Mastering ChatGPT: Advanced techniques for workplace communication and productivity | Hiten Shah

Claire Vo 0:00 I wanna learn how to use ChatGPT the right way. So you gave us one tip, which was, "Be thoughtful about its memories." But what else? What are your tips?

Hiten Shah 0:08 I usually won't start anything without a ton of context or with the intention of giving it context over time. Just show it what great looks like. It's like a human. If a human doesn't know what great looks like, they're not gonna know what great looks like.

Claire Vo 0:19 You do have an example of something that you think is great, which is your boss's operating manual.

Hiten Shah 0:23 My boss's name is Morgan, so it's What Would Morgan Do? So I'm gonna create the project, and then I'm gonna add these files here. I wanna pitch Morgan the craziest product idea I can think of. What is the best way to pitch it to him so we can go after it? [laughing] Now it's like, this is what you want Morgan to say. Yeah, that sounds like him.

Claire Vo 0:45 This is a tip for all the ICs out there: Go and replicate your boss, and prep for conversations with them. [upbeat music] Welcome back to How I AI. I'm Claire Vo, product leader and AI obsessive, here on a mission to help you build better with these new tools. Today, we have a great conversation with Hiten Shah, who's been building B2B SaaS for over 20 years. Not only is he a great founder and product thinker, but he is, by my estimation, a total ChatGPT power user. He's figured out how to load up ChatGPT Projects with information about yourself and about your boss to figure out the best way to communicate with each other and get work done. Let's get to it. This episode is brought to you by Paragon, the integration infrastructure for AI SaaS companies. Are AI features on your 2025 product roadmap? Whether you need to ingest data for RAG from your users' external apps, like Google Drive files, Gong transcripts, or Jira tickets, or build AI agents that automate work across your users' various tools, integrations are key to building useful AI features. However, building every one of these integrations costs months of engineering, time you simply can't afford, given the rapid pace of AI advancement. Paragon is an all-in-one embedded integration platform for AI products. Industry leaders like AI21, you.com, and Copy.ai use Paragon to connect over 100 of their users' SaaS applications to ingest data for RAG and provide their AI agents with thousands of integration actions. They've accelerated their integration development velocity by up to 50X, allowing their engineering teams to focus on core product features. Want to fast-track your integration roadmap this year? Visit useparagon.com/howiai to learn how. That's useparagon, P-A-R-A-G-O-N, .com/howiai.

Claire Vo 2:57 Thanks for being here! I am excited to talk about everybody's favorite friend, ChatGPT. Today is all about ChatGPT, right?

Hiten Shah 3:06 Yep.

Claire Vo 3:07 How did you get to ChatGPT being your, your favorite?

Hiten Shah 3:11 You know, it's, it's really funny. I, I obviously started using ChatGPT first. Uh, it existed first. There's a bunch of experiments I did with the API and all that prior to ChatGPT itself, so with OpenAI's APIs, and then when, when, when it came out, started playing around with it. It really wasn't doing what I wanted in terms of my output that I was looking for. I think it was still helpful in a bunch of areas that we'll talk about today, but it wasn't quite getting there. So then I started using Claude when it came out, and I was... In the initial days of using this stuff, I was actually way more Claude-heavy by, like, more than 50%. Even though ChatGPT came out first, I had access, I upgraded and all that. And I think the turning point for me, which I can't prove because I don't know the dates, was when memories started working, and that made it so that... And, and I'm a very, like, um... I don't try to talk to ChatGPT about everything, and if I'm gonna talk to it about things that are, like, a little random, I'll actually start

Hiten Shah 4:12 a temporary session.

Claire Vo 4:13 I wanna learn how to use ChatGPT the right way. So you gave us one tip, which was, "Be thoughtful about its memories." But what else? What are your tips? Uh, I'd love to see them.

Hiten Shah 4:22 One of the biggest things here, we talked about memories, but temporary for things that you don't care for it to remember is really important, and then I'll archive chats 'cause they don't, they don't count, so to speak, in the memories from last I checked. You can also unarchive them. At one point, I actually archived all my chats, and then I started pulling out the ones I liked. It was definitely a pain for a number of reasons, 'cause I have a lot of chats. I have, like, thousands and thousands. That's why we're not going into [chuckles] my ChatGPT today. I have a blank one with the $200-a-month plan or whatever, which was also a game-changer. Once ChatGPT introduced that plan, it definitely changed a lot of things for me in terms of my usage. I still get limits with Claude. I barely get limits with ChatGPT, although I use Projects a lot, so I definitely hit decent amount of limits. But Claude's limits were also another very, like, brutal thing for, for when you're trying to use it and trying to get something out of it, especially when you need it now. I think my next point is something everybody knows, and that's that context is everything. So, for example, um, um, I don't have it shared here, but I have a project to make projects. I have a project to make deep research prompts.

Claire Vo 5:39 Well, this is, this is some of the feedback that we've consistently gotten from guests, which is, you can try to improve your output, or you can try to improve your input, and everybody says, "Improve your input, and the output will come." And so they, they really recommend, which I think you're gonna show us, is how do you front-load the time to make sure your prompts make sense, that your projects are set up accurately, that your GPTs are well-structured? If you can do that, then you're gonna get that consistent, high-quality outcome. So what are those-... kind of go-to practices that you think everybody should do as they're building these things?

Hiten Shah 6:14 Yeah. You know, if you don't love frameworks, then you need to start loving frameworks. Um, I love frameworks, and I don't mean that in the, oh, seeking wisdom, Warren Buffett's mental models and frameworks, although I kinda do, but those are more like a bunch of tools. I think the strategy is, if I'm trying to get something done, can I find something that's good enough or great that already works for other people and that someone's already shared? Oftentimes, the models already know the framework, so now you're just trying to, like, tease them into the right spot or give them the framework itself, and then it sort of gets even more contextual about it. And I've just noticed this over and over again. So I usually won't start anything without a ton of context or with the intention of giving it context over time. 'Cause I don't wanna say that, like, I won't just do a one-sentence prompt. Yeah, I'm, I'm- I do that all the time, but I don't expect that one-sentence prompt to just magically work. I actually will even oftentimes be like: "What context do you need? This is what, what my goal is." Or... And another one that, like, I don't see enough people doing is taking great outputs, whether it's from AI or not from AI, and using that to help AI codify that kind of output, and it lets you build more outputs like that. That's probably one of the magical ones that I've found, which is just show it what great looks like. It's like a human. If a human doesn't know what great looks like, they don't have to know what great looks like. So then you're just get in, in, in some ways, feedback hell, so to speak, of going back and forth until they get it or until you've put something together. But typically, as a, as a human, if you give them an example and say, "Hey, this is what the thing I'd like from you," they can do a pretty good job of filling in the blanks, so to speak.

Hiten Shah 7:58 I think AI can do a better job. [chuckles]

Claire Vo 8:00 So you do have an example of something that you think is great, which is your boss's operating manual.

Hiten Shah 8:06 Yes.

Claire Vo 8:06 So could we use that as an example to get into-

Hiten Shah 8:09 Yeah

Claire Vo 8:09 ... exactly how you might do this?

Hiten Shah 8:11 Yeah. Uh, we'll, we'll create the whole thing. So I would first create a project, and my boss's name is Morgan, so it's What Would Morgan Do?

Claire Vo 8:18 [chuckles]

Hiten Shah 8:18 Or What Would Morgan Say? So it's WWMD. So I'm gonna create the project, and then I'm gonna add these files here. And so he has an operating manual. Really great human being, built this out, sent it over, uh, when he first joined the company and became my boss, and then I've been sending this to everyone. And now, everyone's already seen it, [chuckles] and new people that join, 'cause someone else is sending it to them. Great. So that's his manual, and for anyone that doesn't know, these are things that oftentimes, you know, it used to be a bigger thing back in the day, I think, where people would create their operating manual, so you knew how to work with them. Now, there's some other tactics and stuff that are actually part of other projects that we can talk about. Then he also tends to share a few different sort of docum- uh, not documents, but, like, articles he likes, and so there's one he shared called Job Is Communication. I'm sure it's out there somewhere in the world. Uh, if it's not, whatever. So I would put those two things in, and that's the project files, right? Then I would add instructions. So I'm gonna open a new chat and be like... I, I actually would do exactly this. I would add this, add these two in here. So this is a chat. I'm gonna keep stick to what I, what I know, which is 4o.

Claire Vo 9:25 4o!

Hiten Shah 9:27 Yeah, yeah, yeah.

Claire Vo 9:27 Classic.

Hiten Shah 9:28 Yeah, yeah. I'm a, I'm very classic on that.

Claire Vo 9:31 Mm.

Hiten Shah 9:31 So I'll do that. I'll pick 4o. I really like 4o, b- but all of them tend to work. 4o is just faster and, and gets me what I want quicker. And so here I'll say, "I'm, I'm creating project, uh, with these documents, and I want to be able to converse with it, with the project, and simulate type of feedback or advice that my boss would give me. Can you create the instructions for that project, knowing that I'll be using these files in the project?"

Claire Vo 10:07 I love that you corrected your typo because I just roll right through it.

Hiten Shah 10:11 Yeah.

Claire Vo 10:11 That was very polite of you.

Hiten Shah 10:13 Yeah. I, I... It's funny, like, the way I type is always correcting anyway.

Claire Vo 10:18 Mm-hmm.

Hiten Shah 10:18 So [chuckles] it wasn't for ChatGPT's sake, uh, 'cause, you know. Uh, yeah. So yeah, but that's great. So okay, so here it already gives me the objective, fine. Boss is Morgan, great. Uh, system configuration, personal. Okay, okay, interesting. It has not given me the actual instructions, right?

Claire Vo 10:40 Yep.

Hiten Shah 10:40 So I'm like: "Okay, where are the actual instructions?" So I'm gonna go, and this, this is, like, in another doc in there, which I call AI Shaping, 'cause you're basically trying to shape it to get it what you want. So this is a classic example of, I could paste all this crap in there, so to speak, but... And it says project setup, but it doesn't feel right, so it's like, "Can you give me the specific instructions that I can use in the project so I can just paste those in there?"

Claire Vo 11:08 So I find this a useful kind of, like, prompt, which is just tell it where it's gonna go, [chuckles] and maybe it'll give you a little bit better... Oh, now we got some... Oh, well, there's some instructions.

Hiten Shah 11:19 Yeah, so usually what I've already done is, uh-- and, and this is another trick. I've done this in, in several chats. So when I find one where it actually did what I wanted, which, eh, we can argue. We'll, we'll probably just try this one, or I can copy one over if this doesn't work from where I have it. But this is what I find most useful. If I can get a good output, I will always try to get it to codify the output, especially if it's one where I'm gonna have to do this over and over again. So in this case, I said, "Hey, give me the instructions." I think just for shits and giggles, so to speak, I'm gonna use these, um, 'cause I think they might be good enough. Um, and I'm not trying to review it, and I also don't really care too much about this, as long as it looks good enough, because I can always iterate, uh, and do things. And I also use the word simulate. I don't always use the word. In this case, I felt like, okay, let's try it like this. So there's just a lot of, like, "Okay, we'll do it like that." But if this were really, really good, and I'm like, "Oh, crap, this is impressive," I'd make it codify it, and I'd turn it into... I could turn it into a project or something. So one thing I would do if I love this is, "Hey, can you help me create a project so I can create these type of instructions in the future?"... and that's what's got me a very good system of projects helping me use-

Claire Vo 12:30 Project for projects.

Hiten Shah 12:31 Project for projects, yeah, whatever. I know it's-

Claire Vo 12:32 So you're gonna add the instructions right here?

Hiten Shah 12:34 Yep, right here. So they're all in here, cool. Let's say... And then now I'm just going after it. I wanna pitch, wanna pitch Morgan the craziest product idea I can think of, of for us. What is the best way to, to pitch it to him so we can go after it? [laughs] Now it's like, "This is what you want Morgan to say."

Claire Vo 12:59 Oh, yeah!

Hiten Shah 13:00 It's like, "Thank you. Okay. Yep, that sounds like him. Yep." And, and this part, this is key. So he's our VP of Product and Growth. I report to him. I'm a PM, and this is a- th- this- I know his manual. This is the type of stuff he would go for. So now it's, like, very specific. Lead with a question, AI-first CRM, whatever, whatever crap, right? So it basically gives me exactly what I should do when it comes to him, and it even says what will he respect, et cetera. Right.

Claire Vo 13:29 Oh,

Claire Vo 13:30 man. So we've had a lot of people on the pod, or a couple people, who have explained how, as a manager, you can replicate yourself, but this is a tip for all the ICs out there. Go and replicate your boss, and prep for conversations with them. And, you know, it's funny, I, I read this, and I was thinking, "Oh, is this just kind of generic?" And I was like, "Some of that would work on me, some of it wouldn't," but I bet people that work with me could pluck a couple artifacts and, um, make a, make a project or a GPT that really does... Uh-oh, find out everything about me in a professional sense. There you go. So you're doing it.

Hiten Shah 14:11 So what I would do if y-

Claire Vo 14:12 Yeah

Hiten Shah 14:12 ... you were my boss-

Claire Vo 14:13 Yeah

Hiten Shah 14:13 ... is I would literally do this, 'cause I know there's enough-

Claire Vo 14:15 Mm

Hiten Shah 14:16 ... about you out there.

Claire Vo 14:16 Yeah.

Hiten Shah 14:16 Now, in another kind of example we can give, or it'll be in the doc, I, I do the same for myself. I call it, like, a personal OS, and it has my personality type, any of the personality tests, Myers-Briggs and everything, and then it helps me a lot. But if I know that about you... So M- Morgan is actually an Enneagram 5. I'm an Enneagram 9. What I would actually do is feed that into the instructions in that as well, if I wanna make it specific to me and our relationship. Because these personality tests help you... Well, Enneagram in particular, helps you with relationships and how to sort of, uh, meet in the middle around communications and stuff like that.

Claire Vo 14:54 Yeah, that's actually a really great idea, because I do all this. When I start as a new leader on a team, I do the work of gathering, you know, has anybody done DISC or Enneagram or any, you know, um, Myers-Briggs? We kind of do that. We also have this questionnaire, which I'm happy to share in the show notes, which is 10 questions about me. What am I motivated by? How do I like to receive feedback? How do I work? All this stuff, and I actually do... I have this repository in a Google Drive file of everything I know about the team that they've been told or that they've told us themselves, and I haven't yet put it in a GPT, and that's so smart. 'Cause then I could say, you know, and, and sorry, sorry, Zach and Amanda, I can say, "You know, Zach and Amanda are not seeing eye to eye on project X. What would be an effective way, given what you know about them, to come, to come together?" So it's a really interesting idea. Uh-oh, psychological warfare in the workplace, all for the greater, [chuckles] all for the greater good, empowered by OpenAI.

Hiten Shah 15:56 So we'll, we'll, we'll let it do its thing. It's... What I'm doing is I'm doing research on you, and I said to get stuff in the professional sense or whatever. Sometimes I'll even include, "Hey, she's blah, blah, blah at this place." That way it doesn't get the wrong person. So it might, but we'll see. I don't think there's too many Claire Voes, but maybe there are. So we should be good. Uh, and then, yeah, so we'll see what happens there. We can continue with whatever

Hiten Shah 16:20 else.

Claire Vo 16:21 Great, so I, I love... You know, you, you showed this about someone else, and you're, [chuckles] you're doing deep research on me, which, you know, is always gonna be thrilling to watch on the other side. But I wanna go back to this thing that you said about your personal OS-

Hiten Shah 16:34 Yeah

Claire Vo 16:34 ... this self-aware, you know, addition. How did you build that? Why did you build that? Um, you know, who uses it? How are you adding to it? I'm so curious about this specific use case.

Hiten Shah 16:46 So I, I have a bunch of these links. This one is the personal OS. So I used my project creator to create this, um, and it's basically a project description, what to use it for, and then these are the actual instructions, right? So what I would do is basically take this or take the instructions here and put them into a project. So we'll just go do that so that we can show, not tell, right? And then it even says things like files to upload, and then even has, like, a kickoff prompt if you wanna play around or whatever. So I will go here, make a new project. I don't need to put any files in there, but I'll probably do that too. But what you'll notice is up here, it's like, what are these things, right? So this is 9. This is 1/3. If you don't know Human Design, you should know Human Design, but I'm not gonna get into that-

Claire Vo 17:35 You know, this is the-

Hiten Shah 17:35 ... 'cause it's on the woo-woo side. Yeah.

Claire Vo 17:37 This is the third time Human Design has come up. I live in San Francisco. I don't know if this is an SF meme that's happening right now. It has come up in the last week. So yes, it is quite woo-woo. They'll send you a PDF for free, but why not?

Hiten Shah 17:52 Yeah, I highly recommend. Um, okay, so I also have a voice and tone guide that was made for me a long time ago, 'cause I used to write a lot more. And so, uh, I'll just put that in there, just for shits and giggles, so to speak, although it probably isn't as relevant. So here, here we are with this. Let's, let's play some games. So I'm gonna take Morgan's manual and basically talk to it as if I just got this boss, 'cause this is kind of what I did. So this manual is from my boss, Morgan. Uh, I'm newly reporting to him and would love any and all advice on how best to work with him, considering my own personality, et cetera.... By the way, he's an Enneagram type five. So already, it's like I didn't have to do the work, and it's already giving me the contrast. Here, I would have to do a bunch of Google searches to be like, type, "Enneagram type 5 and type 9 relationship," and then I gotta wait, right? But this is kind of what's interesting, where it started mixing in the 1.3 from Human Design, which I love, and that's why I love giving it that context. Then there's like, "Here's our working style alignments," right? All this is true. Like, I, I've done this before. I know, I know that this is all very accurate, and then this is where things start getting interesting, 'cause it starts coming up with shit. Then that's why, like, the one-sentence ones are great, 'cause it's like, wait, a check-in format? Huh. So this is how I could give him a weekly update on whatever, or this is how I could put our one-on-one doc in ManageUp, and he's gonna love it, 'cause this is how he thinks, right?

Claire Vo 19:28 The- is this how, how you- what, what you would like? Do you think it's accurate?

Hiten Shah 19:31 Yeah, I've read his manual.

Claire Vo 19:32 Yeah, yeah.

Hiten Shah 19:33 And it's... I h- I know it by heart.

Claire Vo 19:34 Yeah. [chuckles]

Hiten Shah 19:35 And this is, this is, like, exactly it. And one of the big ones for him-- And again, everyone will know this if they know how to lead, but that's a different story. But y- if you're a good manager, my opinion is that, and so is Morgan's, is that you're just unblocking all day. And, and he said that in the manual: "My job is to unblock you so we can get shit done." Not, not in those words, those are my words. So yeah, this is perfect. This is exactly what he'd want. Because if there's a blocker, I know he's the first person I should hit up. If I can't solve it any other way, he will just solve it. Um, and so that's cool. There's some really interesting things around conversations. He does ask you to defend things, but he never makes you feel, uh, like you're offended when he asks you to defend things. So he's always got a good way of doing it, so that's an interesting point there. This is kinda interesting. This is interesting, right? So this is in his doc.

Claire Vo 20:26 So it's quo- it's quoting directly from the manual.

Hiten Shah 20:29 Yeah, and that's why I make these projects, so-

Claire Vo 20:31 Got it

Hiten Shah 20:31 ... If I didn't make it a project, which I've done before, and done the same thing, I don't get this level of cont- context coming back to me.

Claire Vo 20:39 Got it.

Hiten Shah 20:39 I get more of like, "Oh, you put this in. I'm gonna go fill in the blanks, or I'm gonna do something like that." Here, I get these kind of references and all that kind of stuff that I just... It's just much harder to get that in a chat. 'Cause I used to try this with a chat, maybe before they had projects. I know they had... And custom GPT is a whole 'nother mess I don't

Hiten Shah 20:55 like getting into. But yeah.

Claire Vo 20:57 So one of the things that, that I notice here that I wanna call out is, you said you like frameworks. And so what I saw in some of this, this prompting, and is here are personality-type frameworks or, you know, guides that-

Hiten Shah 21:10 Yeah

Claire Vo 21:10 ... have been well-established or well-documented, the LLMs probably know, probably know about.

Hiten Shah 21:14 Yeah.

Claire Vo 21:14 You also listed a set of interpersonal dynamic or workplace dynamic frameworks for conversations in that list, and then loaded that up with an individual's context. And then you're able to, in a pretty structured way, address common workplace things, whether, "I'm gonna have a hard conversation," "I'm gonna pitch something," "I'm in conflict with someone." And I think that idea of mixing structured frameworks and personal context is really useful. And one of the things that I'm really empathetic towards new leaders are... I get lots of new leaders in, in my organization. They're in the biggest job of their career, and they always ask, "Can I get coaching?" And I don't know if you've... The coaches out there are really- you're really making money out there. Coaches are expensive. Good coaches are expensive, but a lot of times they're doing this work. They're doing 360s to gather information about you. They're applying frameworks to help you understand yourself and someone else, and then they're helping you get out of, you know, mental patterns or, or giving you other ways to look at, um, problems in a structured way that allows you to move forward. And I think you just basically built, like, both a kind of coach for you to work with your boss, but I can imagine just a coach for yourself. So if you loaded this up with your own personality [chuckles]

Hiten Shah 22:28 So this works.

Claire Vo 22:29 Yeah, yeah.

Hiten Shah 22:29 Like, this is a coach for me, right? Like, like, I'm really mad right now about someone... Let's, let's come up with a scenario. Uh, let me make one up. Someone trying to get control over one of the projects I'm working on because they are trying to lick the cookie, as, uh, as I've been told happens. So I just did that without even giving it a bigger problem, and then this is where it gets interesting. This reaction makes perfect sense. Okay, fine, ChatGPT, you wanna be positive? Cool. But then it tells me why, right? It threatens my control. Not just my control, my integrity of what I'm building, and this is accurate. Like, when someone says, "Oh, I'm threatening your control," I'm like, "Yeah, I'm a control freak, but that's not the problem here." But then it said the

Hiten Shah 23:20 integrity. There you go.

Claire Vo 23:21 Okay, so you've done deep research on me. I'm gonna give you my Enneagram and my Myers-Briggs, and because we're fun in my team, I'm a Sagittarius, so we're gonna toss that into the mix. And I, I just wanna test, I wanna test if this passes my brand Nira to it sniff test.

Hiten Shah 23:38 Okay.

Claire Vo 23:38 Well, while we're waiting for that to happen, let's switch over to one more use case, which is your sales use case, which I think is super interesting.

Hiten Shah 23:45 Yeah, this one's really fun. Um, I find myself over, over time getting really excited about frameworks that I have applied in work, and also that the people who built them applied them, and I can prove it. So whether that's Lean Startup from Eric Ries or I guess they call it the Sean Ellis score now. It's, it's al- it was always called the Product-Market Fit score in my world, but that and a bunch of these other frameworks. So when I landed on this Winning by Design framework, I was, like, over the moon. And so I did what a lot of people would do if you were really curious about it. So I did a site: their website filetype pdf in Google, and this is before deep research and all that, of course. So I found all their PDFs, which was before AI, too. So I was like, "Okay, great, now I can use these and, and do what I'm about to show you." It took hours to do this kind of stuff, and now I can do it in seconds, literally. So basically-... I have, I have this entire kind of, uh, context put in. So this one's already set up. I can obviously set it up again, but the deal is, there's a bunch of these documents, and these documents are all about the framework and the various processes. These are all publicly available. They also have a bunch of private ones you can get if you go through their courses. Then I, I use my instructions project, you got it, to basically build this, and it's pretty simple and straightforward. It's not really crazy, 'cause a lot of the context is in the files. So, and again, this is easy to tweak anytime.

Claire Vo 25:12 This will be familiar to anybody that's in a revenue organization. All leaders come in, and they have their favorite sales framework, enable the field on it. You get a bunch of PDFs, but then I'm, I'm hoping you're gonna show us how to apply this-

Hiten Shah 25:26 Hundred percent

Claire Vo 25:27 ... consistently over time. Okay, great.

Hiten Shah 25:28 So I actually did this already. So I actually asked it, "What can you help me with?" So I didn't even wanna, like... Again, I do a lot of that sometimes with these projects, just 'cause I want it to tell me what it thinks it can help me with. Just like a human, right? Like, "What can you do for me?" Or, "What, what, what's your best skill?" Whatever. So it's like, I can create and refine, create or refine sales processes. Wow, okay, so I can build a discovery script, call script. Cool. Perfect discovery call playbook, draft that. Cool. Design customer-centric stories, messaging, run impact meetings, coach your GTM team. Cool. Cool. Okay, great. Where do I start? Okay, can you create a SPICE discovery guide? And I did it for your product, and it didn't know much about your product 'cause it didn't, I didn't go do a web search. I solved that problem, too, as you probably can guess. But... And then it's basically like, okay, VP of product, director of product management. Cool, this is a problem space. So now I know that if I give it this context myself, it will probably tweak things. So already I think I'm winning, 'cause I can-- I have now learned something of how to use this thing, which is another piece I go into. And then now it goes into situation, 'cause the whole premise of this framework is there are questions you should be asking on sales calls, so you're only demoing, like, five or ten minutes of a twenty or thirty-minute call. Not even five or ten, five minutes. So what you're trying to do is diagnose pain. So we imagine these are like d- those discovery calls. So these are a set of calls, uh, or a set of questions, which is like how many PMs are on your team, whatever. Great. I, I literally, every time I do this, I'm surprised by at least one or two questions in there, which would've been super challenging for me to think of on my own. So that's where I find a lot of the value. Otherwise, you can look at it and be like, "Yeah, yeah, yeah, this is obvious." Yeah, a lot of it is, but that's not the point. The point is that if you're gonna go get on a call and you wanna use this sales process, you wanna do a really good job, hopefully. These questions help you figure out how to do that.

Claire Vo 27:15 Yeah, this question under pain, I don't ask, but I should, which is, where do things typically break down? I'm not good at, at pain discovery, so this is already giving me, giving me some ideas.

Hiten Shah 27:25 Yeah, exactly. And so, like, and pain is like, uh, you know, as you know, one of the most important things when you're trying to actually do sales, and then the impact questions are always valuable. But again, this is part of the value of the framework. So I think the key thing to highlight is, I love that you like that. It makes perfect sense. It's the framework producing it. It's not ChatGPT. I mean, it is, but it's that framework that taught ChatGPT how to answer these questions very precisely, 'cause it has, like, twenty PDFs from these people, you know? Um, and then, uh, critical event is a big one. So if you're into sales, you know that you need to figure out if there is a date tied to the init- and there's an initiative tied to it, so this is great. So yeah, I, I had it do it, right? So then, to your point, oh, maybe it doesn't have context, maybe it doesn't know enough. Great. You can imagine what I did. I did a, a, deep research prompt. I asked for a deep research prompt on it. I did it in the sales Winning by Design project on purpose. Even though the project is not meant for that, of course it can do whatever the heck I try to get it to do. But I purposely did that because I wanted to improve the Winning by Design outputs. So already did this yesterday. It's right here. So I did some kind of analysis. It got me all kinds of things about your product company, whatever's going on. I... You know, same prompt it gave me, did all the corrections. So now all I do, and this is how dummy I am about it, because it usually works, I'll just copy it.

Claire Vo 28:47 Hit the copy button. [chuckles]

Hiten Shah 28:49 Just don't even need to read it. I don't care. It's all good. Don't worry about that. Okay, now here is context on ChatPRD in terms of the history, et cetera, of the product and research, uh, related to it. Fine, whatever, it could say whatever. Then I would probably go in and say, "Please improve the deliverable above using this con- this new context, this updated context."

Claire Vo 29:19 I am a dash caret, uh, dash, you know, arrow little instructor as well.

Hiten Shah 29:25 [chuckles] All the time.

Claire Vo 29:26 This arrow-

Hiten Shah 29:27 It just feels like it splits it up-

Claire Vo 29:29 Yeah

Hiten Shah 29:29 ... but it probably doesn't matter, but I just wanna do that. I love it.

Claire Vo 29:32 I completely agree.

Hiten Shah 29:33 So this is, this is my point about the memory, right?

Claire Vo 29:35 Yes.

Hiten Shah 29:35 It'll tell you when it does that.

Claire Vo 29:36 Yeah.

Hiten Shah 29:37 So oftentimes I'll sit there and just start pruning it.

Claire Vo 29:39 Yep.

Hiten Shah 29:40 And so that's what I like about it, is like, just makes it really easy to go cut that out of the memory-

Claire Vo 29:43 Yeah

Hiten Shah 29:43 ... if I want to.

Claire Vo 29:44 I don't need that. Yep.

Hiten Shah 29:45 Right. Exactly. So here you go. Purpose, SPICE journey map. It did a little bit of a different thing, which is fine, like, but at the end of the day, it's not even telling me what the gap analysis is. Maybe it got confused about something, but I don't, I don't know, maybe not. But yeah,

Hiten Shah 30:00 that's, that's the way I would do this stuff. And I'd probably prompt it a little differently and be like, okay, and I'll probably say this: "Can you give, give me a demo script for the most relevant, uh, ICP/persona," whatever, in case it doesn't get it, "that the research identified?" And I don't even know if the research identified this, but ChatGPT tends to-

Claire Vo 30:20 'Cause you just, you just clicked the Copy button.

Hiten Shah 30:24 Yes. Uh, savvy sales and... Oh, interesting! Huh.

Claire Vo 30:28 Well, so I think what it did, if I read above-

Hiten Shah 30:31 Yeah

Claire Vo 30:31 ... you didn't prompt it very well.

Hiten Shah 30:32 No, not at all.

Claire Vo 30:32 Uh, which is-

Hiten Shah 30:33 Yeah

Claire Vo 30:33 ... it, it tried to integrate the framework into my product-

Hiten Shah 30:37 Yeah, and that was the-

Claire Vo 30:37 ... I was supposed to integrate the framework into the kit.

Hiten Shah 30:40 So let's do it again.

Claire Vo 30:42 Yeah, but I, this is one of the things I love about AI is-... you know, imagine I told some research analyst, "Go do this for me," and they came back three days later, and I was like, "No, no, sweet child, totally wrong." So expensive! You get this wrong, whatever, it's 90 seconds we just lost, and we can just give it another go.

Hiten Shah 31:04 Yep, exactly.

Claire Vo 31:06 And I think this goes back to what you said at the beginning, which was, who cares if you have to re-prompt it? It's so cheap, it's so fast. Like, who cares if we have to do this again? It totally just works.

Hiten Shah 31:19 Yeah, and I'll keep doing that. I don't care, like, it takes nothing, right? And, and I think that's great. Like, I, I kinda love the mistakes it makes, actually. Sometimes I learn all kinds of weird shit from the mistakes it makes. Yeah, it keeps saying rev ops and product persona, which is interesting.

Claire Vo 31:34 [laughs]

Hiten Shah 31:34 I know ...

Claire Vo 31:34 Well-

Hiten Shah 31:35 I should...

Claire Vo 31:35 I think this gives us, whether or not it gives us-

Hiten Shah 31:37 Yeah

Claire Vo 31:37 ... a good output-

Hiten Shah 31:38 Yeah

Claire Vo 31:38 ... it gives us some good ideas. Let's see if our research, um, your deep research on Claire is done. I don't

Claire Vo 31:44 think there's actually that much, um-

Hiten Shah 31:46 Yeah

Claire Vo 31:46 ... interesting about me on the internet, but let's, let's see.

Hiten Shah 31:48 Oh, we're about to find out.

Claire Vo 31:50 Yeah.

Hiten Shah 31:51 Okay. Let's see.

Claire Vo 31:52 Oh, we got it. Okay, so you got a novel-

Hiten Shah 31:55 Yeah, that was the point

Claire Vo 31:56 ... on Claire.

Hiten Shah 31:57 Yeah.

Claire Vo 31:57 I do have board roles, so...

Hiten Shah 31:59 Oh, you do? Okay.

Claire Vo 32:00 I do. Look at me.

Hiten Shah 32:01 Okay. Okay.

Claire Vo 32:02 Okay.

Hiten Shah 32:02 Okay, I just wanted to check.

Claire Vo 32:03 Move along.

Hiten Shah 32:03 Okay. Yeah. No, I, I'm never surprised at how much it gives me these days, 'cause it tends to give me a lot when I do these kind of prompts.

Claire Vo 32:11 Yeah.

Hiten Shah 32:11 I do these kind a lot. Okay-

Claire Vo 32:12 Okay, so you're copying it.

Hiten Shah 32:14 What is the best way we should do this? Should we make a project?

Claire Vo 32:16 Yeah, make a project.

Hiten Shah 32:17 Okay.

Claire Vo 32:19 Claire.

Hiten Shah 32:19 Claire. Okay.

Claire Vo 32:21 Claire. [chuckles] It's just Claire.

Hiten Shah 32:22 Yeah. Okay, download that. Let me put that in here.

Claire Vo 32:29 Okay.

Hiten Shah 32:30 Okay.

Claire Vo 32:30 Very exciting.

Hiten Shah 32:31 Yeah, yeah. Okay. Claire is my boss. There is research on her from the internet in the project files. Want your help as I'm working-

Claire Vo 32:45 With her

Hiten Shah 32:46 ... with her. Okay.

Claire Vo 32:48 Okay.

Hiten Shah 32:48 Um-

Claire Vo 32:48 An Enneagram 1.

Hiten Shah 32:50 Claire is an Enneagram 1, okay. I don't know-

Claire Vo 32:54 I'm a ENTP, Myers-Briggs.

Hiten Shah 32:57 ENTP, okay.

Claire Vo 32:58 Hard D DISC. [chuckles]

Hiten Shah 33:01 Hard D DISC, okay.

Claire Vo 33:03 Um, and a Sagittarius, of course.

Hiten Shah 33:05 Sagi-

Claire Vo 33:05 Fire sign.

Hiten Shah 33:06 Sagittarius, ah.

Claire Vo 33:09 It's I-U-S. Yeah, whatever.

Hiten Shah 33:11 Yeah.

Claire Vo 33:11 It is like, eh.

Hiten Shah 33:12 Sagittarius-

Claire Vo 33:12 Yeah, there you go

Hiten Shah 33:12 ... something like that? Close enough.

Claire Vo 33:14 Double G.

Hiten Shah 33:14 Okay.

Claire Vo 33:14 Yeah.

Hiten Shah 33:15 Yeah, double G. So, all right.

Claire Vo 33:16 Yeah, there you go.

Hiten Shah 33:17 I don't know any Sagittariuses-

Claire Vo 33:18 No, see if we-

Hiten Shah 33:18 ... obviously. Wait, did we get it right?

Claire Vo 33:20 Yeah.

Hiten Shah 33:20 Okay, we got it right.

Claire Vo 33:20 Yeah, okay.

Hiten Shah 33:20 I might as well capitalize it. Okay, cool. What do you wanna ask it?

Claire Vo 33:25 I have to tell Claire no on a roadmap item she wants.

Hiten Shah 33:29 Like, I don't even know if this will work. Uh-

Claire Vo 33:31 Well, we're gonna find out.

Hiten Shah 33:33 Yeah. Well, it's 'cause it's just so dirty with the research, right? And then, but we did get the Enneagram, so... Okay, so now you're gonna have to judge this. [laughs]

Claire Vo 33:41 Okay, I'm gonna judge it.

Hiten Shah 33:42 [laughs]

Claire Vo 33:42 All right. Let's see if we, it got... We, "She's known to empower strong convictions that are backed by evidence or framed in terms of business impact." Okay, I wanna be up front. I do like, "I want to be upfront." That's great.

Hiten Shah 33:54 Okay.

Claire Vo 33:54 Um, I famously, though, this is very meta for this conversation, famously hate frameworks, so that framework, prioritization framework-

Hiten Shah 34:01 ...

Claire Vo 34:01 ... is not gonna work for me. Okay, it doesn't meet the bar. Yep, that's good. "If helpful, I'd like to set it up as a structured experiment," totally would work on me. Lightweight experiments, that's the easiest way to get, to get what I want. Yeah, "Use experimentation language. It opens the door without giving false hope." [chuckles] Sure, hope springs eternal. [laughs] This is... I don't know, you know? It, it doesn't s- s- it's decent advice. It would probably work on me.

Hiten Shah 34:29 That's kind of my point, right? So now-

Claire Vo 34:31 Yeah

Hiten Shah 34:31 ... here's another little fun one I did. I actually did this. I did deep research on myself, and I happened to have it sitting there, and I happened to be at dinner with a very, uh, old friend of mine from college, and um, and, and I did something like this.

Claire Vo 34:42 Mm.

Hiten Shah 34:42 I don't remember the exact prompt, and I basically took my deep research and said, "Please describe he."

Claire Vo 34:48 Uh-oh. Oh, good- goodness.

Hiten Shah 34:51 Right?

Claire Vo 34:51 Yeah.

Hiten Shah 34:51 And then I just showed him, 'cause I wanted to show him the power of this thing, and he's like, "That's you." Like, literally everything it said was me.

Claire Vo 35:00 I am pragmatic. Uh, okay. Okay, leadership style, high velocity. That's the one I want. When people ask what it's like to work with me, I say, "I go very fast," so that really got me. Okay, let's go. Product philosophy, let's see if this one gets right. Um-

Hiten Shah 35:18 Mm.

Claire Vo 35:19 Cross-functional by design. Yeah, I think that's right. Sh- yeah, okay, let's go to my core traits. Oh, they got me right! Builder at heart, that's the one I want. Be- [laughs] I like the, the quote is ridiculous, so we'll highlight this on the podcast. "Be more entitled." [laughs]

Hiten Shah 35:41 I, I have found this to work for pretty much everyone I've done, as long as it's the right person.

Claire Vo 35:46 Yeah.

Hiten Shah 35:47 So-

Claire Vo 35:47 Okay, this was, this was great. Okay, so you've shown us... I just am going through everything you've shown us. You've shown us how to make projects to make projects or prompts to make prompts. You showed us how to manage knowledge and prune knowledge, as you say, which I think is a really useful tactic. You've given us a lot of great GPT frameworks for knowing thyself, knowing your manager, knowing each other, and using that to, um, give feedback to each other. And then you've shown us that despite my personal anti-framework bias, this has actually compelled me to think maybe I'm just applying frameworks the wrong way, and if I had an easy tool where somebody else did it for me, I would use it. So you've, you've shown how you can scale sort of like training and frameworks into

Claire Vo 36:30 always on GPTs or models. Um, so this is y- it's been all about ChatGPT, but it's been super useful. Uh, we're gonna do a couple lightning round questions, and then I'm gonna, I'm gonna get you out of here. So question number one, you've said that most of your AI usage is ChatGPT. You mentioned, um-... Claude a little bit. What else is in your stack?

Hiten Shah 36:53 I use AI three to six hours a day, is what I'm clocking in at.

Claire Vo 36:56 Mm-hmm.

Hiten Shah 36:56 It's always three at the minimum.

Claire Vo 36:58 Yep.

Hiten Shah 36:58 It used to be about two, and the specific tools are ChatGPT and Claude. The most useful tool I use is not public, so I can't talk about it or the name. It's actually a desktop app. Um, and, and that's outside of ChatGPT. Uh, outside of that, I try probably 10 to 20 new tools a week. I, I guess here's what I'll say.

Claire Vo 37:27 Yep.

Hiten Shah 37:29 Uh, I think the tools are fine.

Claire Vo 37:31 Yeah.

Hiten Shah 37:31 But when I go back to AI tools, nothing beats a blank canvas for me from ChatGPT today, and there's a f- bunch of reasons why, but the one that I keep coming back to, that I see people making a big mistake of, which we kind of did here, too, to demonstrate it, which is y- so many people try to go to automation when I need to l- understand what's the prompting that's gonna work, and how do I get this thing to spit out reliable prompts? Again, we call it evals and stuff in the product world, which I have lots of opinions on, but the bare metal, the chat, messing with that... Like right now, in five minutes, I could fix that sales winning by design thing so it actually gives the output we want. It's probably the instructions. It probably doesn't understand that when I give it context, it needs to use the context and nothing else, right? Simple thing, I could fix it. If I tried to automate that and say, "I wanna do automated sales scripts," where I just toss it in and do it and use one of these automated tools ... I know there's all these fancy, awesome ones, even Zapier, et cetera. I'm probably gonna be stuck with all the outputs suck. Okay, and then I'm done. But I don't wanna be done. I wanna go build the plan on how to build the automation, and the only way to build the plan is to manually do it repeatedly. Then there's the which model and all that, which is a whole 'nother can of worms, uh, 'cause, like, you use the API for obvious reasons with the product you're building. You have a specific model that you know every time you throw the prompts at it, that you like to throw at it, it gives you the output that the customers like. So I think people are racing to automation, and that's a mistake.

Claire Vo 39:02 Yeah, and I'll, I'll just call out my friend Pete Koomen at YC did this article basically begging, begging a- AI product people to let users manage the prompts a little bit better, so.

Hiten Shah 39:14 Yeah, I, I have so many opinions on that, a- and, and... Yes and no. I think the big one that I would point people to if they need actual, true, good inspiration on that-

Claire Vo 39:24 Yeah

Hiten Shah 39:24 ... is a product called Lex, lex.page-

Claire Vo 39:27 Mm-hmm, yep

Hiten Shah 39:27 ... L-E-X. They've instrumented the whole product so you can tear up all the prompting-

Claire Vo 39:32 Yeah

Hiten Shah 39:32 ... make it your own, et cetera, in a very impressive way. So the yes and no is, like, some products, it makes sense. I think others, it will lead you in a bad direction, and I haven't told Pete this yet, but I've read his whole thing, and I have similar thoughts in general. Some of the products we're building, the Lex model makes sense, and I call it-

Claire Vo 39:48 Yep

Hiten Shah 39:48 ... the Lex model because Nathan's one of the best product people out there.

Claire Vo 39:51 Yep.

Hiten Shah 39:51 Don't tell him I said that, although he's gonna hear it.

Claire Vo 39:54 Yeah, well, we love him-

Hiten Shah 39:55 Um, and that's Nathan from Lex

Claire Vo 39:55 ... we love Lex.

Hiten Shah 39:55 Yeah.

Claire Vo 39:56 Yeah.

Hiten Shah 39:56 So, so but the way he thinks about this is very much aligned with what Pete was saying. I would just advise people that some products don't make sense to make the user see all that or do all that kind of work.

Claire Vo 40:08 Yep. Okay.

Hiten Shah 40:08 So then all this is back on us product people, so.

Claire Vo 40:11 All right, and then, uh, speaking of work, I am gonna wrap with my final question, which is, when ChatGPT is not working for you... Yes, you're a prompt master, but if you need to get inside, I, I won't ask you what Enneagram ChatGPT is, but if you need to get inside its head, what's your prompting technique? Do you get mad? Do you do all caps? Do you offer money?

Hiten Shah 40:33 Yeah, I know people like to treat it like a human, and that's because the interface is making us treat it like a human, not because it is a human. And I think there's a whole world out there to talk about that, would, that would take us many, many hours, um, and a bunch of drinks. But high level, I bluntly just say, "This is incorrect." So right now, I would say, "Hey, this is incorrect. Your instructions must be wrong," 'cause I'll give it an assumption of mine, of what's wrong, "Please blah, blah, blah, or do blah, blah, blah." But I don't, like, try to bribe it or any of that. I know those tricks work, but to me, that feels artificial, and I'd instead like to find the s- the same, the same way to get what I want without the bribe. Just like with a human, I wouldn't wanna bribe them all the time, because they're gonna get trained on wanting the bribe, and I know ChatGPT, with the memories and all that, is gonna get trained on your bribing. So please don't bribe it unless you wanna head down that path, just like with a human. And so I guess the, the TLDR is, if you're gonna treat it like a human, think about incentives.

Claire Vo 41:38 You are a- this, this just made me think of my eight-year-old and my six-year-old, and you're right. The second they know a bribe works, they look for the bribe. Okay, this has been so great, very useful. Where can we find you, and how can we be helpful?

Hiten Shah 41:51 I'm on Twitter, hnshah, H-N-S-H-A-H. I tweet way too much. Um, yeah, that's it. Like, w- whatever else, like, uh, just, j- just a person that's been in tech probably way too long and loves playing with these things. Uh, but more importantly, I think we're, we're... Everyone says we're in an incredible time, and this and that, but it's like, if you have access to the tool, you should use the tool, and if you need help using the tool, like, hit me up. I don't know. [chuckles] Most of my friends hit me up, and I build things for them, but, like, that's why, you know, I... Thank you for having me on here.

Claire Vo 42:22 I love it. Well, thank you so much.

Hiten Shah 42:24 Thank you, Claire. [upbeat music]

Claire Vo 42:25 Thanks so much for watching. If you enjoyed the show, please like and subscribe here on YouTube, or even better, leave us a comment with your thoughts. You can also find this podcast on Apple Podcasts, Spotify, or your favorite podcast app. Please consider leaving us a rating and review, which will help others find the show. You can see all our episodes and learn more about the show at howiaipod.com. See you next time!