Most folks who follow me know I’m a bit of a late adopter and skeptic of AI. Yes, it can do some impressive things that takes humans a long, painful amount of time to replicate. But, they’re also stereotype machines whose reasoning and capabilities, especially in the creative and novel needs that marketing demands, isn’t replacing even 1% of jobs (despite what my LinkedIn feed has promised each of the last 4 years).
However, I’ve seen how incredibly thoughtful and creative some SparkToro customers have gotten by combining audience research data with LLM prompts to solve thorny tactical and brainstorm-ideation-focused marketing challenges, and even I’m impressed. On balance, I have to agree that combining real audience data from the right sections with smart prompts and the right models yields powerful solutions.
In this week’s 5-Minute Whiteboard, I’ll walk through one of the most useful things ChatGPT’s o3 reasoning model (which also has the ability to search the web and crawl websites) can do when given a SparkToro audience research report CSV.
If you want more prompts, ideas, and walkthroughs, please join us June 25th for the next episode of Office Hours: How to use SparkToro (and a bit of LLM magic) to solve 10 big marketing problems. It should be a doozy 😎
Transcript
Okay, friends.
Some people think AI is gonna take their jobs. I think you’re gonna do your job better with AI. This is a SparkToro audience research report that you’ve probably seen me show off a bunch of these. This one in particular you can see is about, folks who have the word care facility in their bio or their profile, meaning they work at a health care facility or, a medical care facility, something like that. My my grandfather is actually in one of these places. And, you know, so the the managing director of that place who makes buying decisions about what software to get or which HVAC contractor to use, that that’s the people that I was trying to help a a SparkToro customer find earlier today.
And one of the things that this person mentioned to me is they’re they’re like, gosh. These websites are great. This list of websites is great. Can SparkToro tell me which ones offer advertising or sponsorship opportunities?
No. But, actually, I’ve been talking to a lot of our smart customers at SparkToro, and they’ve been telling me that what they do is they take whatever project they need to accomplish in SparkToro, and then they they export this data. Right? So this big list of all these websites that people who go to care facilities or or working care facilities are visiting. And they export the CSV, and then they upload that to ChatGPT or their their favorite LLM.
And, actually, if you use the right model, in this case, I had to switch to o three, the advanced reasoning model for ChatGPT because, I usually, I use four o, but it was not great for this task, as you can you can see from some of the results I got. But, SparkToro so I uploaded the SparkToro report of people who used, the word care facility in their bio and the specifically the websites that they visit, this list, the same list that you see here. And then I said, hey. I would like you to crawl through these and uncover any that offer event sponsorship, an email newsletter I can sponsor, an on-site advertising that I can purchase direct rather than going through Google or or Meta, or any other opportunities for partnership collaboration advertising.
And then spark ChatGPT actually spent a healthy amount of time, searching. This this took it about three minutes. It’ll describe its process a little, and you can see whether it’s on the right track. And then it gives me a sample CSV, which actually does do the job.
So it shows me, hey, what are partnership opportunities, which ones offer advertising, etcetera. And then it says, hey. Do you want me to do this for all two hundred? It’s gonna cost a bunch of tokens, etcetera, etcetera.
I think this is insanely cool because we’re you’re you’re not taking away jobs. You’re just making it so much more efficient than trying to do this on your own. This is a process that, you know, a lot of marketers are gonna have to do. They they see a big list and they’re like, okay. Well, I have a specific task I need to accomplish. Can can someone just do that for me? And rather than them spending hours and hours and, you know, building a big spreadsheet themselves, ChatGBT is really good at this and it will make you better at your job and more efficient.
I actually have a huge list I have been assembling of all sorts of tasks that you all have told me you do with SparkToro plus AI. I’ve been watching people’s LinkedIn posts and some of the Reddit posts that have been up there or some of the, stuff that people have put on threads and blue sky and Twitter.
And so what Amanda said is, Rand, we should do specifically a an office hours episode where you walk people through your sort of favorite use cases of people using SparkToro plus a large language model to solve a bunch of these marketing problems. So that’s what we’re gonna do. June twenty fifth at eleven AM, just a few weeks from now. I’m gonna be walking through a whole bunch of these.
Now this one is deeply cool, and it is amazing, and it can really help you. But I think there’s a ton more. And if you’d like to come to this office hours and and share your other ideas and see how creative some of your fellow SparkToro users are getting with this stuff, see what AI plus SparkToro can do, I think that’s great. You know, if you’ve been following me, that I’m a little bit of an AI skeptic.
Like, it’s cool, but I worry that we are overusing it in some cases. And so in these use cases, I’m gonna gonna be very particular about which ones I think AI does a great job with. And if I don’t think it does a great job, I’m not gonna show it to you. Alright.
That’s that’s my pitch. Come join us June 25th; hope to see you there!