8: The 80/20 Rule for AI: A Conversation with John "JK" Kornegay
Episode 8 of Kinwise Conversations · Hit play or read the transcript below
-
Lydia Kumar: Welcome to Kinwise Conversations, where we explore what it means to integrate AI into our work and lives with care, clarity, and creativity. Each episode we talk with everyday leaders navigating these powerful tools, balancing innovation with intention and technology with humanity. I'm your host, Lydia Kumar.
Today I am joined by John “JK” Kornegay, a professional who uniquely bridges the analytical and the artistic. By day, he's a digital marketer with over a decade of experience in SEO and strategy. By night, he's a photographer capturing human stories.
We'll explore how he leverages AI as a creative companion and an efficiency engine in both worlds, how his early awareness of automation shaped his career, and his practical strategies for using AI to get to the crucial last 20% where human expertise shines.
Lydia Kumar: Hi, JK. Thank you so much for being on the show today. I'm really excited to hear about your story and how AI has been a part of your journey. Before we dive deep into our conversation, I want to learn more about who you are and what listeners need to know about you to understand how you approach AI and how this is a part of your work.
John “JK” Kornegay: Oh, well, thank you everyone. I am a digital marketer by day, and a photographer once I clock out. Ironically, a lot of the tools that we use on the marketing side and in photography, when we talk about post-processing, AI or machine learning has always been a part of it. Now we're just at such a new frontier of understanding that we can kind of have a companion that really helps us work through things. The kind of the stuff that you would see in sci-fi, the stuff that you see in comic books and things of that nature. It's really now here with a little bit of ingenuity and a little bit of creative thinking.
My journey starts, you know, again marketing for about roughly 12 years. I feel like I'm aging myself in that regard, but I've been into marketing here for about 12 years. I initially started with inbound marketing, kind of learning all those Swiss army knife skills of how to attract folks to websites. For the past six years, I've focused that specifically more so on the SEO, technical SEO, PPC, a little bit of reputation management, social—just kind of that field of digital marketing in capacities, along with being an event and portrait photographer.
So those, ironically, when it comes to skill sets, they kind of hand off to each other in some capacities because I have the ability to have a conversation with a client or a potential business to say, “Hey, you are using these images for what purpose?” And then I can kind of give some perspective insight as to what these images may or may not convey. Or on the other side of it is, “Hey, who is your target audience?” You know, based on some of the demographic of what your website copy is looking or how it reads, you may need to kind of tweak. So I'm armed with an interesting skillset of these two types of creative and kind of analytical things. And so I kind of merge 'em together and with, you know, all these LLMs out here now today, it kind of has really amplified just one, my creativity and two, the sheer ability to test and try things out.
Lydia Kumar: Right. It feels like you're able to create people's stories, like through images you're able to say, what is the story that you wanna create? What do you wanna highlight in your website, in the image that you have and you're able to bring that out. How did your passion for storytelling or perspective shaping start, or when did you realize that you were really skilled at that?
John “JK” Kornegay: It actually happened by accident. Prior to getting into digital marketing, I worked for the community college system. Also, I do have a little bit of experience in higher education. And so in running a particular program, you kind of have to put all of these skills together. So it was literally me realizing that, okay, I have a student story here. And the ‘why’ is the most important. And so like, why is the student kind of pursuing their dream of, you know, again, obtaining certificates, obtaining a degree, things of that nature. But then there are also the other whys, what are some circumstances that would cause them not to? I had to learn how to craft these stories to kind of get buy-in for particular resources to be attributed to the students. So it became an understanding of like what makes the most compelling story.
And then taking that along the way, and over time, kind of just learning some of the tips and tricks as it pertains to best practices or emerging practices, I'd like to say now, because with our big topic of discussion being with AI, AI allows us to take best practices, amplify them to the point where they're no longer the best practice because we now have new emerging or bleeding-edge practices that we can now attribute. So it's just been a combination of, I guess, trial and error of understanding what's the main thing that we're trying to accomplish. And over time, just working my way through that.
Lydia Kumar: You brought up AI, so I'm gonna take us there. When AI first came on the scene for you, did you see that as an opportunity? Were you concerned? Like, how did you start thinking about the technology when you first encountered it?
John “JK” Kornegay: I like to think, I'm one of those people who runs towards the actual thing. I am of age to understand Terminator, Terminator 2, the whole—I understand Skynet. I get that. But for me, it was one of those things where I was already using Grammarly. Grammarly was already embedded into my Microsoft Word, my Google Docs, and then my emails. So to a degree, Grammarly is a certain form of AI-assisted writing. There were just other tools that I was using already that weren't just called LLMs openly and flatly the same way that we kind of have them today. So when the first option to get OpenAI or ChatGPT came in that first run, my background is in computer information systems, my undergrad degree. So it's not foreign to me in understanding coding and programming and stuff like that.
I just applied it and just kind of made it something super, super simple as, “Alright, I see this Excel spreadsheet that I have to kind of go through and synthesize this data. Could ChatGPT do that?” And then once that kind of clicked and it was like, “Yes, not only can it do it, it now takes X amount of time off my plate just from me having to do the busy work.” That's when the aha moment kind of clicked for me. It was really swift for me.
Lydia Kumar: I think a lot about these kind of time switches. If you're not having to do that busy work, then you can do other work that may feel more important for you in terms of being able to amplify a story or whatever you're most passionate about. And so I guess I'm curious about a specific way that you've been able to figure out how to make those trade-offs. Like when do you use AI? When do you not? What do you think is most important for you in terms of the gifts that you have or the things you see yourself being most skilled in?
John “JK” Kornegay: So my default is I love information, I will hoard information. So if I research something, I'll research maybe four or five different sources just to kind of make sure I kind of understand it. But sometimes that is analysis by paralysis because I am a recovering perfectionist in some capacity too. So it has to be this particular way before I can launch. But what I've learned over the past, maybe let's call it 23 months to two years, is just get to 80%.
And what I realized is that how I leverage AI, I let AI get me to 80% and the last 20% is all my subject matter expertise, my creativity, and that 20% is the thing that really matters. Whether we're talking about creative briefs or doing things in that manner, that last 20% is on me, and that's what's going to be the deciding and differentiating factor between whether it's something that the client is gonna love or hate, because otherwise we know that our clients also have access to ChatGPT as well, and we know that they're tinkering and they're playing with it as well. So we're all becoming savvy in real time to kind of understand, and we can read and we see the nuances of how, okay, this might've been written with ChatGPT, you just copy-pasted it. Whereas, no, I'm still going through some of the tried-and-true processes that I would before, but I'm just not doing all of it now. So ChatGPT takes over the thing that is the mundane, the repetitive, and then I can now have a lot more brainpower to think more creatively or from a more strategic perspective in terms of providing the end result for the client.
Lydia Kumar: That's really cool. It's like you're able—a recovering perfectionist—but also you're able to use that perfectionism or that ability to see what this could look like at its best and instead of devoting that time at the end instead of devoting it to the whole process. Right. So what does your new workflow look like right now?
John “JK” Kornegay: It depends on what I'm working on. So let's say, for example, if I need to write some copy for a particular website. What I would do is first go through and look at each particular webpage, using other various plugins to see if it's marked up, if it has its proper H1s, H2s, things like that. So that's the low-hanging fruit. But then there's the other aspect where I will use an LLM—Gemini, Google's version in particular—and run some deep research about not only the company but who the company serves. Is there a socio-economic aspect of who this company serves? What's happening in the location of that particular area? Are there things that I can glean just from a local level?
Having all this information then allows me to better synthesize. Okay. It would be great to highlight that we have a new development that is being built over here and highlight how that will have an impact on the community, or this new restaurant is now open. And now that the LLM—Gemini in particular—has seen that and has gathered that, I can now create social media posts that will automatically highlight that particular new restaurant that's in town that people are enjoying, just for the sake of having it more cohesive and allows us to not just have the cookie-cutter, run-of-the-mill, copy-paste job that one would do when trying to create any type of content. From a secondary perspective, understanding that LLMs are now becoming a part of search, now it's how do we make sure that our current websites are no longer optimized just for human intent, but also being crawlable for LLMs? Because you want your information to show up there, to be shown as a credible and trusted source.
Lydia Kumar: And so then you have to know how these LLMs think and how they work. Right. You have that background in computer science. Do you feel like that has helped you learn that or have you done your own research? Like how have you figured out how to think like an LLM or write for a human and an LLM at the same time?
John “JK” Kornegay: Well, it's one part—this is the beauty of being at the ground floor, that everybody's figuring it out together in real time. So you Reddit, you YouTube, you have a bunch of these folks who are really just A/B testing in real time and sharing their results. It's kind of almost like an army of many really going out there and figuring things out like, “Okay, this use case would be great for this sector, this use case would be great for this. If this doesn't work that much in this sector, but it works really good here.”
So that's where I think the beauty of it is, is it’s really adaptable. And it's really from a perspective of just individual creativity. Yes, you could have someone who has a very similar approach, but the uniqueness, as long as we still have our own human ingenuity, our own thinking, our own creativity, our own style of how we like to write, our own prompting and perspective, that still helps the output be that much more different.
Lydia Kumar: That's so smart. I've been thinking a lot about in this new age of AI and how you can do so much collaboration with a tool, but at the same time, I think networking and your relationships with people are becoming even more critical because there are people on the ground figuring this stuff out. We all have access to this technology, but we only have access to the people that we interact with. And it sounds like you're doing a ton of interacting with people, even though like probably in real life too, but on the internet, like YouTube, Reddit, all of these places where you're reading things that people are saying. I also love Gemini for deep research. It is legit.
John “JK” Kornegay: It is legit. I love it.
Lydia Kumar: But I think that combination of saying, “Oh, I can do this research with this LLM tool, this Gen AI,” but also you have to know who are the people to talk to, who are the people to listen to, what's the YouTube channel that I should be watching right now to learn this stuff. That's also important.
John “JK” Kornegay: I'm a big Tim Ferriss 4-Hour Work Week fan. So when I first graduated college, I read his book and a lot of what that book was, was breaking the cycles of what is traditional and streamlining through productivity and optimization. So his book is the precursor of everything that we kind of see now in terms of like optimize for this, do this. And kind of taking that not-so-traditional approach to solving and optimizing for what's really important.
I say that to say that is also a key to consider: books now, or utilizing knowledge in book form for AI also is something that I strongly tell people to do. Because if you have your favorite author or your favorite person from the financial sector. If you're a Dave Ramsey fan and you yourself have done some Dave Ramsey savings, and if you wanted to now write your journey or document your journey and now become, not a Dave Ramsey acolyte of sorts, but if you're like, “Hey, my journey in becoming financially fit,” you can have your tenants still be very much how Dave kind of structured it because this was something that you enjoyed, but then you still can tailor it to your voice, how you write, how you want to come off. And again, AI allows you to do that.
Lydia Kumar: Yeah. I sometimes think of it as like calling a muse where you're like, “Oh, I wanna write something. I wanna talk with someone who has these tenets of Dave Ramsey.” and tell ChatGPT or Gemini or whatever tool to use that. But you have to understand who the people are out there. You have to know that you appreciate Dave Ramsey's approach or you wanna know what Dave Ramsey and some other person you respect might critique Dave Ramsey. I think those thought experiments, at least for me, have been really useful. But you have to have some base knowledge that's outside of the LLM to be able to interact in new and creative and useful ways.
John “JK” Kornegay: Right.
Lydia Kumar: So do you have a recommendation of someone that you're reading or listening to right now that you think is interesting and useful, from the AI perspective?
John “JK” Kornegay: From the AI perspective? There is a woman, her name is Sabrina Romanoff, I think. I enjoy her. She's pretty much on all platforms. I've never seen her on LinkedIn, but I know she's on Instagram, she has on YouTube, and I feel like she also just throws her content on TikTok as well. She builds, she comes back, she reports, she builds, she comes back, reports, and then she now has just positioned herself to have a newsletter, and you can just pretty much know that she will talk about all the new things.
So for me, I just synthesize, “Okay, is this applicable to the work that I do? Is it not?” If it's something, I'll kind of go back and revisit. Funny enough, because I've been playing around with LLMs for so long, recently at my job we watched this town hall meeting. They pretty much were talking about everything I've been talking about for the past two years, and everyone was like, we're in a meeting and people just looking at me and I'm like, “Uh, guys, I could have seen that we would start adopting some of these things.” I know how to do a good portion of everything that they were talking about that, “Hey, we're now going to be showing this to the client.” So it allows us to kind of shift from a more doing-the-work approach to now having more of a strategic engagement with our clients going forward.
Lydia Kumar: Yeah, I love that. I think what you're saying about how you've been kind of ahead of the curve and now everyone's catching up. I think it has been hard for people to understand how revolutionary this technology is. It's uncomfortable, even if it's useful. It does feel a little bit unbelievable. And so I think it takes people—some people are able to grasp that more quickly—but I think it's taken time for individuals to realize this is gonna have an impact on how we work. And it has to, or we're gonna be left behind.
John “JK” Kornegay: So I actually, it's funny you say it 'cause you just made me remember when it dawned on me. When Andrew Yang was a presidential candidate, he went on a platform and he was talking about universal basic income. And the reason why he was talking about universal basic income is because he said automation is coming and there are gonna be autonomous trucks here. Dot, dot, dot. I registered that, but I did it in reverse.
For me it just made more sense that it would probably come for jobs that are not blue-collar first, just because of the sheer—where it is, it's in a computer, it's in a machine. Those who work on computers would be impacted first, not the people who are actually doing manual labor. So that's what it made me really kind of like, I need to really pay attention to this and at least understand what this is and if I need to upskill or repurpose or re-skill something too, so I can still have some level of employment in the upcoming workforce.
Lydia Kumar: I've had the same thoughts, JK, but you were ahead of me. That's really perceptive that you were able to hear that and make that connection so quickly. Yeah, it definitely took me more time to get to that point where I was like, “Oh, I need to change the way that I'm thinking about how I work because this is gonna change what things look like.” Agentic AI is wild and has progressed so much in the last couple of months. And so we're seeing the shifts happen quickly that are gonna have a real impact.
Right before this call I was thinking about how Silicon Valley is measuring the productivity of a company by how much money is made per employee. That is gonna reduce the value of employees. Because if there's gonna be a surplus of labor, what does that do to the value of actual labor? Because if we need less people, then we become more replaceable.
John “JK” Kornegay: Yeah. And I wanted to be the former, I didn't wanna be the latter.
Lydia Kumar: What are the biggest ways you've shifted? Where have you been upskilling? How have you been trying to change your skillset to respond to this wave?
John “JK” Kornegay: Last year I got my Certified Scrum Master certification because I wanted to get a little bit better at the whole project overview, the consultative approach to certain things. So I figured if I can arm myself with this ability to do Scrum, then that kind of puts me in a different position in my current job. And so for me, that has then allowed me to say, “Okay, well how do I want to solve the problem today based on what my workday is looking like, based on which particular AI tool I want to use?”
I literally yesterday was kind of, to your point, talking about the whole productivity aspect. You know, we look at productivity in terms of time logged in cases and stuff like that. I said, “Okay, well I'm sure that there will be a shift in how we are being measured in how we're handling our caseload in terms of our clients.” So I had ChatGPT kind of reconfigure me a workflow to really implement in Q3. I was like, well, let's see, we are gonna ride this one out, but Q3 we have a new workflow based on breaking down my clients. We're gonna tier the clients out and we're gonna figure it out because these are some things I have to do on a daily, weekly, monthly cycle. So I have that roadmap planned out and now it's just plug and play and just put it into the calendar. We may have to have a secondary call just like, “Hey, how did that go for you?”
But yeah, for the most part, for me, it's just being prepared to know that I know what I know from a skillset perspective. There are two ways you can go about it. I have enough skill to create an SOP for a new hire, which I have done before. But then I said, “Well, what if I took that SOP to train a custom GPT to do that one specific task?” and just keep doing that one. So I was like, oh, once that was the first idea to fall, I was like, “Oh, well I can do it for another thing, then I can do it for another thing.” And then now I'm just playing review, approval—I'm managing my own ecosystem of agentic employees that I've kind of created for myself to help me be a better employee myself.
Lydia Kumar: That's really cool. I would love to hear about how Q3 goes for you. I am so interested in how people's workflows are changing right now. We have these tools that allow us to work differently and be more productive. And how do you optimize that? How do you manage a human and a human-robot, cyborg kind of workforce? I mean, we're interacting with people who might just be coming solely as themselves, but we might be interacting with people who are prepping with AI. What does that look like and how does that change how we work with each other and how we accomplish tasks? So yeah, I'll be interested. It sounds like you are very intentional in how you are using these tools and your approach and how you're setting yourself up for the future.
John “JK” Kornegay: Yeah. I also use it as a knowledge gap closer, so I'm also using it to teach me as well. And then sometimes I get real granular. I'll start mixing them together. So I'll start something in OpenAI and then I'll take it to Gemini and then I'll take it to Perplexity, then I'll take it to Claude to see which version gives me something the best, or like the best version of 80%, and then that's where I work off of. I make the tools give feedback to each other too. So I'll be like, “I got this from Gemini. What do you think?” Because they're trained on different data, and so you're going to get different outputs and different perspectives.
I personally think that Claude is better if you're looking at creating something that is front-facing for humans to read. Claude probably gives you the best version of that. There's not a lot of red-lining and changing, but it's really, it has more of a humanistic read and touch in my opinion. But Gemini, if I had to hedge my bets on who's going to win this whole thing, I like what Gemini has, just from the ease of use, because I have Gmail. My personal business is a Gmail workspace. It's already integrated. It already kind of does everything already on the backend anyway. You can pull and call inside of Google Drive with Gemini. So ease of use, I would think Google would probably figure it out. But I love that there is healthy competition so we don't feel like we're just being monopolized into one particular company.
Lydia Kumar: Absolutely. It just opens up possibility and ways of thinking. I'm about to take us really left field because I feel like you'll have something interesting to say about this. You know, there's that MIT study that just came out that says, “your brain on ChatGPT” and how it's reducing your critical thinking skills. Have you read it? And if you have, what are your thoughts?
John “JK” Kornegay: I saw it, it's tabled for me to read. I think that's, I don't know if that's our problem or if it's my son's problem. And what I mean by that is, my son does not—I'm old enough to understand going to the library and punching out—I'm that old. So I remember the world before we got here. I buy books. I still need the tangibleness of the paper and the page. I don't use LLMs in general for things I know the answer for. I use it more so as a creative branch to like, “How, what would be the best starting point to start this idea?” and then let it go back and forth that way.
But no, I understand the concept of if you don't use it, you lose it. And if we now have this thing that will critically think for us, potentially, that now is a lost skill. I personally make sure that I'm reading things produced by humans. And I even go far back. Like I'll go to something as early as the seventies, like '78, '79. I have some books from advertising in the seventies and eighties. I just peek through just to kind of see what still survived, what has definitely evolved. Same thing with old textbooks. I'll find 'em just to see if this knowledge could be rehashed because we know there's nothing new under the sun and we know things go in cycles. So I think that I practice trying to still have a certain level of analog learning that I feel like I won't lose some of that critical edge.
Lydia Kumar: Yeah. I've been thinking the same thing. And I've also been wondering about how, if you're a leader or if you're an educator, how do you develop that in students or employees? This desire to keep up that critical edge because you can outsource your thinking to a tool, but that long term you lose something. Maybe in the short term, there are immediate gains, but long term, what does that look like?
John “JK” Kornegay: You still need to have the skill to prompt. In order to properly prompt in a particular format, you have to understand the syntax of what you are asking for to produce end results. So I think prompting will probably be as critical as typing was when we were coming up.
Lydia Kumar: Right. And your ability to chat—all these LLMs are based on language and so you have to have the right language to get the right output. So we also need everybody to have a baseline skill because if you don't have the baseline skill, then you can't ask. If you don't know what SEO is, then you're not gonna be able to prompt very well around SEO. So you do have to have some of that basic knowledge.
You have a ton of information you gather, you have technical expertise, but you also have this artistic side, your photography business. How do those things connect for you? The technical, the artistry, and now AI, which can generate content.
John “JK” Kornegay: I've utilized AI in photography, but it's more on the marketing side. It's, “Hey, give me this, this is a goal of mine. I wanna get to this goal by January 1st, 2026. Let's reverse engineer how we're gonna do this.” And I pretty much just had ChatGPT give me a full blueprint month by month: a content plan, keywords to talk about within my platforms of choice, which would be Instagram, LinkedIn, and YouTube, and of course my website. So from the creative aspect, I'm still using it very analytically from a marketing side.
I do have a couple of ideas from a creative perspective that could actually help creatives that are coming up. I haven't done 'em yet, so I don't wanna put it out there, then someone else might steal it. But yeah, I think just having the requisite knowledge and working with the tools now, and then understanding a pain point as a creative, this would actually help a lot of creatives coming down who still have a certain level of DIY. So my idea is creating some type of application, whether it's an actual app itself, a web app, or something that I can add to my website as its own independent tool to help creatives in that capacity.
But in terms of the creative aspect, I think that is the part I still let be its own process that is very present. I do edit on Adobe. Adobe products now are very in tune with generative AI in terms of being able to remove things. So I'm still not escaping it, but the only thing that has now assisted me from the AI perspective is the ability to actually navigate, pick, and select images now. So I used to have to manually go through each image like, “Okay, this was good. I'm gonna keep this.” If I do event photography, I can probably take about 2,000 photos in two hours. So that's going through a lot of repeats, eyes closed, you know. There are AI tools that allow me to really streamline that process and that's been tremendous. It helps with turnaround time, helps with the workflow.
But when I do look at that, I sometimes take that same approach and think about it from the desk job in the sense of just because you're more productive doesn't mean that you need to be that much more productive to give yourself more work. So you still kind of gotta keep the same expectations. I now have that much more time to really focus on the images that I have now selected and the images that we are gonna work with.
So there are trade-offs in terms of being more productive, but it doesn't necessarily mean that you can produce twice the work. And that's one of the things I think about in the day job. If I'm that much more productive, does it mean I can take on that much more clients? Sure. But over time we would then have a conversation of quality of work. So managing those expectations based on the productivity is something I'm always conscientious about. So I make sure that I kind of build in a certain level of pace that is in my work so I don't feel like I'm just twiddling my thumbs or going too slow.
Lydia Kumar: That's great self-awareness. If the goal is just to do as much work as possible all the time, LLMs are gonna lead to a lot of burnout. If you were taking on twice the number of clients, even if you could get the work done, you would be doing maybe higher-level thinking work all the time. That's exhausting for your brain. So you have to find that balance. The admin stuff that is a drag that AI tools do really well are also times when your brain can kind of rest. And so it sounds like you're being really careful and finding time for your brain to rest.
John “JK” Kornegay: Yeah. You don't want scope creep either way. AI can make you more efficient, then you can be like, “Well, can you do this?” Next thing you know, you've now doubled, tripled your work just off your sheer ability to be that much more productive. At that point, you lose everything that you just gained with this productivity.
Lydia Kumar: And that's so important. If you can do more work, then you wanna write that in the contract on the front end instead of just letting it creep up on you. All of a sudden you've doubled your project commitment for the same amount of compensation. That is tricky. You don't want that. It changes expectations all across the board.
Do you have advice for creatives or marketers who aren't using LLMs yet?
John “JK” Kornegay: Just try it. I understand the apprehension, don't get me wrong. I totally get it, but just try it. Try it with something that you just think, “There's no way this could work.” And you'd be surprised. Just over time, ask it a question that you know the answer to. “How would you explain this question to a 5, 6, 7, 8-year-old?” And then over time you realize, okay, this is useful. “How would you explain this to your grandmother? How would you explain this to your mom who's a boomer?” Next thing you know, then you can really start. “A creative who has an affinity for working with florists? What's a way you would wanna reach out to them? What is a pain point of a florist? What's the demographic of people who own a florist business?” All these things could be obtained through just utilizing ChatGPT or LLMs.
Lydia Kumar: What do you imagine is coming in the next year, five years? In the future, what do you see happening?
John “JK” Kornegay: Everything bespoke. I think everything bespoke is going to be the name of the game. So there will be—I'm an Android user, I do own an iPhone—but I know that app stores were the thing. I think now because we have these LLMs that allow us to ideate our own apps and to deploy them relatively quickly, it could be a situation where, if I have an affinity for black Labradors, I could create a training program specifically for black labs. You can really just hyper-niche down and have that segment serve that set of people.
Let's go back to the florist idea. You have these apps that you take a picture of a plant, it lets you know if it's healthy or not, it lets you know how to revive it. But what happens if you just have a person who just loves chrysanthemums? Now someone can really just hyper-niche down and have a bespoke experience to their own liking. I think that's where it ultimately goes. I ultimately think it's going to be not choose your own adventure, but it really is gonna be like, “Oh, we got a lot of one-on-one experiences that we can create that sets us apart.”
You can kind of see it from advertising dollars. You look at advertising now in content communities, you have more advertisers looking for regular influencers. When I say regular influencers, not the big names that we do know who have become somewhat public figures or celebrities, but there is a lot of money being shifted to those who don't have the super large following, but they have the visibility and eyeballs of the target demographic that they want to sell to. So I think it's gonna move toward that bespoke approach with these tools.
Lydia Kumar: That's really interesting. I think I love the social media example of thinking about, it's almost like you're having to kind of dig through these layers to figure out what is the right connection point, because we can be hyper-personalized now. And so you have to take the same principles of market segmentation but use that principle in a different capacity to fit the changing landscape.
John “JK” Kornegay: Because the world has gotten a lot larger and connected, but then it equally means you're gonna have some constriction somewhere. And then, you know, everybody who likes this particular video game, we're going over here. And then inside that community, you got sub-people who are creating their own little thing. So again, I just think that the opportunities are actually endless when it comes to how we would kind of use it from that perspective. It's gonna definitely make us hyper-granular just because it's like, okay, we can tailor this to exactly your brand voice, who your target demographic is, nothing more, nothing less. And I think once that is the ultimate goal is to talk to your people. You know, what's the “1,000 True Fans” is the old marketing letter. All you really need is your 1,000 true fans and everybody else is cool, but you get your solid 1,000 true fan army, you're good to go. And I think generative AI allows us to really build that for our own ecosystems.
Lydia Kumar: Yeah, it's really interesting. I was messing around with vibe coding yesterday and I was like, “Wow, I can just sit down here and make an app.” And when you think about it, you could easily make an app for that hyper-segmented audience that speaks to them, and that's a whole new ball game. Because it used to be very difficult. I've worked on product development before and it is time-consuming. And so it was a really interesting experience to be able to give feedback to Gemini and see it implemented in real time instead of having to wait 24 hours or a couple of days for small changes. So that is just a very different world that we're living in.
John “JK” Kornegay: And that's another reason why I think Google is positioned to do well. To your point, they have the LLMs, they have the different models, they have the different ways that you can go text, you can go prompt, you can go video, they have the vibe coding aspect of it as well too. Like they just have so many of these different things that you can really just already package together.
Lydia Kumar: Well, I have one last question that I always like to end on and that's what is the idea or question about AI that you can't stop thinking about?
John “JK” Kornegay: How can AI be equitable for all? That's the question. Because we operate in a capitalistic market. Capitalism needs winners and losers. But this tool is, to the words you say, so revolutionary. Now, I think people don't really understand it in that capacity. This should help everybody. So how can we really make this equitable for all? Like, that's my question. Do I have an answer? Absolutely not. But that's the question of like, can this—I feel this should help everybody, but how do we do that?
Lydia Kumar: That was such a fascinating conversation with JK, full of practical wisdom for any creative or marketing professional. A huge thank you to him for sharing his journey and his thoughtful approach to using AI. I was particularly struck by his 80/20 rule: letting AI handle the foundational 80% of a task to preserve his energy for the final 20% where human ingenuity makes all the difference.
Continuing our exploration of what it means to lead thoughtfully in the age of AI, join me next time on Kinwise Conversations for a chat with Dr. Allison Harris-Waldo, Chief Strategy Officer at Chief of Change and founder of Thrive Life Coaching. We'll unpack how she blends intentional, relationship-first leadership with just enough AI support to keep things human.
To dive deeper into today's topics with JK, I put everything for you in one place. Just head over to the resource page for this episode at kinwise.org/podcast. There you'll find the full transcript, links to JK's work, and a list of prompts and resources inspired by our conversation.
For the leaders and teams listening, if JK's insights have you thinking about how to build a real AI strategy for your own work, I invite you to learn more about the Kinwise Pilot program. We partner with organizations to create practical, human-centered professional development and policies that empower your team to use these tools with confidence and care. You can learn more at kinwise.org/pilot.
I hope this discussion has encouraged you wherever you are on your own AI path. If you enjoyed this conversation, please subscribe and consider leaving a review. It truly helps other thoughtful listeners find us. You can learn more about how to approach AI with intention, explore resources, and join the Kinwise collective by visiting kinwise.org. And if you or someone you know is doing interesting work at the intersection of AI and humanity and has a story to share, we'd love to hear from you. Until next time, stay curious, stay grounded, and stay Kinwise.
-
I hope you enjoyed hearing John “JK” Kornegay’s perspective on the 80/20 rule, bespoke marketing, and building your own “AI assistant” team. If you’d like to follow his work or bring his lens to your own projects, here are a few easy ways to connect:
John Kornegay Photography – Browse JK’s portfolio, learn about his intimate-wedding and portrait packages, and request a quote for your next shoot.
Visit the website johnkornegay.com johnkornegay.comConnect on LinkedIn – Follow JK’s career in digital strategy, SEO, and AI-powered marketing.
Connect on LinkedIn linkedin.com/in/johnkornegayFollow on Instagram – Get behind-the-scenes stories, fresh edits, and #TaoOfJohn inspiration from his latest shoots.
Follow on Instagram instagram.com/johnkornegay instagram.comRead his marketing insights – JK contributes practical SEO and web-design articles to Inside Leads’ blog—perfect for marketers looking to level up.
Explore the blog getinsideleads.com/blog getinsideleads.comgetinsideleads.comFollow on X (Twitter) – For real-time takes on AI tools, photography tech, and digital-marketing trends.
Follow on X twitter.com/JohnKornegay twitter.com
Feel free to share these links with colleagues who’d benefit from JK’s blend of artistry, analytics, and AI know-how.
-
Pro-Tips for Better AI Results
John "JK" Kornegay masterfully leverages AI as both a creative companion and an efficiency engine. His approach highlights how powerful these tools become when you interact with them thoughtfully. Here are some pro-tips for getting better results from any LLM, framed with JK's insights in mind:
Set the Scene (Define Your Persona & Goal): Just as JK bridges his analytical marketing mind with his artistic photography, you should tell the AI who you are and what you're trying to achieve. This helps the AI tailor its output to your specific needs. For example: "As a digital marketer with a focus on technical SEO, I need help synthesizing competitor data for a new client in the [industry] sector," or "I'm a portrait photographer looking to streamline my post-processing workflow for event galleries."
Embrace Multimodality (Beyond Text): While the original prompt mentions voice, think broader. JK utilizes various tools for different aspects of his work – from Google Gemini for deep research (which can integrate with Google Drive) to specialized AI photography tools for image culling. Consider how you can use different AI modalities or tools (like image analysis in photography software or data synthesis capabilities in LLMs) to fit the task at hand.
Provide Rich Context (Your "80%"): JK gets the AI to handle the "first 80%" of the work by feeding it detailed information. Don't just ask a generic question. Upload or provide as much context as possible:
For marketing: "Analyze this website's content for H1/H2 structure," or "Here's demographic data for our target audience; generate social media posts that resonate with them and incorporate local news."
For photography business: "Here's my business goal for next year; help me reverse-engineer a monthly content plan for Instagram, LinkedIn, and YouTube, including keywords relevant to event photography."
Iterate and Refine (The "Last 20%" is You): JK's "recovering perfectionist" mindset means he guides the AI to a solid foundation, then applies his human expertise for the crucial final 20%. Think of the AI as your initial brainstorming or drafting partner. Don't stop at the first response. Engage in a dialogue: "That's a good start for website copy, but can you make it more persuasive for a B2B audience?" or "This content plan is solid, now suggest how I can differentiate it with unique storytelling elements for my photography brand." Remember, the AI gets you to 80%, but your unique insight and creativity make it shine.
Prompt 1: Leveraging AI for Digital Marketing Strategy (Inspired by JK's Workflow)
This prompt focuses on using an LLM for in-depth research and content ideation, specifically for website copy and social media, as JK describes.
You are an expert digital marketer with a deep understanding of SEO, content strategy, and target audience analysis. Your goal is to help a client develop compelling website copy and relevant social media content. Client Information: *Company Name:** [Insert Client Company Name] *Industry/Niche:** [Insert Client's Industry/Niche, e.g., local bakery, B2B SaaS, non-profit advocating for literacy] *Target Audience:** [Describe the target audience in detail: demographics, psychographics, pain points, aspirations, e.g., "Small business owners in their 30s-50s, struggling with online visibility, value efficiency and ROI."] *Geographic Location (if applicable):** [e.g., "Durham, North Carolina," or "serving the entire US"] *Current Website Pages (provide key pages if known):** [e.g., "Home, About Us, Services, Contact, Blog"] Task: Perform deep research on the client and their target audience, considering socio-economic aspects and local events/news if relevant to their location. Then, based on this research, generate ideas for: 1. Website Copy Enhancements: Identify potential gaps or areas for improvement on the specified website pages. Suggest compelling headlines and subheadings that resonate with the target audience and incorporate relevant keywords for SEO. Propose content themes or angles that highlight the client's unique selling propositions and address audience pain points. Consider how to optimize content for both human intent and LLM crawlability, making it a credible source. 2. Social Media Content Ideas: Generate 5-7 engaging social media post ideas for [specify platforms, e.g., Instagram, LinkedIn, Facebook] that are cohesive with the website content. Each idea should include: A catchy hook. A brief description of the content (e.g., image concept, video idea, text post). Relevant hashtags (5-10 per post). A call to action. Specifically, suggest ideas that could incorporate current local news or community developments (if applicable to the client's business) to make the content more timely and relevant, as JK highlighted. *Tone:** Professional, insightful, creative, and data-driven. Output Format: A section for "Website Copy Enhancements" with bullet points for each suggested improvement. A section for "Social Media Content Ideas" with numbered lists for each post, clearly outlining the elements above.
Prompt 2: Streamlining Workflow with AI (Inspired by JK's 80/20 Rule and Task Automation)
This prompt helps users apply JK's strategy of letting AI handle the initial 80% of mundane tasks.
You are a productivity expert and workflow optimizer. Your goal is to help me identify and streamline repetitive tasks in my professional workflow by leveraging AI. I aim to offload the initial 80% of "busy work" to AI, allowing me to focus my human expertise on the crucial final 20% (strategy, creativity, client relationships). My Role/Profession: [Your profession, e.g., "Event Photographer," "Marketing Coordinator," "Project Manager," "Content Creator"] My Current Challenges (specific pain points or time-consuming tasks): [List 3-5 specific tasks, e.g., "Manually sorting thousands of photos after an event," "Synthesizing data from large Excel spreadsheets," "Drafting initial creative briefs for clients," "Writing routine client emails," "Developing basic social media content calendars."] Task: For each of my identified challenges, propose a practical AI-powered solution or workflow adjustment. Think about how LLMs or other AI tools could automate or significantly assist with the initial stages of these tasks. For each challenge, provide: 1. The Challenge: Reiterate the specific pain point. 2. AI Solution/Workflow: Describe how an AI tool (e.g., an LLM, specialized AI software) could address the 80% of the task. Be specific about the type of AI and its function. 3. Human Expertise (The "20%"): Explain what the human's role would be in the final, value-add 20% of the task (e.g., "reviewing AI-selected images for artistic merit," "interpreting synthesized data for strategic recommendations," "adding nuanced client-specific details to AI-drafted briefs"). 4. Anticipated Benefits: Briefly describe the expected gains (e.g., time savings, increased focus, reduced burnout). Output Format: Use clear headings for each challenge. Present the information concisely using bullet points or short paragraphs for each section (AI Solution, Human Expertise, Benefits).
Prompt 3: AI as a Knowledge Gap Closer & Critical Thinking Aid (Inspired by JK's Learning Approach)
This prompt leverages AI not just for task completion but for learning and challenging perspectives, as JK does by "mixing them together" and getting different outputs.
You are a highly knowledgeable and critical thinking assistant. My goal is to use you to bridge knowledge gaps and stimulate deeper critical thinking, similar to how an expert might cross-reference information from multiple sources or even different AI models. My Knowledge Gap/Topic of Interest: [Specify a topic or question you want to explore, e.g., "The impact of quantum computing on cybersecurity," "The history of abstract expressionism," "Understanding the core principles of lean manufacturing."] Specific Questions I Have (Optional, but helpful): [List any specific questions related to your topic.] Task: Act as a Socratic tutor and information synthesizer. 1. Initial Explanation: Provide a clear, comprehensive explanation of the "Knowledge Gap/Topic of Interest" as if explaining it to a highly intelligent but unfamiliar adult. 2. Diverse Perspectives/Analogies: After the explanation, offer 2-3 different analogies, metaphors, or alternative frameworks to understand the topic from varied angles. 3. Potential Criticisms/Counterarguments: Based on common discourse or potential misunderstandings, generate 2-3 common criticisms, limitations, or counterarguments related to the topic. Explain why these perspectives exist. 4. Deep Dive Question: Pose a thought-provoking question that encourages further critical thinking or exploration beyond the initial explanation, similar to how one might challenge an initial understanding or explore complex implications. 5. Comparative Analysis (Simulated Multi-LLM Review): Imagine you have received an explanation of this topic from another LLM (e.g., Claude or Perplexity). From your perspective, what might be a key difference in approach, emphasis, or detail compared to your own explanation? How would you "critique" or add to that hypothetical output? (This simulates JK's process of making tools "give feedback to each other"). Tone: Instructive, analytical, and thought-provoking. Output Format: Use clear headings for each section (Initial Explanation, Diverse Perspectives, etc.). Keep explanations concise but thorough.
