Transcription
I keep having the same conversations with business owners. They say, "I'm slammed. I don't have enough hours." So, they try apps, time blocking, waking up early, and nothing changes.
Here's what I figured out. They don't have a time problem. They have a clarity problem. They're moving fast, but they never choose which direction to head.
So, I built something. It's an AI that interviews you. It asks 20 to 25 questions about how you spend your time, and then hands you an audit where things don't line up. I ran myself through it. I'll show you clips from my interview, how I built it, and then I'll even share a link for you to then have access to this for free to try it yourself. Let's get into it.
All righty. So, if I had to summarize the problem in one single chart, I think this represents it best. We have two axises here. We have one axisy that emphasizes clarity. So, this is the vertical axis. The other axis is speed. That's the horizontal axis. Most people that I meet, business owners and leaders, they spend most of their time in this quadrant here. So the bottom right where the treadmill is where you're running fast but you're going in many different directions and you haven't had much clarity in which direction you want to head. The goal with this interview and using AI to extrapolate what's in your head and putting it on paper and making you think harder is moving you from treadmill to progress where you're actually moving fast and running in a direction that's clear and aligned with your goals.
We understand the problem. Now what's the approach here? So most of the time you interact with chattclaw Gemini etc. you're in this approach where you're the driver and you're kind of guiding the AI where you ask it a question, give it a task, it does the thing, then it comes back and then you give it another question and then it does the thing. That's usually the interaction. You're driving. But what if we flip this? What if AI is the one that's asking the questions and it asks you a question, you give it an answer, and then based off that answer, it asks a new question, and then we keep going back and forth where the AI is actually guiding the conversation. We're not the drivers.
In this approach by simply just flipping the AI and having it ask you questions. I've applied this to many, many different use cases. We're just talking about clarity and time prioritization right now, but again, you can use this for anything. And I found when I work with clients on this specific process, they're constantly surprised and impressed by the types of questions the AI asks and how it makes them think differently about whatever problem they're trying to solve.
Now, this is the approach. Right now, with this specific use case, I've limited the AI's ability to ask you questions to 20 to 25 questions. The reason I do this is I don't want it to take a bunch of your time. Ideally, that's going to give you enough questions to give you good insight into how you're spending your time. This should take between 20 and 30 minutes. It'll be faster if you use dictation. So, instead of typing everything out, just talk to the AI via dictation, hit enter, and that's going to speed up the process. Also, it's going to get more context from your head into the AI's head. And then at the very end, you're going to get a personalized audit back on how you're spending your time and how different ways you could change spending your time based off the goals disclosed in the interview.
Hello. Hello. So, quick pause and you're regular programming. This video is brought to you by me. So, two quick things. First off, below is a 30-day AI insight series, completely free. You'll get 30 insights in your inbox how you can apply AI to your business and your work. The second thing is if you'd like to work with me, blow a series of offerings to see if there's a good fit between the two of us. With that being said, let's get back into the video.
All right, so this is the process. Now, what model are we going to use? The model I'd recommend using is going to be GPT 5.2. Why? Well, within GPT, there's a specific setting where you can set it to auto. When you set it to auto, it's up to the AI to decide how much reasoning it uses for any given response. And that's what this spectrum represents. So, on the fast side is less reasoning. On the deep side, more reasoning. Now, when the AI is asking us questions, we don't want to have to wait forever. So, most of the time, the AI is going to spend a good proportion of its time on the fast side. But when it comes to it synthesizing the entire conversation and giving us our audit, we want it to automatically or we can make it explicitly move over to the deep side and think more before it gives us an audit back. Now, this gets the benefit of both. We don't have to wait, but also we get a good quality response back from the AI when it synthesizes the conversation. This is why I use this model.
Now, for the prompt, this is where all the magic happens. We uh we have a pretty beefy prompt, but luckily AI created most of it. But the things that I specified within the prompt before the AI did anything was the lenses. I wanted to get questions asked through. So there's different personas that we're going to amalgamate into the AI's head. The first is going to be the executive coach. The next is going to be the economist and last is going to be the provocator.
So if I show you exactly each one of these, the executive coach is lens one. Now this the goal of the executive coach is really to understand the what. How are you spending your time today and how does that align up with the goals that you've set out for yourself? So basically what do you want? What are you doing? And how do we close this gap? Because oftentimes people state their goals but what they're doing is completely disjointed from that. So there's a massive gap between the two. The executive coaching questions are going to help us close that gap.
And then we have the economist. So the economist's job is basically to allow us to understand what is the trade-off that we're making here. What is the opportunity cost of when I say yes to something I'm saying no to something else? And we need to be clear on that. So the questions that the economists are going to ask us is going to help us think through that specific lens.
And then finally we have the provocator or the provocator. And uh the premise of this lens is for it to ask us questions that help us think outside of the box in the sense that maybe you set a goal for yourself to say I want to do this in the next 5 years. But what if we said okay you could only you only had the next six months. How can you condense that down to get that done in 6 months or maybe I have 10 meetings a day but what if you could reduce that down to just two. If you had to say no to eight of these how would you restructure your business your days your teams etc to ensure that you can remove those meetings from your day? These questions are meant to stretch you.
Now, those are the lenses that the AI is going to look through. It's important that I don't specify the types of questions that I want it to ask because I want the AI to then take that responsibility on its own. So, it asks questions based off the context you provided based on previous answers.
So, this is the final base prompt that's behind the AI that you're going to be speaking with. So, you can see it is a longer prompt, but again, the important thing here is I did not write this entire thing. I had AI write this for me. All I did was give it context. I told AI that I wanted it to create an AI that would interview me based off of these three lenses. I then gave some constraints on the number of questions to ask and things like that. But outside of that, it researched best practices for prompting GPT5.2 and then wrote the prompt for me. So don't think that I or many other people that you look at are prompt engineer experts. Really, we're giving some guidance and context and allowing the AI to write it for us.
So let me highlight some components in this prompt. We won't talk about the entire thing, but some things I think are important. So at the very top, we're giving the AI a role and objective. So here we're saying you are a time clarity coach. And then for the time clarity coach, we're saying that you need to blend these three lenses that I mentioned previously into the types of questions you're asking because that is your objective is to be an interview to ask questions. Below this, we give it clear instructions on what it needs to do, which is asking one question at a time. And then we're stating the problem up front. So the the persona this AI has for this interview is helping us find clarity because time is the issue. We don't have a lot of time and we need more clarity. And below this we have some rules. So making sure that it is one question at a time because that's really important. Why is that important? Oftentimes when you do these interviews with AI, if you don't specify that, it'll ask you 10 questions at once. And that's not necessarily useful because we want the answer we provide per question to inform the next question. The direction of the conversation should be dictated based off of the context I provide the AI, not it just dumping a bunch of questions on me.
Down here, it breaks down the question methodology. So it's important that these are not the questions, but the methodology of asking the questions. So here, it's basically giving explanation of what the lens is and the angle you should be taking when asking questions but not giving it exact questions to ask. That's very important because I don't want it to have canned templated questions that it asks. I wanted to use its knowledge base to figure out what questions are suitable for that context.
Then we give it the instructions on how to synthesize the overall conversation and what we expect in that output. So in the output we expect it to tell us where our time is going currently, what's the misalignment with our existing goal, what to consider changing, and the question to sit with for a longer period of time. So this is what we expect from the AI at the very end. So that's our base prompt.
Let's actually see what this looks like in action. I ran this interview for myself. So at the very beginning of this interview, I gave it a bunch of context on myself. So I said, I'm the owner of this company. This is what I do. All these things. So I gave it context. After I gave it context, it started asking me questions. And it's important to note when I respond to this AI, it's all through dictation. It's I'm not typing anything. I'm talking. And you can see at the very beginning, it says, "Hey, are you ready to start? This is what to expect." I say, "Yes." So we get into it.
So it starts asking me questions and you can see immediately at the top it's starting with the macro kind of how are you spending your time and the work week. So what does your work week look like based off the different offerings that I gave above. So I gave it context on the offerings that I sell for my business and I said based off of these things that you do where are you spending your most time. I then answer that after I've answered it. It gives me a more a very directed question saying okay give me a rough percentage of how you spend your week per area. Try to guess. I give it that and then we go back and forth. And there's a few different questions in here that really caught me off guard in a good way where it made me think differently about how I spend my time in areas that I care most about for my business going in the next six six to 12 months. And let me show you a few of those questions.
So here's an interesting one. It asked if writing is thinking, what part of your content scripting process is actually thinking and what part of it is production overhead? Now this is something I've not thought about and the context of this conversation right now is basically talking about YouTube. So, I create a lot of YouTube content because that's one of the main drivers for my business. And I was talking through different processes or activities inside of that process of creating a video. And scripting tends to be the highest time suck of all the activities. And I stated that writing is a form of thinking for me. And it asked, okay, well, based off the type of writing you do, because I write out the entire script, can you maybe distill that down to just to write bullets? And if you choose just write bullets, is that sacrificing your thinking process? And that's really the question and the essence of the question. That's something that I've not really thought about andor questioned to my writing process right now. And I found that it actually was thoughtprovoking and made me think about how I wanted to change that process or experiment it with it for the next couple of videos.
Here's another interesting question that it came up with that kind of caught me off guard is actually not just this question, but a series of questions around experimentation and knowing for a fact something impacts something else. And through this question asking process, you start to realize what's actually true and what's just an assumption that you're making in your head and that's not connected to reality. And this is a really good question that pulls that out where it's asking if your goal is understanding, how would you know concretely that scripting is the one that's driving that that specific need that you want and or have from your videos? And this is not a question I've ever asked myself. So I don't really know if scripting is the primary driver for people to understand my videos effectively. And this is something that I've thought through more after doing this interview. and I realize that this is a basis this is a baseless assum assumption that I've not experimented on and can prove that's true. So if that's the case then maybe I need to experiment with these different areas where I spend most of my time on videos to see if I can create as good if not better videos without having to write all the scripts out word for word.
So this is the interview, these are the questions and these are the answers. Now this is something that I like I said I will give access to you so you're going to get access to the shared project so you can play around with it yourself. I'll also share the presentation as well so you can get access to this specific base prompt so you can alter it and tweak it based off of your interest and the things you'd like to run through for yourself or an interview.
And one last thing I want to call out here is the importance of doing this on a reoccurring basis. So once you've established clarity either through a conversation, an interview, you've gone to some sort of retreat for your company or something else, you have the clarity, but over time that clarity degrades and that's what this chart represents where we have time and clarity. So time is on the horizontal axis, clarity on the vertical axis. And you've established clarity here. So maybe you've had some sort of conversation, an interview, whatever else, and you have clarity and you're moving in that direction. Over time, things change in the business and you may need to change your goals and you need need to make sure and make clear that the actions you're taking are heading in that direction of the goals you care about. So it's important to run this audit on a reoccurring basis, either this interview or whatever else gets you that clarity, doing that on a reoccurring basis. And so the chaos that comes with running a business and time, your clarity doesn't degrade. So if you run these audits, you can keep that up.
And as a quick recap, the problem often isn't just being overloaded and not having time, but really it comes down to clarity where you're spending your time. One of the most efficient and effective ways I found getting clarity is using AI for that process. Allowing AI to ask you questions, you provide answers, and then it help you provide that clarity. The interview I've set up here is only going to take you 20 to 30 minutes, 20 to 25 questions. And again, it's best to do this on a quarterly basis so we can sustain that clarity over time as the chaos in our business changes. And that's it.
So, if you enjoyed this, please reshare with your friends. And also, as a reminder, two quick things. First off, Blow is a 30-day AI insight series, completely free. You'll get 30 insights in your inbox how you can apply AI to your business and your work. The second thing is if you'd like to work with me, Blow a series of offerings to see if there's a good fit between the two of us. Now, you just saw AI do the hard thinking for you. I built another system that does the same thing for images. You give it five simple inputs. The AI writes the prompt for you. You then paste that into an image model and get high quality images back. Check it out right here. Go ahead, click that video right there.