Transcription
Today I'm going to talk about 17 ways to come up with startup ideas in the AI age. I'm going to go through some workflows. I'm going to go through some tools. I'm going to go through some prompts. I'm going to show you exactly how to come up with really good startup ideas. Let's get into it. [Music]
You know, and a lot of people listening, they they say to me, "Well, I don't need a startup idea, Greg. I'm already working on a startup." Well, this is going to help you grow your startup. You know, it's going to you're going to come up with ideas for your startup that uh are are going to basically grow the product, grow your customer base. And if you're looking for a startup idea, well, you know, a lot of people here have a 9 to 5. They're looking for a side hustle, are looking for, you know, just another business, a creative outlet, and I didn't want to gatekeep. So, I'm going to go through exactly how uh I would come up with a startup idea uh in the AI age.
And I got to hand it to Sam Alman, the co-founder of OpenAI, because he really inspired me to do this uh this episode of the pod. Uh he just posted on his blog. He says, "For a long time, technical people in the startup industry have made fun of idea guys, people who've had an idea and were looking for a team to build it. It now looks to me like they are about to have their day in the sun." So, if you are an idea guy or gal, this episode and pretty much honestly all episodes of the pod, but this one specifically is a tactical uh saucefilled uh 17 steps how to come up with startup ideas in the AI AI age.
So the first thing you can do if you want to start up idea is you can literally go into chat GPT and say give me 10 tedious workflows and then you insert a job title, for example, social media marketer, does that AI could automate and then you just rinse and niche down. So let's actually go and cook that up and see what what you know how I would approach it. So, uh, I'm going to go ahead and, uh, copy the prompt: Give me 10 tedious workflows that a social media manager does that AI could automate.
Now, you're going to want to uh, make sure that 03 is clicked here. Sometimes, by default, I think 40 is clicked. Um, but I find for for questions like this, you're going to want to use 03. It's just going to be a bit more effective. So it it's going to give you I don't know around 10 I think uh of ideas. Uh so daily content roulette, sifting through trends, brand assets and halfbaked brainstorm docs just to fill tomorrow's posting calendar, captioning, frequent uh Frankensteining, manual hashtag mining, evergreen re-edits. So it's going to give you a bunch of ideas that are really really tedious. And tedious is a a synonym synonym for painful. And if you're coming up with a startup idea that helps solve someone's pain, you will be able to make money from it. So, I think that that's why I like to use the word um tedious uh or painful. Um and then you just got to go through here and pick the one that you think is the most tedious. And you can even sometimes I do this and I go like which one do you think is the most tedious and why? And just let the LLM do the hard work.
Now, if you are a social media manager, you probably know this already. Um, but if you're not, um, you might not know it. And even if you are a social media manager, I I you know, I I find it helpful to just ask anyway. So it says screenshot and pace reporting. It's peak drudgery. Every platform hides its metric in a slightly different maze. Exports break the moment a filter changes and one typo on a slide can nuke everyone's confidence in the numbers. There's zero creative upside. Just an hourlong copy paste gymnastics that the exact team skims in 30 seconds. High stakes for accuracy, low reward for you, and perfectly suited for API hungry API AI bot that never misalign a chart. So there you have it. You know what the pain point is and then you can go ahead and just start prompting it and be like, you know, what would an MVP look like and give me a realistic high growth uh plan to acquire, you know, 100 customers in 30 days. And it's going to tell you this. So this is this is now you you know you you understand a pain point, you know now you're just like planning the MVP and growth strategy you can start in chat GBT. I find Manis also is super helpful uh and then you can uh and then you you you start having a plan of what to do so that's the first way you can come up with a startup idea in the AI.
Quick break in the pod to tell you a little bit about Startup Empire. So Startup Empire is my private membership where it's a bunch of people like me, like you who want to build out their startup ideas. Now, they're looking for content to help accelerate that. They're looking for potential co-founders. They're looking for uh tutorials from people like me to come in and tell them, "How do you do email marketing? How do you build an audience? How do you go viral on Twitter?" All these different things. That's exactly what Startup Empire is. And it's for people who want to start a startup but are looking for ideas or it's for people who have a startup but just they're not seeing the traction uh that they need. So you can check out the link to startupmpire.co in the description.
Number two, scroll your own chat GPT history. If you're repeating the same prompt weekly, that's a product in in hi h h h h h h h h h h h h h h h h h h h h h h h h h h h h h h h h h h h h h h h h h h h h h h h h in plain sight. So, one example of that is like I was just scrolling through my chat GBT history before this and I noticed that I ask ChatGBT to uh take my content and write it more clearly. If I'm doing a YouTube description, like for example, for this episode, I'll ask it, you know, I'll I'll write something and I'll say make it more clearly just so people understand what I'm saying. And by the way, if you're listening to this on YouTube, go ahead and like and comment so I know to continue doing these videos. Um, but yeah, back to the Chat GPT history thing. Uh, this, you know, number two is really only helpful if you use Chat GPT every single day or if you use code every single day or if you use Mattis every day. You're going to basically know how you're using it. And if you're using it a certain way and it's solving a tedious workflow for you, for example, you know, make it more clear, make writing more clear, there's probably other people that uh are suffering uh in the same way. So, um you know, as an example, like yeah, there's probably a business that you can create that's like makemywritingmoreclear.com and it's specifically designed to make writing more clear. And you know, people like me might be willing to pay for that. So that's what you're going to kind of see with Chat GBT is it's going to be like unbundled. There's going to be like very specific um vertical agents, vertical SAS that are specific um to that. And and and you know, there's an opportunity right now and an arbitrage opportunity frankly where you can go ahead and create that, own that niche, own that uh headspace. uh just like you know the guy from you know PDF.ai AI, you know, I think this guy does like at least a million dollars a year in revenue and it's just chat with any PDF document. Um, and like of course you can do this on chat GBT, but he owns like the the the headspace, loved by millions of happy users. So, um, this could have been found, this idea could have been found via just looking at your history if you're uploading a ton of documents. And and by the way, you know, I know that there's someone in the comment section who's going to be like, "Yeah, but like all the, you know, big ideas have been taken. PDF has already been taken. A lot of, you know, h a lot of these ideas have been taken." Fine. Then just, you know, go and niche it down a little more, right? So maybe you don't just do all PDFs, but you do, um, you know, one specific type of finance PDF. Um or maybe and this is a huge opportunity is you just focus on a specific geography. You do PDF.ai for Brazil and that's your business. There's so much opportunity in that.
Uh the third thing is you can go on Google. Yes, Google still exists. Um and search site Reddit uh is there a tool and then your niche. And then you're going to find users begging for tools that don't exist. Let me show you exactly what I mean when I say this. So, I'm going to go ahead and copy this prompt here. I guess it's like not really a prompt cuz it's Google, but you know what I mean. Um, search uh site Reddit. Is there a tool? And then our niche. Let's just say we're doing a fitness. We're in the fitness niche. What is this doing? This is basically searching Reddit for free for um basically people asking is there a tool and then in the fitness niche. So I can go and see uh you know here I can see in the productivity uh subreddit you know what what tools do you use to balance uh and track daily tasks and this is around work and volunteer but also family and fitness. So you can see that you know there's um demand for that sort of thing or you can see in the running subreddit you know hi runners I'm a software engineer I wonder if there's any software tool which potentially is not available and could be helpful in your training and planning communication between trainer and runner and then you have comments uh you know to help influence uh your idea like this person is saying I would just want a free straa you know like that would be amazing okay maybe there's an opportunity to create a free straa and you monetize it a completely different way and by the way Strava I think so you know is like a billion dollar you know company so I think that you know the point is the point is you can search on Reddit and this is going to give you ideas and you do it through Google and then just write down what you're learning um and it's kind of crazy if you think about it like these people are telling you exactly what they want. It's it's literally gold.
Um, number four, audit where you copy and paste uh between tools because that friction is a startup idea. If you find yourself, for example, uh copying and pasting emails into a CRM, well, that workflow is a tedious, painful thing that you can build software for and charge money for it. So, in your own life, whenever you're whenever you see copy and pasting, um there's an opportunity for a startup idea there.
Number five, browse Upwork for repeated AI and grunt work gigs. If multiple clients want it manual, it's ripe for destruction. Um, I think that's, you know, pretty self-ex uh explanatory. Upwork is a marketplace for services and people are pay you can see how much people are paying and what they're paying for um for service. Um, and you know, I think the opportunity right now is of all these services that are happening, how can you productize that with AI? So, there's certain things that, you know, I think you could the, you know, we can do a whole separate Upwork, you know, episode if people are down for that. But the point is uh going through Upwork and and just looking you know what what are what are some repeated tasks and then how can you create agents and AI that can productize that and essentially charge a lot less and there there's your startup idea.
Uh number six, join Discord Slack groups, look for does anyone have a tool for complaints? You can even set up an automation for that on NADN. But basically looking just doing the hard work of going into groups and just looking and scrolling and seeing if people uh if people are asking for tools and you can just search within the tool uh within the group itself.
Um you can number seven you can use ideabrowser.com I built this like for myself. So you know idea browser basically is to find you know startup ideas from Facebook groups and subreddits and Google trends. It's completely free to see the idea of the day. So you know today's idea is AI bulletin boards that actually organize your chaos. So uh every single day it's going to look at trends and it's going to give you a free startup idea and it's going to tell you you know what's the feasibility of it, why you should do it. um you know, is this a big problem or not such a big problem. It goes through all the subreddits, Facebooks, YouTube to help give you the idea. And then my favorite part is it tells you exactly what the offer should be. So, you know, you want to start with a lead magnet, you want a front-end offer that's like $12 a month. Um, but the point is literally every single day we're giving out a free startup idea. So, all you have to do is just sign up, look at it, and be like, "Is this exciting to me?" And if it is, go build an audience and go build a product to go uh to do it. Or you might realize, "Hey, I actually don't really like this startup idea, but like it it it I'd like to take the start idea and kind of like put it into a completely different niche." So, I find, you know, uh I subscribe to a bunch of, you know, people or brands and products that help me in my head give me startup ideas. So, I I recommend you do that as well.
Um, you can join uh number eight, you can join niche Facebook groups and look for workflow hacking in the comments, build a tool that they're manually creating. So, Facebook groups, you know, still has a billion monthly active users, I believe. Uh, don't sleep on them. There's a bunch of, you know, uh, especially older people who, uh, are using it and and, uh, there's just a lot of alpha alpha in Facebook groups.
Number nine, uh, which is similar to, um, the Upwork thing, uh, which is browse browse Fiverr for all use chat GBT quote unquote services. So these are manual uh workflows that are apps waiting to be built. Um, so, um, yeah, that's just like, uh, pretty self-explanatory, but go on go on Fiverr and yeah, you can just see like, uh, what people what services are here and then if they have chat GPT in the like, for example, you know, I'll pay $50 for anyone who writes me an SEO blog post with chat U and I'm going to do it with Chatch GB. BT um like there's people paying for services on Fiverr and and people are actually fulfilling it with uh AI and it's happening. Yeah, it's happening pretty much all over writing, video, animation everywhere. So those are opportunities to build startups around there. Um, but I will say that this there's a a window for that and that window is probably now.
Number 10, literally talk to business owners about their daily Excel updates because that's an AI agent waiting to be built. So um the reason I I I I put Excel is because you know that is uh there is a lot you know businesses are run on Excel in a lot of ways. Um so if you just ask local business owners and by the way you can pay them you can say I'll give you $50 $100 for your time to learn about you know how you use Excel and if there's any way I can make that uh workflow a lot better for you. And it doesn't just necessarily need to be about Excel, but you can also just, you know, I that's one example. You can also just ask them like, tell me about, you know, the struggles you have in your business just generally. Like literally how you can come up with startup ideas is by just talking to business owners, maybe paying for their time, taking them out for coffee, reaching out to them and being like, you know, what are some big issues in your business? And then that might give you ideas around um around uh around your startup idea and then you might have your first customer there too.
Number 11, browse G2 for top rated legacy tools without AI and then you can build the AI letter uh layer. So G2 um is basically a review site for software. So, you know, you can see here Slack has 34,04 and a half stars and there's monday.com and you know, so what does this do? This allows you to just see like what are the issues um with uh the software and you can actually go through the reviews, what are the missing features, which is insane. You can actually like what do you like best about Slack? What do you dislike about Slack? And this can help you not only you know create a slack killer or you know um but it can also I you know I think the opportunity right now is to create AI first versions of a lot of these software pieces. So um think about like okay you know now I have a better understanding what are limiting features and missing features. Now I can actually um not only go ahead and create something that's competing, but I can actually create it in an AI first way.
Uh number 12, ask agency owners what clients beg for repeatedly. If it's annoying and repetitive, it's worth turning into software. So that's similar to what we talked about just asking business owners, but specifically agency owners. I think it's for a lot of people, especially creating some of their first businesses. There's so many agencies. Um, it's a it's an easy business to create. Um, and uh, it's one of those great businesses where you can actually go and cold email people um, cold DM them. They're they're more susceptible to it. Um, and there's o, you know, I think that building software for agencies is a really big opportunity. So, just wanted to include that in there as well.
Uh number 13, read comments on popular uh YouTube tutorials. Uh so a lot of like videos and episodes like this where you know uh a lot of people will um will watch something like this and uh you can learn about you know what friction points exist and what applications uh and software you know niche applications for software you can create. So, I think that watching uh tutorial videos, software tutorial videos is what also is like a really no-brainer way to come up with startup ideas because, you know, in a lot of tutorials, you know, they're they're hinting to what they like about certain things, what they don't like, what are their pain points. Um, and I don't think a lot of people actually go on YouTube and look for startup ideas. So, think about that one.
Um, number 14, find multi-step tutorials using eight plus tools uh and build the one-click version. So, if you uh stumble across tutorials where people are stitching together a lot of tools, well, you'll come up with a startup idea just there. And people are willing to pay for products that make it really easy and faster. Faster is we talked about uh tedious um before, but fast uh is just another code word or synonym for I will pay you to do it faster. So building the one-click version uh super super uh effective.
15. Scan LinkedIn job titles in growing industries. Uh ask what would make this person 10 times faster uh with AI? So um you know take 30 minutes and just you know look around and be like okay this person is a marketing manager at Proctor and Gamble. Okay, what do they do? You know and like you can go ahead and ask Chat GBT or Manis or or Clude and you can like literally be like we can do it right now. Um, you know, I just came across this guy with 12 years of marketing experience. He has been at Proctor and Gamble the whole time. He works in the um digital marketing department. So, you can look obviously you can see uh what he does a little bit. A lot of people put a little bit about what they do. So you can just include that there. Copy and paste that there and then just say what would make this person 10 times faster with AI. And what are you doing here? You're basically leaning on to chat GBT to do a lot of the user research and also act as like an AI co-founder. Um and it's just we've gotten to the point where it's just it's excellent at it. It's just absolutely excellent. So, how do you turn a battle hardened
PNG digital marketer into a 10X machine? June 2025 edition.
Spin up a brand-tuned co-pilot. Fine-tune GPT-40 on 10 years of, you know, Gillette and Pampers, Bree Dant. You know, look at this. This is liter—This is amazing.
Build a creative conveyor belt instead of a studio queue. Concept, storyboard, Midjourney, V7, mood boards—like this would blow some marketing manager at Procter and Gamble’s mind. Like, absolute there. This is many, you know, multi-million dollar opportunity for someone to do like wire gum loop into braise AI for always-on growth loops. They—they are not doing any of this.
Deploy an insight-mining agent. They’re not doing that. Wrap it in a brand safety compliance guide rail. Like, I’m happy that they included that. So, good. Um, and then last, lastly, number 17.
So, find two-star Shopify WordPress plugins with high usage, low—low stars. Sorry. Find two-star Shopify WordPress plugins with high usage. So, you want low stars but high usage. That’s like the holy grail. So people care about the space but the experience sucks. Um, or you clone them—focus on clone them with AI, use like Cursor, you know, Bolt, lovable—focus on different geographies and undercut on pricing.
So, the point is you can go on Shopify’s app uh marketplace and you can go on WordPress uh plug-in directory and you can literally see um what—what people think of the software and then you can just go and build competing software and undercut them on pricing or focus on different geographies and and make the tool better. Um, you know, these are people that have paid millions of dollars to create their software, but now uh in the AI age it’s a lot cheaper and uh I don’t know why more people aren’t doing this and there’s an opportunity to build like a portfolio of these, you know, Shopify and WordPress plugins that are giving you 5, 10, 15k a month uh each. So I think that this is definitely underrated um and you can create uh even like the beauty is now you can create really, really niche uh plugins and it’s not going to cost you a lot of money to build.
Like, I’m even interested in that just for fun. Like, you know, I was talking about an idea in the Shopify ecosystem too today um that I would have never been able to build two year, 3 years ago because it probably would have cost me 500k, a million dollars and I wouldn’t know if people actually would use the product um because um it’s it’s like a nicher version of something that’s popular and has low stars. But now I can actually go and do that. Um, probably cost me $5,000 to build.
So, uh, these—this is, um, this is designed to give—just get your creative juices flowing, to make you smarter about whatever it is you’re working on or to help you figure out what you should be working on. And yes, you can come up with an idea uh, just walking uh, you know, going for a walk in the sauna, I don’t know, just in the real world—just it—it hits you. And those are good ideas, but um sometimes—but what I’m saying, the ideas I like to—to work on are the ideas that I have odds stacked in my favor. So, if I know that thousands of people are saying, “I want this product,” or they’re showing me by, you know, low-star reviews that they’re dissatisfied with this product, or I can go on LLMs and it’s surfacing all the world’s information and it’s telling me in a curated way that um that this is an opportunity, or I can go on ideabouser.com and it’s just going through Facebook groups and subreddits and Google Trends and it’s handing to me on a silver platter, I’m going to go and do that.
So, um, I’m excited for you and I can’t wait to see what you build. Um, thanks for listening to me and uh happy building. I’m rooting for you.