Maybe one other interesting thought is like, since you are talking about, you know, the benefits of ai, but I think there are also perhaps potential pitfalls that people may not realize. yeah. So what are some like key considerations that you also share with teams when they're thinking about adopting and implementing AI strategies?
Nicole Leffer:Yeah, I'm so glad you bring that up because there are a lot of pitfalls that you have to be careful of. So one of the biggest, and I do think people do talk about it to a degree, but it still does not seem to have like fully sunk in with a lot of people. Is the idea of hallucination so that the AI can make things up. it really is a real issue with these tools that they can invent things out of thin air. And when you're talking about marketing, you have to be very careful that the AI does not make up things about your product, about your industry, things like that, right? So certain things, if it made it up, oh, well, who cares? Like, it's not gonna really matter. Like if it makes up something in your persona. It's probably gonna have most of that persona, right? I'm not gonna say it's, oh, well, who cares? But it's, it's not gonna be catastrophic to your company unless the thing that it made up like cost you a lot of money. So, but you know, if it's just like they have, you know, this person is like driven by X, Y, z, internal motivator. Yeah. And like. Really specific level, it could affect you, but at the end of the day, it's probably not gonna make this huge of a difference. However, if you have it, say my product has, you know, like it's writing content for you, and it says. This product can do X, Y, Z and that is completely made up out of thin air. And you add, you post that because you didn't review it.
Audrey Chia:Yeah.
Nicole Leffer:You could have some serious issues, right? So you have to be careful when you are creating the content that you fact check everything, including when it's working off of an asset that you gave it. So you could give it a document with all your product spec. You still need to check that it didn't make up anything about your product. that's the number one thing. Also, depending on where in the world you serve as far as your customers, you do need to make sure you're following the laws based, you know, depending on where you live, where you are working from, but also where your customers are that are gonna see it. what are the legal considerations around using this? and then just like brand credibility, right? Like, ah. I'm not one, like I think it's totally awesome and amazing to use AI in your content creation, but if your content, everything you write, starts with in the rapidly evolving blah, blah, blah landscape, or in the era of X, Y, Z, the things that it's like, oh my gosh, every single AI tool is writing this about everything. Like you're gonna look stupid, right? So you've got to be aware that you are editing out the obvious AI stuff that is gonna make it seem like you didn't even bother to review what it was putting out. I don't think the issue is using ai. I think the issue is. Not editing the AI to bring your own unique brand voice into it and to ensure that it's not sounding like a robot wrote it and because it's just, even if in the rapidly evolving landscape of whatever sounded good five years ago, we were all so freaking sick of learn like hearing it, right? It's the issue isn't that it's such bad writing. The issue is that it is being used so constantly. It started out as probably what was considered good human writing.
Audrey Chia:Yeah.
Nicole Leffer:It's just that like, good lord, if I read that one more time, I'm gonna like smack somebody for like, you know, like myself. I can't deal with it anymore because we're so sick of seeing the same phrasing everywhere. And so if you're seeing this phrasing, this language everywhere, you should take it out of your ai. Outta your AI writing. So
Audrey Chia:definitely, I, I think one interesting thing is like, the more you use ai, the more you start recognizing these patterns, words, or trends. So you would realize that people who don't use AI as often, they just take it wholesale and they post it, you know, as content, because to them it sounds perfectly fine. yeah. And it, it could even sound like, you know, pretty like. Pretty, pretty classy, you know, to some extent. But then because you use AI so often you, you start realizing these phrases that AI likes to use. and I think more and more people are aware of that because there is also an increase in adoption. using chat GT specifically chat GT's, AI voice. Yes,
Nicole Leffer:absolutely.
Audrey Chia:Have you seen, perhaps certain tools perform better when it comes to writing compared to chat GPT or is it still your favorite? I.
Nicole Leffer:I'm like definitely a chat. DPT girly myself for the writing quality. I think that the newer model, like they've gotten much, much better over time as they've approved the models and so that's important. But I also have really good custom instructions. I'm prompt, so like I am always like really surprised at what comes out of chat GPT when I see. See it without custom instructions.'cause I'm used to, what it produces for me has my custom instructions and I use my custom instructions largely around like the writing style that it puts out. And it's like, even though I work with companies like logically, I know how big of a difference it makes. It always still shocks me when I see the version without custom instructions. And then I realize I'm like, that's why people complain about Chad g PTs writing because they don't have those instructions in there to make it better. but personally I think that with some really good custom instructions, Chad GPT is incredible at writing, especially as these. Models get better, but I do know a lot of people really love the Claude Sonnet writing, and so I think it is personal prep for, for me personally, I it just like, I cannot get Claude to perform at the same level as chat DPT. But again, I think it comes down to the custom instructions I have in my chat, DPT backend and in my automations and my prompting, it is just chat. Like I've got it down. To a T. So I think it really is a bit of personal preference. That said, I do, I mean, I still run, like I, I have automations I've built, which is obviously a little bit more advanced and like my automations automatically run to give me like a chat, a GPT-4 oh or 4.5 version, and then a Claude Sonnet version. So I get. The different choices. And so I can kind of watch and see how they evolve. And the newest CLA is good, like the newest CLA is definitely improving. Significantly
Audrey Chia:powerful.
Nicole Leffer:Yeah.
Audrey Chia:Well, I, I would love to also double tap, double click on what you said about automations, right? So I think there are different levels of. AI usage, firstly, using the LLM or the chat GPT itself. And then secondly, figuring out how to bring automation into the picture. So what are some like, use cases for yourself or your clients in which you integrate automation and AI to then build a workflow?
Nicole Leffer:So I think automation, like the first thing is like I use Chad GPT as Chad, GPT all the time, so I don't want it to seem like this isn't either or. but automation is really good if you have to do the same thing with like slightly different inputs. On a regular basis. So first thing I do like with automation is go, is this something like genuinely repetitive enough to take the time to build an automation? Because it's not worth taking the time to refine an automation if you're not gonna do this on a pretty frequent basis, or just like a ton of times in a row, in a short proximity. So my number one most used automation is to write my LinkedIn post. And the way that it is built, I actually have a, it's a combination of chat, GPT and a Zapier automation. I have a custom GPT that I built in chat GPT. That is literally like the entire purpose of it. It has nothing to do with a workflow or working with it. All I do is I use, on my app, I grab my phone and I open the chat GPT app and I open that GPT and I go into where it like you can record your voice and it transcribes it and I just talk to it. I basically tell it a thought dump, verbal version of what I want that. That LinkedIn posts to me about, and then I submit it, and when I submit it on the chat GPT app that sends it to Zapier. Now I'm done with chat GPT that sends it to my Zapier automation, and that automation takes it through a series of prompts that it writes the post it double checks to make sure it included everything I wanted. It strips out the things that. It made up, so it's like fact checking itself to make sure that it didn't make anything up. It's taking out signs of AI generated writing. It's copy editing, so it goes through this whole workflow, and then it emails me a draft of the post. So I will spend two to six minutes rambling to my phone. And then about two to three minutes later, I, after I submit that, I will have draft, a draft, I will actually have three drafts.'cause I run it through three different automations. When I, it's when it submits it, it runs it three different times. Once through GPT-4 oh. Once through clogged sonnet 3.7 and once through like a hybrid of the two working together. Technically I have a fourth one that's GPT-4 0.5, but I actually turned it off'cause it's too expensive. It's actually the best version, but it's just, it's not enough better to justify how much more money it costs. So I don't run that one. So I will have the draft and then I only have to spend a few minutes editing. It means that all of my LinkedIn posts are original thought leadership, my own ideas, but I can create them in a fraction of a time. So that's probably my most used high value automation for marketing. but I have tons of others. Like when I onboard consulting clients, they do a survey, like a question, not a survey, a questionnaire to tell me some of their needs and stuff. I have, as soon as they submit that it triggers an automation that like basically takes, I mean, it's a bunch of stuff I've given it, but it figures out exactly what assets I'm gonna need from them for their training and to draft the email to ask them for the things that I will need for the training. I will have it review emails from pr, like, do you know what Help a report or out is. Are you familiar with that? No, they, it's, so there's a website where like reporters, if they need PO experts in a topic, they'll go put in their request for the experts they need.
Audrey Chia:Interesting. And
Nicole Leffer:it's called Harrow, HARO Help a reporter out. And they send emails every multiple times a day. Oh. So it's like, Hey, we're looking for an expert in this. 95% of them are completely irrelevant to me. Even though it's like tech and ai, it's not anything to do with what I would do. So I have an automation that anytime one of those comes in, it checks, it decides if it's relevant for me, and it flags it for me if it is relevant. So there's all kind so that I don't miss PR requests. So. All kinds of different things that you can do with the tech. It's really pretty cool.
Audrey Chia:and it, I mean, those are just a few, it's so many use cases, right? it's just, yeah. How do you then start streamlining your own processes and figuring out where best to implement ai and what is your, like what is your tech stack look like? right now
Nicole Leffer:it's probably a lot smaller than you would think. this stuff, like from a, I mean, I've got lots of different tools in general, but like, actually like truly AI powered tools. I have chat, GPT. I have Quad, I have Gemini, so those are, and like I also train people on all of those, so I have to have them now chat. DPTI have two teams, accounts, A plus account and a pro account. So like my stack gets bigger of just chat. DPT. I have Zapier and then I have API access for open AI's, different models for quads, different models. I have Perplexity Pro, I have s. What else? I've gotten rid of a few things. I used to have midjourney. I don't pay for it anymore. I used to have runway. I don't pay for it anymore. just'cause I didn't do enough with those two things. Yeah. I would, again, they're great tools. If I needed them I would use them. and then ID script, which is a video ai, video editing, and. I feel like I'm forgetting something important, but those are the biggest. Oh, SUNO for AI generated music. yeah, so I mean, it's a lot. And then one, it's not really like. I have like Adobe, which has AI features in it. So like in Photoshop I'll, so like a lot of times I'll make stuff in chat GPT with Dolly and tweak it with Photoshop or in chat GPT, but sometimes in Photoshop with their generative ai. Adobe also has like audio fixing tools for podcasts, that are included with like an Adobe subscription. So all kinds of stuff, but. The stuff that I use the most. Chat, GPT, Claude Perplexity, Zapier, and the And the A, like the APIs. That's really the big stuff. And sometimes Gemini, but that's more client specific.
Audrey Chia:Yeah. Interesting. I think one follow up question today is like, because for someone who is newer to the world of ai, this seems like. And even though I'm like, that's not that much, they'll be like, oh, where do I begin? So that question is like, for someone who is newer in their journey, who is, for example, in a marketing team, like how, how do they even get started? Where do they begin? You know,
Nicole Leffer:one tool. So I just gave you a bunch. Don't no one and only one type of account on that one tool. The only reason I have four Chad GPT accounts is because, like I need different types of accounts for different types of clients. Not because like any individual actually needs four types of Chad GPT accounts. So one tool, and I would make it a core tool. So when I say a core tool, I would include in that chat, GPT, Claude, Google, Gemini. And if it's the only I. Core tool you're allowed to have, Microsoft copilot would fit into that category too. So that is a tool that you can access these models in a user friendly, or at least theoretically user friendly I interface. And, depending on, you know, like Google, Gemini, you can use it in a few different places. My co-pilot, you can use it in a few different places. but Quad and Chad, GPT are just in their own interface, but you can use it and you can. Bend it to your will, right? So like you can literally like one minute be doing one thing with it, and the next minute be doing something totally different with it. You know, you could be doing marketing and you could be doing meal prep for your children in the exact same tool, and it will perform just as well at either of them. So a core tool in my mind is one of those tools that you can use like that. I would pick that one tool and I would learn it inside out. Learn everything you can. What is it? What are its features? What does it do? What are the models that it offers? Like. You don't need to stress over what model do I use for what? Find the one that's gonna work the best for like the most things. So it's probably gonna be GPT-4. Oh. If you're in Chad, GPT, Claude Sun it. If you're in Claude, like you don't need to get. Stressed about all the complexities, but really learn how to use that one tool. Learn how to communicate, get, write really good prompts, get what you want. Understand like how does it connect to the internet? Does it connect to the internet? How does that work? All of these different things, like what kind of documents can I give it? Can it see, does it have the ability to have vision and understand visuals like a person would? What are all of those pieces? And get really good with that. Do not start building a tech stack with like a bunch of different things, right? So like, yeah, I have a big tech stack, but I'm an AI marketing consultant. I literally like, it is my job to use AI all day, every day and understand how to do some of these things. It's not most people's world to do all of that. So focus, and I think the biggest mistake people make is trying to find different tools for everything they do, different models for everything they do. If you just zone in and go, I'm just gonna learn how to use this one thing, you're gonna build skills that are transferable to anything you're working on.
Audrey Chia:Yeah, and I think what I love about what you said is basically not spreading yourself too thin and learning focus, right? So if you're able to focus and do one thing well, it's like learning a new skillset. You focus, you learn that skillset with that tool, and then start expanding it from there after you've gotten like the foundations cupboard.
Nicole Leffer:Yeah, it's like if you think about like, I think graphic design, like from a historical standpoint is a really good example to look at. I know a lot of people who work in a graphic design, very few of them are super pro experts at like. Adobe Suite and Canva and like every other random tool that is out there, right? Like they're usually like, I know Adobe, right? Like I know Adobe inside out. And they don't start with every Adobe tool. They probably start with either Photoshop or Illustrator, right? Like, so they're, you know, you pick the tool. Once they understand the one, then they can learn the next one and the next one. But they kind of stay within like what compliments each other. They're not learning five different tools to do the exact same thing. And I think we're kind of seeing AI evolve in a similar way where it's like your chat GPT is gonna basically be like your Adobe Suite or your claw is gonna be your Adobe suite, or your Gemini is gonna be your Adobe suite, but you don't have to use every single one of these.
Audrey Chia:You mentioned chat GPTA lot and I think one thing that stood out to me was also how you use it for custom and with its custom instructions with different prompting styles. Now what are perhaps some underutilized features of chat GPT That I know, you know, but not many people know. Of course there are a lot of hidden gems in there.
Nicole Leffer:There is a lot you can do with chat GBT and people just don't have any clue. Right. so I think my number, this is. This is something that's gonna be like, wait, what? Like, did you really get excited about this? But the edit prompt button is like my favorite thing on chat, GPT, and most people don't even know it exists. Like it's kind of hidden. It's kind of secret. But when you put in your prompt in chat, GPT. If you then go hover over your initial prompt, there's a little pencil button, pen button that'll appear, and if you click that, you can edit your original prompt. I think that is so insanely powerful because no matter what you're doing realistically, unless you have like prompted so much, and even if you have, you are not gonna get it perfect on the first go. Right, and most people start chatting to change the outputs, and then that's when your responses start going off the rails and you don't really get what you want. Versus if you put in your prompt, you look at the output, you go, where did I like, not get clear enough in my prompt. And then go and use that edit prompt button and refine your prompt so that it's more specific, more exact to what you want, resubmit it and get your output. You get way, way, way better results. So that doesn't sound like a flashy fun, like it's not really an AI feature. It's just like interface feature. But I absolutely love, and by the. Claude has that, Gemini has that, like most of these tools have brought that feature in. But that is an amazing tool that nobody really seems to know exists in chat DPT, but more exciting, fun stuff. Like, like a hidden, oh my gosh. a lot of people still don't know that if you're using the app on your phone, like one, you should use the app on your phone. Yeah. And this isn't even necessary. Fairly a marketing use case, although it could be depending on the type of marketing you're doing there is if you go into the voice mode where you can have a verbal back and forth conversation, there's now a little video icon and if you push that, it can see and in real time. So let's say you are setting up a trade show booth. For your company, you are like something in person and you want the AI to help you with like placement of things. You could actually be showing it how you've done it and ask it for advice of how you know you can make it look better. That's like a marketing use case for it. Being able to see is, you know, giving it the visual. It can see just the video, almost like you're FaceTiming with it and it can give you feedback. You or like a store display. You know, if you're not a B2B company, you've got big displays in a store, you could have it look at that kind of thing. Or product, packaging, anything like that. and then along the same lines with vision, I think a lot of people still aren't aware. You can put images in. And like just in chat GPT in general, just a screenshot or a photo and the AI can see it and understand not just words that are in the image. I think people maybe realize that, but it can see layout, it can see color, it can see all of it just like you would. And so being able to leverage that is really powerful. There's, there's so many things Chad GBT can do, so, so many different things. but those are just a few that I think a lot of people are not necessarily. And
Audrey Chia:I think it has gotten so much better at analyzing images, even files, right? Like in the past when you uploaded a PDF file, you, it would miss out things. But right now it's so thorough. I was very impressed at how thorough it became when, and even looking at charts and graphs, right? These are things it used to miss, but right now it's like oof getting everything right. It getting,
Nicole Leffer:it's getting so smart. If you have, If you have a paid account, which I mean, I kind of talk about PT through the account. I can't imagine not paying the 20 a month level. But like if you use deep research, which is pretty new, tag, GPT Oh. at least as of when we're recording this, I don't know when people listen as of when we're recording this, it's very new. if you use that, you can give it files as part of the deep research as well. And so it can do some pretty cool stuff of like. Now for anybody that doesn't know what deep research is, what, what that is, you put your prompt in and if you wanna have a file, you attach the file or whatever, but you're giving it the prompt and you're asking it to go do research on a topic. But research could also involve. Bring that research to strategy, right? Like it doesn't have to be just like, tell me about a topic. It could be like, go learn about this topic and about my company and write me a strategy that combines, you know, the two to accomplish X, y, Z goal. It could be very specific, but it will go out. It'll, you submit your prompt, it asks you a few clarifying questions, get some more information from you automatically, and then it goes out and it'll spend anywhere from like. Seven or eight minutes to, I've had it take as long as an hour of doing research. that was a real complicated prompt when it sent an hour doing it. and that was in the pro account, so I'm not sure. I haven't gotten it quite that high on time. Yeah. And plus, but it's still, I mean, it can spend quite a while and then it will write you a in-depth report. Exactly what you asked for. I mean, I've literally seen it put out a 50 page report before. Wow. Like, and this is really high level research now. It can still make things up. but it, so you do need to fact check it, but it is. So insanely powerful. So if you, you could give it PDF files to use as part of that process as well that you specifically wanted to look at. I've even seen it open up like code, like start running Python code to do analysis and mathematical stuff as part of its research as well. It's pretty wild.
Audrey Chia:It is. I have. I particularly love the deep research feature because in the past, I think chat GPT was not that great a research, like even if it could search the web, it would just pull out like, you know, just like five links and then yeah, some of it would be irrelevant. But I think with the deep research feature you can see that AI is thinking at such a thoughtful level, and it's really going into the details and pulling out insights even that you would otherwise have missed.
Nicole Leffer:Yeah, absolutely. No. So at least right now you can only use 10 a month in like a plus or a teams account. If you are like, oh my gosh, I'm using so much of this and I burn through it and I need more, you could always just add accounts, right? Like I think a lot of people are like, I'm either spending$20 a month or$25 a month for plus 13, or I'm spending$200 a month, 200 for the pro like. Let's be real y'all. I think people forget like the fastest and easiest way to double your, your amount that you have is just get a second account of whatever cheaper type. But like if you're gonna use it all the time, the pro account gives you like 120 deep researches a month, which I think would be because I, how long it takes. Pretty hard to burn through all of it. So you kind of have this range. But do keep in mind, I mean that's how I ended up with four tattoo.
Audrey Chia:No, no, no. Take stick. Stick up.
Nicole Leffer:Yeah, because they've always had the best features in Chad, GPD do tend to get rate limited. Yeah, I mean it's kind of always been how it's been. so I just ended up with multiple accounts. So like you run out of that model or that feature on one, you move to the next account. So that's always an option to scale your usage without having to go to like 10 times the price.
Audrey Chia:And that's a secret hit right there. That's a
Nicole Leffer:today. That is definitely, that is how we got today's contact.
Audrey Chia:Yeah, that's right. And And looking ahead, Nico,
Nicole Leffer:we still use pen and paper. Oh yes. That's super
Audrey Chia:important. As a copywriter, marketer, that's still the best. But looking ahead, Nico, how do you see AI perhaps, you know, further evolving, specifically that B two P marketing space? Like what are you excited about? How do you see how teams will change? Will lead generation, you know, systems change? What is exciting for you?
Nicole Leffer:I'm gonna say something that is simultaneously exciting and also terrifying, and I don't know if I want it to happen, but I know it's coming. Yeah, and I think people need to understand this so. AI agents are gonna be something real very quickly, and they even, they're already real. But let me take a step back because everybody's slurring around the term AI agent, and most people are not referring to AI agents when they talk about AI agents, right? So a lot of people call their GPT and Chad GPT, an AI agent that is not an AI agent. Deep research is maybe an AI agent, but an AI agent essentially, or true agent, means that it is a, you give it a prompt, you ask it to do something, and then it autonomously goes and doesn't. So it can like take control of your computer or have its own interface of a computer workspace, and they can go out and accomplish that task. Now, the vast majority of things being marketed as agents aren't, they're automations, and that's different because you still have control versus an agent. You might just be like, Hey, go out, do the research and buy me this thing. And you got like, you're out of the process. As soon as you do that, like a true AI agent, it's like sending a human off. It's gonna figure out how to do it itself. It's gonna make those decisions. It's going to take all those steps. They already exist, right? Like so OpenAI, if you have the pro account, you can play with their operator agent, which can go off and do some of this stuff. There's more and more of these are coming out that are actually starting to be a little bit more functional. A lot of questions like a lot of companies will not allow you to use them because. You're autonomously letting a computer take control and make decisions. but I think this is like kind, this is the big thing that's gonna be like very disruptive within B2B marketing and like everything else. Right. And so you're gonna be talking about being able to go, hey. Go out, research this prospect for me on all of the different platforms, all of the different stuff. Write me an email cadence, send the email, like it just gets to a whole nother level. Schedule the call, like do all of these things. It's. Really crazy and like, it's really cool and in one way I'm excited about it. And in another way I'm like, maybe we should just like not do this. Like I, what could go wrong? Right? So what could go wrong? There was a bit of like, are we sure we're ready for this? For me, about that technology? I think it's cool. It's fun to play with. And it's also a little like, whoa, this is so sci-fi. So,
Audrey Chia:and how do you think, for example, because. AI is already here and it's only going to evolve and, you know, expand its reach from here. How do you think that humans, you know, and marketing teams should react? What, what should people do about it?
Nicole Leffer:Not, not to shut it down just because you're scared of it. Right. So I think that there's a lot of people I'm seeing that are terrified of what AI is doing. And so they refuse to have anything to do with it. And that's not a good response because like, here's the reality check and I, the reason I say that's not a good response is'cause it's gonna hurt you to be that, to have that response. To be honest, like I'm also afraid of ai. Like, I don't just like, I love it. I think it's really fun and really cool and really exciting. And also it terrifies me like legitimately, like keeps me up maybe more than other people because I see where we already are and what's possible and where it's going. But I think that what so many people do is they shut it down. I want nothing to do with it. The reality is that the people who are gonna be the most secure in their jobs for the longest amount of time. Are the people who are learning how to be the ones that control the ai, that understand inside out what the AI is capable of, how it works, how to get it to do what you want it to do, how to control it, how to reign it in. Like all of those things. Those people are gonna be significantly more valuable in the job market for the longest. Yes. So the best way to combat the fear is to learn it, like the fear when it comes to what does this mean for my job, my future, all of that. The best way is to be the person that knows how to control it. And I mean, that's part of why I, you know, I recognize what's coming. It scares me. I don't like it. And also, I'm gonna protect myself as much as I can for as long as I can by understanding this technology. And then the other piece of that is. Like, that's why I want to teach as many marketers as possible to use this because everybody who learns it is gonna have more job security than those who don't.
Audrey Chia:Yes. And I also actually said this to myself at the beginning of my AI journey. I said, I'm not gonna wait for a tool to replace me. I'd rather be the one learning how to replace myself. So, right. Learn how to leverage it.
Nicole Leffer:You know, I'm old. I am like old enough to remember when people lost their jobs because they refuse to learn how to use the internet, right? Like they refuse to learn how to use Google. They refuse to send email. Like they really, you know, that was the thing. And like most people probably are either have forgotten it or they. Didn't live through it to experience it, but like it wasn't always a given that you would just know how to use the internet. You had to learn how to use it, and there were a lot of people who didn't. And you know what? At some point, companies would've stopped hiring the people who had no computer skills. Right. Like, and I like if you're younger and you're like, oh my gosh, like the boomer in my office that doesn't know how to do this or that, like multiply that by a thousand because that's what you're talking like, they literally didn't know how to go on the internet. And they refuse. Like, I'm not gonna have anything to do with this whole internet thing. Like y'all. Those people didn't keep their jobs. And that makes sense to you now because you can look at it in hindsight and gamble, like why would you think you would've kept your job? We're at the same place with ai. Yeah, right. Like refusing to learn how to use this technology is refusing to learn how to use a Word document or email or Google in those days. So we just have to realize that. Like that is where we are. And like it or not, this is the world we live in now.
Audrey Chia:Yes. So really the best time to get started learning is really right now. Right now, right now the best
Nicole Leffer:time was last week. And if you haven't already started last week then like today, right now, right this minute is right now, obviously if they're listening to this, they're probably already at least a little in their learning journey. but, and it's not too late, but like, just running away out of fear is not the solution to protect your job.
Audrey Chia:Absolutely, Nico. So thank you so much for sharing your insights today. Welcome. It has been amazing. Now. Where can our listeners find you and who should reach out to you?
Nicole Leffer:Yeah, so I post a ton on LinkedIn. That's my primary channel where I'm like educating lots of free stuff. Like I'm not like trying to sell anything. so LinkedIn, just my name, Nicole Effer. if you are coming from like hearing me on the podcast and you wanna actually connect, please just put a note in to tell me where you came from and it like, that's why you're connecting.'cause I have to weave through and decide who's trying to spam me and. Like try to just guess so much rather know. Yeah. but definitely if you are a marketer or B2B marketer, especially, but you, any kind of marketer, I would definitely come check out my content on there. If you have a marketing team, you want to up. Skill with ai. Then my website is just my name, nicole effer.com. I train marketing teams on how to leverage this technology. so that's the thing. And I also do have an online course, the foundations of generative AI for B2B marketing. So if you really wanna learn this stuff in depth, I teach it there. So all kinds of, all three different places and that you can find that course from my website as well.
Audrey Chia:Awesome. We will drop the links in the podcast and thank you again, Nicole, for joining us. It was a pleasure having you on the show. Thank you guys. Thank for tuning in and don't forget to subscribe to the AI Marketers Playbook and hit the develop for more actionable marketing insights. We will see you next time.
Nicole Leffer:Thank you. Thank you so much. This is fun.