Latest / Ops Cast / From Tasks to Transformation: Scaling AI Adoption in Marketing with Spencer Tahil
Transcript
- 0:48Michael Hartmann: Hello everyone.
- 0:49Michael Hartmann: Welcome to another episode of Opscast, brought to you by MarketingOps.com, powered by all the no pros out there.
- 0:54Michael Hartmann: I'm your host, Michael Hartman, flying solo this week or this episode again.
- 1:01Michael Hartmann: But joining me today to dig into how to move from experiments to experimentation to execution with AI is our guest Spencer Tahill.
- 1:11Michael Hartmann: I should have asked Spencer so you can thrust me if I mispronounce your name, is the founder and chief growth officer at Growth Alliance, where he helps companies design AI and automation workflows that drive growth across go-to-market, revenue operations, and marketing.
- 1:25Michael Hartmann: His systems-driven approach focuses on using AI as a practical tool to free up strategic thinking and improve operational performance.
- 1:32Michael Hartmann: Liz Spencer, welcome to the show.
- 1:34Michael Hartmann: And uh remind me, where where in the world are you right now?
- 1:37Michael Hartmann: I forget.
- 1:38Spencer Tahill: And so I'm usually in Prague, uh, but currently I'm in Genoa.
- 1:42Spencer Tahill: I'm in Italy.
- 1:43Spencer Tahill: So uh yeah, that's uh that's a little bit of a change up.
- 1:46Spencer Tahill: So yeah.
- 1:46Spencer Tahill: It's a pleasure to be here.
- 1:48Michael Hartmann: Well good.
- 1:48Michael Hartmann: Glad to have you.
- 1:49Michael Hartmann: I I knew it was Europe, I couldn't remember where, and I would have like if I had to actually remember, it sounds like I would have been off.
- 1:55Michael Hartmann: So I what I do know is it's but at least early afternoon, if not late early early evening, there, right?
- 2:03Michael Hartmann: Yeah.
- 2:04Spencer Tahill: It's uh it's just turned almost 6 p.m.
- 2:07Spencer Tahill: almost.
- 2:07Spencer Tahill: All right.
- 2:08Michael Hartmann: So there you go.
- 2:09Michael Hartmann: Yeah.
- 2:09Michael Hartmann: All right.
- 2:10Michael Hartmann: So you're seven hours ahead of me.
- 2:12Michael Hartmann: So good to know.
- 2:12Michael Hartmann: All right.
- 2:13Michael Hartmann: Well, let's when we start like with a little bit about your background.
- 2:17Michael Hartmann: Like, how did you get into AI and automation, especially for like the go-to-market rev ops kind of domain?
- 2:27Spencer Tahill: So it's kind of a long story made short.
- 2:30Spencer Tahill: And I was always told to if you're gonna do something, do it right the first time.
- 2:35Spencer Tahill: And even if it takes you longer, just make sure it works the first time.
- 2:39Spencer Tahill: So um, when I was starting to work in kind of like a HubSpot sphere, I think there's probably gonna be a lot of people maybe in the Salesforce sphere listening.
- 2:48Spencer Tahill: Um, I did a lot of freelance work in the HubSpot world.
- 2:51Spencer Tahill: It was setting up marketing hub, starting up service hub, setting up sales hub, um, all of these different modules, as well as connecting everything through everybody's favorite marketing operations platform, Zapier.
- 3:01Spencer Tahill: Um, at the time that was at the time, uh that was kind of like the leading edge of the market.
- 3:07Spencer Tahill: Um, and as my kind of freelance business evolved, I was able to step away from you know a full-time engagement where I was a marketing manager at a startup actually here in Europe.
- 3:17Spencer Tahill: Um, and I dove into doing full-time freelance work for a lot of American um Fortune 500 companies, like creating their HubSpot instances, helping them keep up to date with their modules and all their automations there.
- 3:29Spencer Tahill: So how I fell into it was very naturally.
- 3:33Spencer Tahill: It was just playing around with the workflows and the sequences features inside of HubSpot.
- 3:38Spencer Tahill: And then when I wasn't able to build something inside inside the platform, just moving over into no code tool of Zapier was the first one that I learned really well.
- 3:47Spencer Tahill: And then I played with make, but um, it wasn't until at least a few years ago that I kind of doubled down into into AI development.
- 3:55Spencer Tahill: So that's what we do now.
- 3:57Spencer Tahill: Um we we build out HubSpot portals mostly, sometimes in addio, but everybody wants to do more.
- 4:06Spencer Tahill: And AI is kind of like putting the pedal to the metal at that at that rate.
- 4:10Spencer Tahill: So what used to take a team of five or ten, you can do it in if you get a nice prompt going or an AI agent chain going, you can do it in a matter of like minutes or hours.
- 4:22Spencer Tahill: You know, it's it's crazy the things that you can build now.
- 4:25Spencer Tahill: And it's really interesting.
- 4:26Spencer Tahill: So that's kind of long story made short, kind of fell into it.
- 4:29Spencer Tahill: I love engineering, it's it's business engineering.
- 4:32Michael Hartmann: So it's an interesting way of thinking about it.
- 4:34Michael Hartmann: I hadn't heard that.
- 4:35Michael Hartmann: Yeah, it's it's um it's interesting because so I'm just really curious.
- 4:41Michael Hartmann: Like, do you because I've had to start doing this in my own head, like do you differentiate between AI and automation?
- 4:50Michael Hartmann: Because I think to me, what I my view is that automation is a thing where AI can be a part of an automation automated workflow or whatever.
- 5:01Michael Hartmann: Um, and I think I feel like a lot of people sort of conflate those.
- 5:06Spencer Tahill: Yes.
- 5:06Spencer Tahill: There's a huge issue.
- 5:08Spencer Tahill: Um, the long answer to this is that they're two different things.
- 5:11Spencer Tahill: Uh, or the short answer is that they're two different things.
- 5:13Spencer Tahill: The long answer would be that there's a lot of misnaming in the industry, especially when it comes to anything regarding automation, anything regarding AI.
- 5:23Spencer Tahill: And probably you've heard it.
- 5:24Spencer Tahill: I just built an AI agent on LinkedIn or something like this.
- 5:27Spencer Tahill: I've replaced a 20-person B2B sales.
- 5:29Spencer Tahill: No, you didn't.
- 5:31Spencer Tahill: And uh sometimes you can, you can do it.
- 5:34Spencer Tahill: You know, like we've done cool things with Slack integrations and you know data across the world, but these types of linear flows, and I'm talking, you know, you have you know, single input, single output, you've got multiple inputs, multiple outputs.
- 5:51Spencer Tahill: The more complex you go, the more essential AI is going to become.
- 5:56Spencer Tahill: So when you build a workflow, my definition of a workflow is A, B, C.
- 6:01Spencer Tahill: You have input, you have some sort of transformation step or processing, and then you've got C, you've got an output.
- 6:07Spencer Tahill: And then you have an AI enhanced workflow, which is you have an input or you have multiple inputs, and then you have processing, the orchestration layer, and then the thinking that is going on, instead of saying, you know, if A put one, if B put two, if there's some sort of lead scoring and the lead scores have different weights, and you know, you have to calculate it on different weighted.
- 6:32Spencer Tahill: It's a really good use case for AI if it's way too complex for formulas.
- 6:37Spencer Tahill: In that instance, that becomes an AI-enhanced workflow.
- 6:40Spencer Tahill: Where the industry is going now, and where there's a lot more confusion, is that there's workflows, AI-enhanced workflows, AI agents, and then there's MCP, which is model context protocol.
- 6:52Spencer Tahill: So there's it's all different.
- 6:54Spencer Tahill: And I would say the more, the more inputs you guys you have, the more context you have, the more transformation steps you have, and the more output possibilities you have, yeah, the more the more AI is gonna accelerate that.
- 7:09Spencer Tahill: So that's how I define it.
- 7:11Spencer Tahill: Workflow is just gonna be linear or it's gonna be a single path input, and then it's gonna do a formulation about output.
- 7:17Spencer Tahill: And AI powered workflow is gonna do a lot more thinking, thinking that's been behind the scenes.
- 7:24Spencer Tahill: Right.
- 7:25Spencer Tahill: Yeah.
- 7:26Michael Hartmann: Yeah.
- 7:26Michael Hartmann: And it's uh you brought up MCP, which I still haven't quite wrapped my head around, but we can maybe we'll get there in the course of the conversation.
- 7:34Michael Hartmann: But yeah, I want so this leads really to one of my next questions is that one of the steps that you think of towards I think a lot of people struggle with like, where do I start?
- 7:44Michael Hartmann: Right.
- 7:45Michael Hartmann: I want to do like I want to AI, I need to learn AI, I need to do something.
- 7:49Michael Hartmann: And but what what one of the things that was interesting to me is when we talked before, you said one of the first steps towards doing automation, especially AI enabled or AI powered, whatever you want to call it, is identifying the things that piss you off, right?
- 8:04Michael Hartmann: So I think that was a phrasing you used, but like what do you what did you mean by that?
- 8:08Michael Hartmann: And like I how like maybe expand on it a little bit.
- 8:13Spencer Tahill: Yeah.
- 8:13Spencer Tahill: So um some people know this about me, some people don't know about me, but um, I I am at the end of the day, I want to do the least amount of work possible.
- 8:22Spencer Tahill: And that's probably not a very professional thing to say.
- 8:24Spencer Tahill: But I'm an Amazon, right?
- 8:26Spencer Tahill: But but you know what?
- 8:28Spencer Tahill: Like it's a game of efficiency and it's a game of focus.
- 8:32Spencer Tahill: And when I used to work for a few different startups, even when I was like freelancing, it was all about trying to like optimize my day.
- 8:39Spencer Tahill: If I can try to like build something that takes somebody eight hours to do and I can do it in two, oh my god, like you know, the value goes you know up the ladder, right?
- 8:47Michael Hartmann: Right.
- 8:48Spencer Tahill: When I start to think about when I want to not do something anymore, it's it's more philosophical.
- 8:55Spencer Tahill: It's just like, I just don't have I don't have a great feeling, you know.
- 8:58Spencer Tahill: Like I don't want to wake up, I don't want to turn the coffee machine on at six in the morning to have coffee at six and six thirty in the morning, right?
- 9:03Spencer Tahill: Like you press a timer on your coffee machine the night before and then it's done, right?
- 9:07Spencer Tahill: That's like very simple use case.
- 9:09Spencer Tahill: When you start talking about it in like the business aspect, everybody has different pain points.
- 9:14Spencer Tahill: And I'll get into that later, but I start by looking at the real ground level threat of what do I just hate doing every day?
- 9:22Spencer Tahill: Like, yeah, okay, let's say I'm a sales guy and I hate picking up the phone and dying.
- 9:27Spencer Tahill: Okay, maybe that's not the right, you know, industry for you to be working in.
- 9:31Spencer Tahill: Right.
- 9:31Spencer Tahill: But but if you if you're a sales guy and you hate going into Salesforce and HubSpot after every call, then what do you do?
- 9:39Spencer Tahill: Like if you're a really good closer, but you're just terrible at just like putting stuff in this ARM, okay, that's your pain point.
- 9:46Spencer Tahill: Let's solve that, let's automate that first so that you can focus on what you're good at and you guys can produce more, you know, you can be more efficient at your job.
- 9:54Spencer Tahill: So that's why I always ask it's like what pisses you guys off the most?
- 9:57Spencer Tahill: Ah, you know, I just want to get on calls all day.
- 9:59Spencer Tahill: I just want to close, you know, make whatever, every whatever everybody wants to do what they want to do.
- 10:03Spencer Tahill: But I start by looking at it as like, I don't like taking notes.
- 10:09Spencer Tahill: Very simply.
- 10:10Spencer Tahill: Like I enjoy the process of listening, I enjoy the process of conversing, but I don't, I'm not able to focus fully on you, Michael, if I'm just like, you know, off to the side writing my notes.
- 10:21Spencer Tahill: So now you see all these AI note takers, and that's why I use, you know, like something like Fireflies or Fathom to be able to take my notes for me, because then I can focus on what I'm good at.
- 10:30Spencer Tahill: So I've gotten rid of everything that I that annoys me.
- 10:32Spencer Tahill: Now I can 100% give my focus to you.
- 10:35Spencer Tahill: So I think like when you start doing, when you start making automations on the things that, you know, the processes that piss you off every day, that's where you should start.
- 10:44Spencer Tahill: That's the zero to one.
- 10:45Spencer Tahill: Get the idea of like what you don't want to do.
- 10:47Spencer Tahill: Because eventually, once you get all the stuff out of the air that you don't want to do, the path is gonna light up.
- 10:53Spencer Tahill: You'll be like, okay, this is this is the direction we want to take it.
- 10:56Spencer Tahill: This is what I want to build.
- 10:57Spencer Tahill: And then you bring somebody in to build that or you piss yourself.
- 11:00Michael Hartmann: Like I and I'm with you, like the things that I don't like to do, but like to some degree, there's every job right now has things that we don't like to do, right?
- 11:08Michael Hartmann: You brought a salesperson, I think in the marketing context, the one that pops to my head is you know cleansing and you know getting lists cleaned up to be able to upload into a marketing automation platform or something like that, right?
- 11:21Michael Hartmann: Which is necessary in a lot of cases, right?
- 11:24Michael Hartmann: And it's an important piece of work.
- 11:26Michael Hartmann: But like I'm sure there's a handful of people out there who actually enjoy that work.
- 11:31Michael Hartmann: I'm not one of those, and I would love absolutely love to just replace that in some automated way.
- 11:38Michael Hartmann: So I think it feels like there's a little bit of like I what I'm hearing from you is not just like things that annoy you and bother you, but like those that are still important, but are like things you don't enjoy doing.
- 11:49Michael Hartmann: Maybe it's a slight, slight tweak from what you were saying.
- 11:53Spencer Tahill: Yeah, it's it's uh if you've ever heard of the eisen, I think it's called the Eisenhower matrix, it's uh skill and will, and or it's gonna be something along between is it urgent and important?
- 12:04Spencer Tahill: And it wherever it falls on that matrix, if you have to delegate that work, you should automate it.
- 12:12Spencer Tahill: You should automate it to the best of your ability.
- 12:14Spencer Tahill: You should save a hundred percent, you should save 80% of your brain function to focus on the tasks that are the most important and the most urgent because nobody's gonna be able to replace that, right?
- 12:23Spencer Tahill: But if you're how I got into it, very true story is you know, my girlfriend at the time was a virtual assistant, and I was trying to figure out what does a virtual assistant do so that I can work from my laptop.
- 12:35Spencer Tahill: That's how it all started.
- 12:36Spencer Tahill: And I was like, oh, they, you know, in my silly head, like, you know, you book the calendar meetings, you you keep an eye on the email, you keep an eye on the Slack, you you maybe take a look at the car reservations, you know, you do your job.
- 12:48Spencer Tahill: And I was trying to figure out a way, okay, like what if I just got a notification from all of these different platforms in one place at every, you know, all summarized in one place.
- 13:00Spencer Tahill: Would it make my job easier as like a person?
- 13:03Spencer Tahill: So I built something for my own inbox that that does that for me.
- 13:06Spencer Tahill: So um these annoying things, the repetitive aspects, like how many times a day, how many times an hour, how many times a month are you doing it?
- 13:16Spencer Tahill: What is the impact that it's gonna have in the short term and the long term?
- 13:19Spencer Tahill: I think this is the way that you could start thinking about kind of like going from manual processes, you know, like you just said, like enriching your CRM.
- 13:27Spencer Tahill: That is a pain in the butt.
- 13:28Spencer Tahill: Um, that is that is something like people come to me for.
- 13:32Spencer Tahill: It's like we have 400,000 contact records in HubSpot.
- 13:37Spencer Tahill: What do we do with them?
- 13:38Spencer Tahill: Oh, there's a strategy.
- 13:40Spencer Tahill: Let's go pay for Zoom info, let's go, you know, use HubSpot Breeze to enrich it, right?
- 13:45Spencer Tahill: That's like kind of like the whole thing now.
- 13:47Spencer Tahill: But to do it in a sequential order and to do it in a way that you can actually get money out of that project is a completely different strategy.
- 13:56Spencer Tahill: So I think like before you do any automation, before you do any like thing, you should probably think about the strategy before you do that and how how you might make money back, you know, if that makes any sense.
- 14:07Michael Hartmann: Yeah, it's like there's a like uh I mean, I'm people with long time listeners know I'm a big fan of having sort of thinking in the financial terms, right?
- 14:16Michael Hartmann: So there's an ROI, right?
- 14:17Michael Hartmann: Like whether that investment is a time and effort one or actual dollars, like doesn't really matter.
- 14:22Michael Hartmann: It's like are you gonna get more back from that in the long term or whatever time horizon you're looking at, right?
- 14:30Spencer Tahill: Exactly.
- 14:31Spencer Tahill: It and it and that's not even just talking about I do the same thing.
- 14:35Spencer Tahill: It exactly what you just said.
- 14:36Spencer Tahill: It it is everything has to be ROI adjusted, whether that's gonna be you're gonna build an automation, you know, somebody comes to me like, oh, what is the, you know, we're gonna pay $10,000, $20,000, $30,000 for something, you know, monetary value, but what is the actual realized ROI?
- 14:54Spencer Tahill: Okay, within three months, you're gonna your SDRs are gonna be spending 35% less time because we've timed them, we've looked at their data entry times, we've averaged it.
- 15:05Spencer Tahill: That's gonna be like the time that they save.
- 15:08Spencer Tahill: But what is the percentage of their salary that you're paying them that's gonna be freed up?
- 15:13Spencer Tahill: What is the 35-25% of their salary that they're gonna be able to go and focus and do something else?
- 15:17Spencer Tahill: That's your margin, that's your ROI right now.
- 15:19Spencer Tahill: And that's how I, you know, that's a whole different story.
- 15:22Spencer Tahill: But yeah, I agree completely.
- 15:24Spencer Tahill: It's everything needs to tie back to revenue.
- 15:27Michael Hartmann: It's interesting.
- 15:27Michael Hartmann: The other like the tie back when you bring in other, like now you if you're if you're opening up resources to be able to apply their skills and experience in in ways that are more productive or more uh provide better return for your organization or whatever, right?
- 15:45Michael Hartmann: It seems like it's a really good mesh with I don't know if you are you familiar with Strengths Finder.
- 15:50Spencer Tahill: No.
- 15:51Michael Hartmann: Okay, so this is like one of those personality test kind of things, right?
- 15:54Michael Hartmann: But the the real uh the precursor to that book and then the the actual assessment was a book called Um First Break All the Rules.
- 16:04Michael Hartmann: And it was kind of a one of the first management books that I read where I was like, oh, they like it actually really changed how I thought about things.
- 16:09Michael Hartmann: And the basic premise is a lot of modern um like management was about right uh fixing people's weaknesses, right?
- 16:20Michael Hartmann: Uh and this sort of flipped it on its head and said, really, what you should be doing is take advantage of people's strengths and minimize the impact of their weaknesses.
- 16:28Michael Hartmann: So like it's not ignoring it.
- 16:30Michael Hartmann: But where I'm going is like this meshing of like what do I like what's like what could I get?
- 16:36Michael Hartmann: If I if I could free up somebody who's got a strength in a certain area, but they're not able to take advantage of it because they're spending a lot of time on stuff that they don't like, they're not good at, but it's a necessary part of the job, um, you start to free up more of that time.
- 16:49Michael Hartmann: Now you can apply that strength to something else where it's a bet going to be a better return and it's probably gonna be more satisfying for them.
- 16:55Michael Hartmann: I think this is a really interesting way of thinking about it.
- 16:59Michael Hartmann: That like literally in real time, I'm like it's hitting me.
- 17:03Spencer Tahill: So uh this is like very common.
- 17:07Spencer Tahill: And I know I know we're kind of tangenting, but I I think it's very common when people will come to our door and they'll say, We want to take call intelligence and we want to figure out what competitors are being mentioned.
- 17:24Spencer Tahill: Okay, that's that's a whole call intelligence workflow, you know, that has to be built out.
- 17:29Spencer Tahill: But why?
- 17:31Spencer Tahill: Why would that, you know, like what where where's that revenue gonna be made back?
- 17:36Spencer Tahill: What type of insight does that data give you?
- 17:40Spencer Tahill: And what are you gonna do with that insight?
- 17:41Spencer Tahill: What's the action you're gonna take?
- 17:42Spencer Tahill: And and I think being able to steer the course before you start building is very, very important.
- 17:49Spencer Tahill: Because uh what I see a lot, and I don't I'm not sure if you see this a lot, is like sales works in their little bubble, and then you've got revenue team and marketing team kind of like sitting over here, and but nobody had nobody actually sees each other's data, and I think it's a huge problem.
- 18:03Spencer Tahill: So it's I think it's about when you're starting to think about like making AI workflows and agents and stuff like this inside of a like you know, I work in mid-market, so if you have to connect the teams, you have to make everybody play nice together somehow, and it's not an easy job to do.
- 18:21Spencer Tahill: Um, it takes a lot of coordination, but there's a lot of issues with the industry.
- 18:26Spencer Tahill: It's that like you just build one thing and call it an AI workflow, and it that's just not what it is, you know.
- 18:32Michael Hartmann: Yeah, I think you're hitting on something that really like I think it I've always thought it's important to go like what's the end result?
- 18:38Michael Hartmann: Like, what do you want to do with what you're asking?
- 18:41Michael Hartmann: Very often how I it pushbacks maybe not the right way, but I I want to understand the why when I like somebody has come to me as a marketing apps leader.
- 18:49Michael Hartmann: Um, and I think example I could use is actually before marketing apps was really a thing, but I ran website for a big global semiconductor company.
- 20:09Michael Hartmann: And when the product marketers came, he's like, hey, we want to get like we want to get asked, you know, put up a form and ask for feedback from customers.
- 20:16Michael Hartmann: Great, easy.
- 20:17Michael Hartmann: What are you gonna do with it?
- 20:19Michael Hartmann: Because I'm gonna tell you right now, if you open that up and people start providing that feedback, they're gonna want to know what did you do?
- 20:25Michael Hartmann: And they hadn't thought through that part of it.
- 21:28Michael Hartmann: And I think it feels like that is gonna needs to be even more and more important with these things where they're less deterministic if you're especially automating things, right?
- 21:39Michael Hartmann: Where you've got some sort of LLM or AI sort of component in the middle of it where you don't always know exactly what's gonna come out of the other side.
- 21:51Spencer Tahill: Yeah, I think um I think that comes down to all, it just comes down to operations.
- 21:56Spencer Tahill: And it comes down to processes, it comes down to nailing goes down, it comes down to hey, if you're gonna do one, this second thing is gonna happen.
- 22:05Spencer Tahill: If you're gonna do A, B is gonna happen.
- 22:07Spencer Tahill: Um, but it's the edge cases.
- 22:10Spencer Tahill: Like humans, you know, like we make mistakes, but then we get reprimanded and we're like, oh, we're not gonna make that mistake again.
- 22:18Spencer Tahill: But the the the bigger issue that I'm seeing now is that like when you start to replace humans with machines, is how do you make sure that machine handles that edge case a different way the next time?
- 22:29Spencer Tahill: And I think like you can tell somebody what to do and they're gonna learn not to shove the circle into the square hole, but a machine doesn't know inherently like what they did wrong.
- 22:40Spencer Tahill: So you have to tell it what to do before it does it, before it tries to do that thing, right?
- 22:46Michael Hartmann: Yeah, you have to give it it can't be totally free form.
- 22:50Michael Hartmann: Yeah, you need to provide it some rules.
- 22:52Michael Hartmann: It's interesting because I hadn't really like my own experience, just personal level of using ChatGPT is I've had a lot of several times where I've been frustrated, where I get an output that is say 80 to 90 percent of what I was hoping for, and I give it very specific instructions, go change this or that on the output, and it completely changed, like it does some something completely different, like it actually goes away from where it was going.
- 23:17Michael Hartmann: And you know, that's like my fear would be if I'm doing something that's automated and that's like something like that's in the middle of it, that if it like call it an edge case, call it a hallucination, call it whatever, right?
- 23:31Michael Hartmann: That's a risk that we have to deal with.
- 23:34Spencer Tahill: Yeah, and I I think that's gonna happen anywhere.
- 23:37Spencer Tahill: It's you're gonna have that in organizations, you're gonna have that, you know, in in user, you know, a user level, you know, things like that, right?
- 23:45Spencer Tahill: Um when I started using GPT, you know, no hate here.
- 23:50Spencer Tahill: Like OpenAI did a great job by getting GPT into the hands of the consumers, the consumer market, right?
- 23:58Spencer Tahill: But behind the scenes in the organization level, like there's already intelligence layers that are crunching and formulating and you know, like spitting out intelligence and insights.
- 24:07Spencer Tahill: Um on the consumer level, I was not a power user at all.
- 24:12Spencer Tahill: I was a power user of workflows, I was a power user of like, you know, formulas and API calls and you know, sequences and HubSpot and stuff like this.
- 24:22Spencer Tahill: But damn, I hated it.
- 24:24Spencer Tahill: I hated the process of like interacting with it because, like you just said, it's like a lot of people get frustrated.
- 24:30Spencer Tahill: I was one of those people, you were one of those people, you type something in and it's like, uh-huh.
- 24:35Spencer Tahill: I'm gonna go get that thing over there, and it's not anything relevant to what you do.
- 24:39Spencer Tahill: And and now it's been like two or three years since we've had like this kind of like this power in our in our hands to be able to tell it what to do.
- 24:47Spencer Tahill: And now you can just like you can do so many crazy things.
- 24:50Spencer Tahill: Yeah.
- 24:50Spencer Tahill: Um but it takes so much precision.
- 24:54Spencer Tahill: Like if you want it to do a very specific thing, you need to be super precise about it.
- 24:59Spencer Tahill: And and the problem is actually not being able to produce the quality result once.
- 25:03Spencer Tahill: Um I feel like on a on a consumer level, the hardest thing to do is to aim for consistency.
- 25:10Michael Hartmann: Yeah.
- 25:11Spencer Tahill: Can it do it again?
- 25:12Spencer Tahill: Can you do it again, basically, right?
- 25:15Michael Hartmann: Yeah, yeah, no, it's um I think I heard someone the other day talking about like training in AI on, especially if you're thinking about like for your purposes and like as a it's almost like um teaching a baby, right?
- 25:32Michael Hartmann: Because that's kind of where it starts.
- 25:34Michael Hartmann: It doesn't really though it could it's just what's weird is it actually speaks your language already, right?
- 25:39Michael Hartmann: Which is not what a baby can do.
- 25:40Michael Hartmann: So like that's that part's different.
- 25:42Michael Hartmann: So it's a very interesting one.
- 25:43Michael Hartmann: So uh you kind of hinted at something.
- 25:46Michael Hartmann: So one of the other things you and I talked about before was um systems thinking and and how that needs to come into play.
- 25:55Michael Hartmann: Um, when you think about AI and adoption of AI.
- 25:58Michael Hartmann: Like, what did you like?
- 25:59Michael Hartmann: So what did you mean by that?
- 26:00Michael Hartmann: Like what like from your from your frame, maybe start.
- 26:02Michael Hartmann: Like, what is systems thinking from your perspective and then why you think it's so important?
- 26:07Spencer Tahill: So system thinking is imagine you have, I'll just put it very eloquently.
- 26:13Spencer Tahill: Imagine you have a car, right?
- 26:15Spencer Tahill: And each component of that car makes up the whole machine, right?
- 26:19Spencer Tahill: You've got the wheels, you've got the axles, you've got the engine, you've got the motor, like whatever.
- 26:23Spencer Tahill: If you take the wheel off the car, can it run?
- 26:28Spencer Tahill: Maybe on three wheels.
- 26:30Spencer Tahill: Depends on what you mean by run, but yeah.
- 26:32Spencer Tahill: Depends on what you mean mean by run, exactly.
- 26:35Spencer Tahill: But if you change one thing in a in a s in a machine, if you take that belt out, if you take the engine out, it's probably not going to work as well as you designed it to.
- 26:45Spencer Tahill: System thinking is the idea in my head that before you change one component, you should look at the ramifications of the impact across the whole system.
- 26:58Spencer Tahill: And you see this a lot in like workflows, especially in like, especially in Zapier, right?
- 27:04Spencer Tahill: It's like you've got one Zapier account, you've got 20 folders in there, and you've got 18 different zaps in there that are all floating around.
- 27:12Spencer Tahill: Ideally, they're organized.
- 27:14Spencer Tahill: If you change one custom field inside of HubSpot, how many of those zaps break?
- 27:19Spencer Tahill: If you if one of your interns goes in and breaks one thing, how many of those things get you know jeopardized?
- 27:25Spencer Tahill: And I think this idea of like systems thinking and the way that you can approach making a minor impact is that that one thing can domino affect down the whole system and shut it down.
- 27:35Spencer Tahill: And I'm sure that has happened to many organizations, you know, but I've seen it.
- 27:41Spencer Tahill: Um, but this idea of don't change anything until you're know until you know don't change micro until you understand the macro.
- 27:50Spencer Tahill: I think this is the idea, and this is how I try to formulate like what is systems thinking?
- 27:55Spencer Tahill: Understanding the whole embodiment of an organization, of a system, not just HubSpot, not just the ad campaign, but the whole creative process, looking at like creating the copy, go, you know, go to Canva, post it on LinkedIn.
- 28:09Spencer Tahill: Okay, when you post it on LinkedIn, it's an ad, then you boost it.
- 28:12Spencer Tahill: But how do you boost it?
- 28:13Spencer Tahill: Do you boost it with a UTM?
- 28:14Spencer Tahill: Where does that UTM get picked up?
- 28:16Spencer Tahill: Does it go to HubSpot?
- 28:17Spencer Tahill: Where does it go?
- 28:17Spencer Tahill: You know, like tracing the line back and being able to map out the whole system like a skeleton to be able to see, yeah, if I change this little thing.
- 28:28Spencer Tahill: That's how it changes everything, and that's the impact.
- 28:32Michael Hartmann: Got it.
- 28:32Michael Hartmann: So this goes back, as we had a guest on uh it's probably been a year ago now, who talked kind of taught me about something called the Kinebin framework, which is a Welsh word.
- 28:41Michael Hartmann: So it's done it's spelled C-Y-N-E-F-I-N, I think I'm close.
- 28:46Michael Hartmann: But it's pronounced very differently.
- 28:47Michael Hartmann: Anyway, long story short, like this the it differentiates across different kinds of problems.
- 28:52Michael Hartmann: And one is the the one that's most relevant here is complicated versus complex, right?
- 28:57Michael Hartmann: And what you described, like in fact, car is a great example of complicated, right?
- 29:01Michael Hartmann: Where um if you have enough knowledge and training, you can understand how it all fits together.
- 29:06Michael Hartmann: But if there's a problem, you can usually kind of figure out and fix it, right?
- 29:10Michael Hartmann: Because it's a singular problem, uh, even if it has downstream effects.
- 29:14Michael Hartmann: The difference in a complex one, which is feels like we get into very quickly it uh, but doesn't always like we don't always know the different like when we're going from complicated to complex, is it like you actually can't fully understand it?
- 29:27Michael Hartmann: And I think this is where AI starts to come in, where it's like you have one input over here and you change it, you actually don't know what the impact is gonna be in that one case, right?
- 29:37Michael Hartmann: And now you start at like you've brought like you got a new case, the input's slightly different.
- 29:41Michael Hartmann: Do you get the same kind of result?
- 29:43Michael Hartmann: Maybe, maybe not.
- 29:44Michael Hartmann: And that's where I suspect we're headed is that we're gonna be moving from complicated, where it's not easy, but you can kind of figure it out to complex.
- 29:54Michael Hartmann: Um, and it will be really, really difficult to figure out if something changes and you don't know because it's all sort of uh hidden inside this quote black box.
- 30:07Spencer Tahill: Nail that you nailed it.
- 30:09Spencer Tahill: Okay.
- 30:10Spencer Tahill: Yeah.
- 30:10Michael Hartmann: But I mean, but understanding how those pieces fit together is important, right?
- 30:14Michael Hartmann: I I like I don't want to discount that.
- 30:15Michael Hartmann: Like so that systems thinking I I now that you described, right?
- 30:19Michael Hartmann: Yes, it's gonna be really important.
- 30:21Michael Hartmann: Um, but it feels like there's gotta be this um acknowledgement that even with systems thinking in this new kind of world we're we're either in or headed towards, like it's hard to tell.
- 30:34Michael Hartmann: There's gonna be stuff that is gonna be really hard to figure out.
- 30:38Michael Hartmann: Um both to set it up and to like figure out if something doesn't I wouldn't even say broke breaks, but like produces an output we don't expect.
- 30:51Spencer Tahill: That and that is probably the hardest part of my job is edge case handling.
- 30:55Spencer Tahill: I can I feel confident telling you like I can build a research prompt or I can build a prompt in two or three days, right?
- 31:02Spencer Tahill: Like stress test and everything like this, but it is the testing period that I tell people about I said, like, listen, like I could build it, but we're gonna test it for two weeks because I need to monitor the edge cases before I hand that over to you.
- 31:15Spencer Tahill: And it's just I think you try to some so many people try to go so fast on burning something or implementing something that they don't do the proper QA, and then you get issues, right?
- 31:26Spencer Tahill: And and that's that's the way of the world now where everybody's going so fast, AI is here, it's at your fingertips, and you're like, it's gonna solve everything, you know, like a a brand new cooking pan is not gonna make you a better cook, right?
- 31:39Spencer Tahill: But like, but it's gonna help.
- 31:41Spencer Tahill: Maybe a sharper knife will help, but it's it's inherently not going to it's not gonna fix the root issue.
- 31:48Spencer Tahill: And I think if if a lot of people more people try to figure out you know, root cause analysis, the systems thinking approach, is they will understand that they have a lot different problems than they actually think they do.
- 31:59Michael Hartmann: So yeah, I mean, do you so do you think that's one of the big mistakes people make when they're trying to adopt AI?
- 32:04Michael Hartmann: They're moving too quickly, they're not like what yes, like one of the biggest red like things, things that kind of call bear traps, maybe like the people you watch out for when they're trying to adopt AI.
- 32:15Spencer Tahill: I think when I I don't really get it so much anymore across you know, across my desk, but a few months ago, I mean, even I would say even the past like six months, right?
- 32:30Spencer Tahill: What I've noticed is that more and more people that come to me, they know like, okay, we're gonna do something complex.
- 32:35Spencer Tahill: Going back to complexity, yeah.
- 32:37Spencer Tahill: Um, I want to go, like, for instance, I want to build a Slack bot that connects one community to another community, but also routes into Adio and uses call intelligence and everything like this to make content.
- 32:48Spencer Tahill: Whatever.
- 32:48Spencer Tahill: That's very complex, and you need different API nodes.
- 32:52Spencer Tahill: But I think a lot of people that come and they try to use a new tool.
- 32:57Spencer Tahill: Tool.
- 32:57Spencer Tahill: They use this like shiny object syndrome where it's like you see AI, they're like, oh my God, like I see all these things on LinkedIn.
- 33:04Spencer Tahill: This is like the stemming of the problem.
- 33:06Spencer Tahill: Is that they see everybody else doing something just super complex and it looks complex and it looks crazy, and it somewhat sounds like it can solve their problem.
- 33:18Michael Hartmann: Right.
- 33:19Spencer Tahill: But they don't realize that every problem is unique.
- 33:24Spencer Tahill: And I think like that's the very like pinnacle of the problem, like, not even regarding AI adoption.
- 33:30Spencer Tahill: I think it's how you how you hire people.
- 33:32Spencer Tahill: Like, why are you hiring them?
- 33:34Spencer Tahill: Is the problem unique?
- 33:35Spencer Tahill: Are you making a workflow or something like this?
- 33:37Spencer Tahill: So I think understanding not being able to understand the situation before you try to adopt something like AI and not being able to define the problem properly and not being able to define possible solutions to implement is a huge problem.
- 33:53Spencer Tahill: And and I don't I don't touch that because it's it makes work very difficult when you've got a project manager.
- 34:00Spencer Tahill: It makes it makes everything slow down and you're like, oh wait, uh-oh, I built something and you didn't want that.
- 34:06Spencer Tahill: Everybody's been there, you know, and then it costs everybody time.
- 34:09Spencer Tahill: So I think it's rather just slow down, define the problem, define the desired result, work backwards, look up any roadblocks, try to, you know, set up the task matrix, just who's gonna handle what based on their strengths, delegate, start at the beginning, and then say, okay, this is gonna take this much time.
- 34:29Spencer Tahill: And then that's done.
- 34:30Spencer Tahill: And I think I think companies and people that fail to do this in this order, just bound to fail.
- 34:35Spencer Tahill: And I it just it's too fast, not enough thinking, it's just failed.
- 34:40Michael Hartmann: It's the whole the whole like to try to eat an elephant in one bite is like doomed to fail, right?
- 34:45Michael Hartmann: Um so what I'm curious what your take is.
- 34:48Michael Hartmann: I think I think when we talked before, you're switching gears a little bit, is that yeah, a lot I think there's a lot a lot of fear out there that AI is going to replace people.
- 34:58Michael Hartmann: And I think if I remember right, your take on it was that AI wasn't gonna replace people, is gonna replace people who didn't know how to use AI.
- 35:05Michael Hartmann: Is that has it is that a fair assessment?
- 35:08Michael Hartmann: And like has it evolved at all since then?
- 35:11Spencer Tahill: Has it evolved?
- 35:12Spencer Tahill: It's probably I I'll probably double down on that statement.
- 35:15Spencer Tahill: It's we are apex predators as humans because we know how to use tools and because we know how to think ahead.
- 35:26Spencer Tahill: That is that is my double-down statement.
- 35:29Spencer Tahill: AI is a tool that allows you to do more way past your cognitive abilities.
- 35:36Spencer Tahill: I'm not a P, I don't have a PhD in microbiology, I don't have a PhD in you know, mathematics or anything like this.
- 35:44Spencer Tahill: But these LLMs that, you know, OpenAI and Propic, and you know, any of the other LLMs out there that are amazing, they've been trained to do this.
- 35:55Spencer Tahill: I, you know, they've they've gone through the steps, they've gone through the data, they've there you have you have a billion perspectives and a billion different data points outside of your own cognitive ability to be able to tap into.
- 36:10Spencer Tahill: You just need to ask the right questions.
- 36:13Spencer Tahill: And I think the people that know how to ask the right questions, the people that are inherently inquisitive about new tools that are, you know, how to become a better prompt engineer, how to use AI better.
- 36:25Spencer Tahill: This is step number one.
- 36:26Spencer Tahill: It's like, don't ever just use GPT, login and it's gonna say, hey, help me do this.
- 36:31Spencer Tahill: No, no, no.
- 36:32Spencer Tahill: Go to GPT, open two GPT windows, say, help, I want to do this, write a better prompt for me to help me do this that I can copy and paste.
- 36:41Spencer Tahill: Do this.
- 36:41Spencer Tahill: It's gonna be way smarter than you about prompting.
- 36:44Spencer Tahill: Take that, put it in the second window, do your task, and then after you're done with your tasks, say, This is what I did, this is the prompt I used, this was the original prompt that I wanted to use, but help me learn how to go from my original idea, my original intelligence level through the process and teach me how to do it at a higher standard.
- 37:06Michael Hartmann: Yeah.
- 37:06Spencer Tahill: And I think like the people that know how to do this are gonna soar over the people that that don't know how to ask the right questions or don't adopt AI.
- 37:15Spencer Tahill: The people that just you know don't adopt it, I think it's quite frankly, I think it's negligence.
- 37:20Spencer Tahill: And uh it's a tool.
- 37:22Spencer Tahill: You don't need to use it to be great at what you do, but I think if you're if you're trying to do more and accelerate, it's clearly the next step.
- 37:30Spencer Tahill: And yeah, I saw that's just my my forte, Anna.
- 37:33Michael Hartmann: I saw somebody post on LinkedIn, I don't remember, but it was recent, like the last week or two.
- 37:38Michael Hartmann: Somebody was saying, like kind of echoes what you're saying, AI is a tool.
- 37:43Michael Hartmann: And he's like, as a writer, like my right, like I didn't lose the ability to write when I started going from pen and paper to a typewriter.
- 37:52Michael Hartmann: I didn't lose my ability to write when I like from that to a computer to no.
- 37:55Michael Hartmann: It's but it's it and it didn't change when I started using grammarly or something like that.
- 38:01Michael Hartmann: Like it just helped me get better and more efficient at it, it's as opposed to atrophying, which we we could talk about.
- 38:09Michael Hartmann: Like, I I think it's gone by the wayside now.
- 38:11Michael Hartmann: There was all those headlines about like AI causes brain atrophy, but I think that was just headline fodder, but I don't know what your take was on that.
- 38:21Spencer Tahill: That's the MIT study, right?
- 38:22Spencer Tahill: Yeah, yeah.
- 38:23Spencer Tahill: The one I was covered a few months ago.
- 38:25Spencer Tahill: Yeah, I think uh that's a whole discussion.
- 38:30Spencer Tahill: Um but I had read I had read the synopsis of that study uh like a few whenever it came out, a few months after it came out.
- 38:40Spencer Tahill: Um for the people that didn't read it, it was like a study by MIT with a control group, and it was it was it was roughly talking about like people that use AI and the people that don't use AI and kind of like the adoption and the competence levels there.
- 38:53Spencer Tahill: Um, but roughly uh the synopsis there that I would sum up is that the way that it was presented in the news that like GPT users have you know lower competence of like XYZ.
- 39:05Spencer Tahill: Um that's canon fodder, but I would say that there was like my personal beliefs aligned with that article in the sense that somewhere in there it said something along the lines of like higher competence users of AI will lapse lower competence users at a baseline level, but lower competence users will fail to learn if they continue at this pace of low learning like aptitude.
- 39:32Spencer Tahill: Yeah, long story short, lazy people get lazier because they depend on AI as a shortcut, but the higher performers, the ones that are more inquisitive, will use that as an acceleration jump in their learning.
- 39:42Spencer Tahill: So it's that's how I understood it.
- 39:45Michael Hartmann: Yeah.
- 39:46Michael Hartmann: Yeah, I'm with you.
- 39:47Michael Hartmann: And I it it's consistent with other stuff I've heard uh more in the education realm, um, where the highest performing kids who started using AI as a part of their learning process continue to like distance themselves and those that didn't use it, and and those who who were not performing as well and used it also improved.
- 40:12Michael Hartmann: They just their improvement was less, right?
- 40:14Michael Hartmann: So it's not quite what you were saying, but it's very similar kind of results.
- 40:17Michael Hartmann: Um I'm sure there's studies there.
- 40:19Michael Hartmann: I just I that one I've heard just anecdotally about.
- 40:22Michael Hartmann: Um okay, so I'm trying to decide where to go here.
- 40:28Michael Hartmann: So you we kind of talked about the system prompt, you know, like using the tool to help you write better prompts.
- 40:34Michael Hartmann: And um one of the things you you and I talked about, and I still feel like I don't quite understand, is the difference between I think you called it a system prompt and a user prompt.
- 40:44Michael Hartmann: Like so is this well, I'll I'll let you explain.
- 40:48Michael Hartmann: What's the difference?
- 40:49Spencer Tahill: Yeah, so in in the context of a user-level interaction, this is it layman's term, you log onto your computer on your phone and you type in the GPT.
- 41:00Spencer Tahill: You're the user.
- 41:01Michael Hartmann: Yeah.
- 41:02Spencer Tahill: On the system side, that is inherently going to be how the system thinks, right?
- 41:09Spencer Tahill: It's going to be a blank slate.
- 41:11Spencer Tahill: You know, like imagine you as a human, just in a machine form, right?
- 41:17Spencer Tahill: You're a system as a person, your brain is a system, it's wired to think a specific way.
- 41:21Spencer Tahill: Your brain, it's wired.
- 41:23Spencer Tahill: You've learned all these different interactions, you know how to speak, maybe French, Italian, Spanish, English, whatever.
- 41:29Spencer Tahill: Maybe you have a maybe you have a doctorate, maybe you have you're in a bachelor's degree of engineering, maybe you're in business.
- 41:35Spencer Tahill: You have all of these unique different ways of connecting the synapses, right, in your brain.
- 41:40Spencer Tahill: Everybody is unique.
- 41:42Spencer Tahill: That's what makes us us.
- 41:44Spencer Tahill: But we're not a machine.
- 41:46Spencer Tahill: And if you want to replicate the way of thinking, like a machine, this machine thinking, quote unquote, right?
- 41:53Spencer Tahill: I'm gonna kind of like simplify.
- 41:54Spencer Tahill: A system prompt is the back-end design of what you're talking to.
- 42:00Spencer Tahill: A user prompt is what you're actually talking, like how you're interacting with the machine, if that makes sense.
- 42:07Spencer Tahill: So when you put in, hey, help me make do this research, you are prompting it as a user.
- 42:15Spencer Tahill: But you are prompting it as a user and not a machine.
- 42:18Spencer Tahill: The machine, the system, will think in a very specific way.
- 42:23Spencer Tahill: And and it's trained to do this.
- 42:25Spencer Tahill: So you have very specific use cases like I want you to act like a market researcher.
- 42:32Spencer Tahill: I want you to tell me how to make a podcast episode, but have you ever given it extra pot like context right before you tell it what to do?
- 42:39Spencer Tahill: Like, hey, I I want you.
- 42:41Spencer Tahill: This is contextual prompt prompting.
- 42:44Michael Hartmann: Yes.
- 42:44Spencer Tahill: The system prompt is behind the scenes.
- 42:48Spencer Tahill: So you can already give it all of the intelligence and all of the use cases.
- 42:53Spencer Tahill: You can fuse shot it, you can give it edge cases to look out for, you can tell it what to think.
- 42:58Spencer Tahill: So not a lot of models will, well, nowadays you can, but not a lot of models six, seven, eight months ago made it easy for you to actually design the system prompt.
- 43:09Spencer Tahill: You were only able to interact with it from a user prompt point of view.
- 43:13Spencer Tahill: So when you are building a workflow, you know, maybe you there's some listeners that, you know, understand API calls and stuff like this.
- 43:22Spencer Tahill: If you have the chance to put information inside of a system prompt window, this is where you're going to give it its thinking ability.
- 43:30Spencer Tahill: This is where you're going to give it full context.
- 43:32Spencer Tahill: This is how you're going to give it all of its edge cases, its limitations, and everything like this.
- 43:36Spencer Tahill: The user prompt is to drive the conversation and the contextual output forward.
- 43:41Spencer Tahill: So these are the two differences.
- 43:43Spencer Tahill: And I think like when you are, you know, you're just in your computer or you're on your phone and you're just trying to like find the answer to something, you're prompting it.
- 43:50Spencer Tahill: You've not designed it.
- 43:52Michael Hartmann: Right.
- 43:52Spencer Tahill: The first iterations of a system prompt came with like a fine-tuned LLM where you would kind of design it to you know handle these use cases to think in a specific way.
- 44:02Spencer Tahill: It's not thinking.
- 44:04Spencer Tahill: I mean, it is, but it's connecting synapses, it's connecting nodes of data together.
- 44:08Spencer Tahill: That's thinking.
- 44:09Spencer Tahill: That's a system prompt and I think those are two inherently different things that you can.
- 44:17Michael Hartmann: I was thinking like programming languages like that most people would use versus assembly language, but I don't think it's quite the right analogy there, right?
- 44:25Michael Hartmann: This is like um, if I understand it right, it's almost like um providing that baby, right, um as much ex like experiential stuff you can, right, and knowledge before you then go ask it to go either answer something or do something or whatever, right?
- 44:45Michael Hartmann: Is that the way more of a better way of thinking about it?
- 44:48Michael Hartmann: Yes.
- 44:48Spencer Tahill: I I would I would up the complexity a little bit, but on a on a back end, yes.
- 44:53Spencer Tahill: Uh imagine like you have that one friend that's really annoying that you just don't want to like talk to about anything, but you know that if you tell him about this other thing, that he's gonna be there for you or XYZ, right?
- 45:05Spencer Tahill: This is like a very complicated scenario.
- 45:07Spencer Tahill: But the system, that person is a system, and they're gonna handle stimuli in a very specific way.
- 45:14Spencer Tahill: When X occurs, that's gonna happen, right?
- 45:17Spencer Tahill: Because it's you've already tried to talk to that guy about you know, 20 times, you know, two in the morning comes over, whatever, you know.
- 45:24Spencer Tahill: Yeah, but the way the the consistency, like the training of this, that's that is like the personality, that is the way that the system operates.
- 45:33Spencer Tahill: It's the operating model, it's not like assembly, but the user input is the scenario.
- 45:39Spencer Tahill: So instead of instead of looking at it as like, oh, the you know, like you're gonna give the toy to the baby and it's gonna try to shove it through the box, whatever, right?
- 45:48Spencer Tahill: You teach it, you tell you tell it what to do and how to do it.
- 45:53Spencer Tahill: And when it doesn't do that, these things happen.
- 45:55Spencer Tahill: That's a system prompt.
- 45:57Spencer Tahill: And then when the scenario happens, when you when you kind of give it the scenario in the context of like what's going on at the user level, then it'll know inherently, oh, this scenario matches scenario 47 that I already have knowledge base about.
- 46:13Spencer Tahill: Let me handle it in this specific way.
- 46:16Michael Hartmann: Interesting.
- 46:17Michael Hartmann: Okay.
- 46:17Michael Hartmann: Yeah, I think I'm I'm getting that nuance now a little better.
- 46:21Michael Hartmann: So um I think I brought I mentioned the word hallucination at some point.
- 46:26Michael Hartmann: Um, and I think pretty much anybody who's used this stuff has experienced a hallucination from one of these LOMs.
- 46:32Michael Hartmann: But um so one of the things you and I talked about, you said that you suggest there's like five things that you suggest that you do with every LLM every time you try to go through something.
- 46:46Michael Hartmann: I don't know if this is a system prop or user promp, so maybe there's a different shade there.
- 46:50Michael Hartmann: But what are those five things uh um that we our listeners or audience could benefit from?
- 46:57Spencer Tahill: There are I'm gonna explain it terribly, but I'll do my best.
- 47:02Spencer Tahill: Um I always get it context, background, goals, constraints or limitations, and formats.
- 47:12Spencer Tahill: So if you can give it those five things, and in not it doesn't have to be in that order, but and you can look up on the OpenAI website and Anthropic COD website, like how to make a good system from, how to make a use good user prompt.
- 47:23Spencer Tahill: But if you give it context, you know, you are this, you have this data set, you you know, you're connected to the API and FBI, whatever, you've got all that data.
- 47:34Spencer Tahill: That's context.
- 47:36Spencer Tahill: Background, you've handled a hundred and a hundred thousand iterations.
- 47:40Spencer Tahill: These are the a hundred thousand results that you got that you produced.
- 47:43Spencer Tahill: Out of these a hundred thousand, only 50,000 of them should be produced, whatever, right?
- 47:47Spencer Tahill: That's that's gonna be the background of like, what is it?
- 47:50Spencer Tahill: I'm a senior marketer research, everything like this.
- 47:53Spencer Tahill: That's step two.
- 47:54Spencer Tahill: So context and background.
- 47:56Spencer Tahill: The third one is gonna be the goals.
- 47:58Spencer Tahill: So it's gonna be the directed output.
- 48:00Spencer Tahill: What are you trying to achieve?
- 48:01Spencer Tahill: Okay, I'm a market researcher.
- 48:03Spencer Tahill: I have act, I now I have access to Google, I have access to web search capabilities, I have access to NPI, you know, all of these different web pages, whatever.
- 48:13Spencer Tahill: But now my goal is to be able to crunch data.
- 48:16Spencer Tahill: I want to analyze the data, I want to synthesize the data, I want to triangulate the data sources, I want to see if there are any data discrepancies.
- 48:23Spencer Tahill: Okay, awesome.
- 48:24Spencer Tahill: Now the fourth step.
- 48:26Spencer Tahill: Now we have context, background, and goals.
- 48:28Spencer Tahill: Very clear the system knows what to do.
- 48:30Spencer Tahill: Then you have limitations or constraints.
- 48:33Spencer Tahill: This is the fourth one.
- 48:34Spencer Tahill: This is like what a lot of people forget, and this is leading into the discussion of like hallucinations, you what not to do.
- 48:41Spencer Tahill: Because you you summed it up perfectly.
- 48:45Spencer Tahill: You aren't on a GPT months ago and you were super frustrated because it produced something completely different.
- 48:51Michael Hartmann: Yeah.
- 48:51Spencer Tahill: What you had forgotten, what you had forgotten, exactly.
- 48:54Spencer Tahill: You had forgotten, and naturally I have too many times, is you didn't tell it to retain its memory.
- 49:02Spencer Tahill: You didn't tell you told it, you're like, hey, I want you to make these changes.
- 49:06Spencer Tahill: But then it did, hey, I'm gonna make these changes across the whole thing.
- 49:13Spencer Tahill: But you didn't, you but what would have been better is hey, great job.
- 49:18Spencer Tahill: Give it feedback, just talk to it like a person.
- 49:20Spencer Tahill: This is what I do.
- 49:21Spencer Tahill: I do Windows key H on my keyboard, or you open up speak to type.
- 49:25Spencer Tahill: There's a lot of speech to type paths, and I speak to it.
- 49:27Spencer Tahill: I give it feedback.
- 49:27Spencer Tahill: That's how I work, but with the limitations, do not do this, do not output a string, do not output an array, output rich text or you know, take out all the spaces, take out all the vowels.
- 49:40Spencer Tahill: I don't know why you do that, but do not do this, do not list it down because it's not gonna know unless you tell it.
- 49:47Spencer Tahill: A child isn't gonna know that the the letter of the alphabet starts with A unless you go A, B, C, D, you know, right?
- 49:55Spencer Tahill: So you have to tell it, you have to treat it like a child, right?
- 49:59Spencer Tahill: It doesn't know.
- 50:00Spencer Tahill: They're so smart now that most of them do, but if it's never seen the scenario before, then it's not gonna know, right?
- 50:06Spencer Tahill: So now we have context, background, goals, limitations.
- 50:09Spencer Tahill: Now you have the fifth one, which is almost the most important, which is the format.
- 50:14Spencer Tahill: So a lot of the form the formation of the output is the result.
- 50:18Spencer Tahill: That's the thing that you need to get ROI from, right?
- 50:21Spencer Tahill: Like if you're making a workflow or a program or transformation program, that's probably what you care the most about as a result.
- 50:26Spencer Tahill: The result doesn't matter if it's not formatted properly.
- 50:31Spencer Tahill: If you go into HubSpot and you have a first name field and it says Spencer Tahill dot whatever, but it's it's not spaced out and it's my full name, it's not parsed correctly.
- 50:43Spencer Tahill: Like, what's the point?
- 50:44Spencer Tahill: You need to do output and then format the output in a way that makes sense.
- 50:50Spencer Tahill: I'm a senior researcher, don't do researching about this, don't take X threads, don't look at social media, but I want you to make the result in an array or like in a summary, a one-page summary in this format executive summary, sources used, exclamations, data discrepancies.
- 51:07Spencer Tahill: That's the format.
- 51:08Spencer Tahill: And that's if you can nail that, you've pretty much got a great prompt on your hand.
- 51:14Michael Hartmann: Yeah.
- 51:14Spencer Tahill: Context, background, goals, limitations, format.
- 51:17Spencer Tahill: And you're pretty much good.
- 51:18Spencer Tahill: Outside of that, you can do all of your learning on Open AI.
- 51:21Spencer Tahill: Yeah, I think I I won't.
- 51:23Michael Hartmann: Yes, yes, yes.
- 51:24Michael Hartmann: I think that constraints one was the one that was the biggest help for me when we talked because I started go like to write the elucidations come from that.
- 51:31Michael Hartmann: And I started doing things like, you know, only use the stuff that I have provided you specifically for this task, and don't use the other resources because otherwise that's where that's where I found that it was finding stuff it had access to, but wasn't really relevant.
- 51:48Michael Hartmann: Um yeah, I think that's that was helpful.
- 51:52Michael Hartmann: I still I I had already started doing the context.
- 51:54Michael Hartmann: I think the the big thing I've learned.
- 51:56Michael Hartmann: So I've got some things where I've got long threads of things where I've provided a whole bunch of context and examples of other up with it I've done to give it guidance, and it was really good.
- 52:06Michael Hartmann: It's gotten better over time.
- 52:07Michael Hartmann: But what's happened is it's all in one big long chat, and the performance has been bad.
- 52:12Michael Hartmann: So I actually did kind of what you said.
- 52:14Michael Hartmann: I was like, I'm gonna go ask again.
- 52:16Michael Hartmann: I'm not advocating for Chat GPT, it's the one I happen to use.
- 52:20Michael Hartmann: Um, and in fact, I feel like I should like I there's a part of me is like I should be looking at others more, but I like I've invested in this one.
- 52:27Michael Hartmann: But I actually said, like, I'm having this, like the results are good, the performance isn't is getting worse and worse every time I use it.
- 52:34Michael Hartmann: How should I restructure this?
- 52:36Michael Hartmann: Like in if I'm gonna do it again.
- 52:38Michael Hartmann: And it gave me a great like we went back and forth, and it gave me it actually gave me like here's the structure do this, do this, do this.
- 52:44Michael Hartmann: And um that's how I'm gonna implement it in the future.
- 52:48Michael Hartmann: So I get the benefit of all that, but continued better performance.
- 52:52Michael Hartmann: So yeah, I love it.
- 52:54Michael Hartmann: That idea of using the tool to help you get better at using the tool.
- 52:59Spencer Tahill: That's that's how I learned.
- 53:02Michael Hartmann: Yeah.
- 53:04Spencer Tahill: I sucked at it and I got better at it.
- 53:08Spencer Tahill: The first step of being great is sucking, unfortunately.
- 53:14Michael Hartmann: Yeah.
- 53:15Michael Hartmann: Um, well, this has been awesome, Spencer.
- 53:17Michael Hartmann: Thanks for for this.
- 53:19Michael Hartmann: Um, if folks want to kind of keep continue the conversation with you or learn more about what you're doing or what you're doing with uh with Growth Alliance, what's the best way for them to do that?
- 53:31Spencer Tahill: Um guys, anybody that's listening, book a call with me.
- 53:34Spencer Tahill: It's uh or text me on LinkedIn, you know, whichever.
- 53:38Spencer Tahill: Um I'm completely open.
- 53:40Spencer Tahill: My you know, my eyes can be your eyes in the industry.
- 53:43Spencer Tahill: Um building a lot of complex things.
- 53:46Spencer Tahill: It takes a lot of time.
- 53:47Spencer Tahill: So networking chat, dude.
- 53:49Spencer Tahill: Like uh book a time on my calendar, tell them that Michael Hartman sent you from marketing ops, and we'll just catch up.
- 53:56Spencer Tahill: We'll just get a coffee.
- 53:57Spencer Tahill: I like meeting people.
- 53:58Spencer Tahill: So um everything is on my LinkedIn.
- 54:00Spencer Tahill: It's just like slash Spencer Tahoe on LinkedIn or something like that.
- 54:03Spencer Tahill: Gotcha.
- 54:04Spencer Tahill: Um, but yeah, don't think that you're alone in trying to learn a new thing because I sucked at it too.
- 54:11Spencer Tahill: So it's okay.
- 54:12Michael Hartmann: Yeah, no, like I've it's definitely been a like you have to invest a ton, right?
- 54:19Michael Hartmann: And it's I wish I had more, but it's it's been I feel like I've gotten better and better over time.
- 54:25Michael Hartmann: So yeah.
- 54:28Michael Hartmann: Anyway, well, it's a pleasure, Spencer.
- 54:30Michael Hartmann: Thank you so much.
- 54:31Michael Hartmann: Thanks for staying up late.
- 54:32Michael Hartmann: Uh I I know this was a long time coming, so I appreciate it.
- 54:36Michael Hartmann: Thanks also to our our listeners and our rest of our audience.
- 54:39Michael Hartmann: Uh, we always appreciate you.
- 54:41Michael Hartmann: If you have suggestions for topics or guests, or you want to be a guest, you can reach out to Naomi, Mike, or me, and we'd be happy to get the ball rolling.
- 54:48Michael Hartmann: Until next time, bye everybody.