Biggest wealth creation opportunity is SaaS

I walk through a complete 30-step playbook for building a modern SaaS company using AI agents, media, and sub-niche positioning. The core argument is that SaaS is evolving rather than dying, and the builders who win are the ones who combine a focused workflow product with a media flywheel and agent-powered execution. Drawing on my experience advising TikTok, Reddit, and building three venture-backed companies, I lay out a step-by-step framework any solo builder or small team can follow from niche selection through to becoming the default execution layer in their market. I'm hosting a free workshop so you can build your business in the age of AI. Sign up here: https://startup-ideas-pod.link/build-with-ai-2026 Timestamps 00:00 – Intro 01:18 – Step 1: Start with a sub-niche inside a big market 02:21 – Step 2-5: Map Workflow end to end 06:37 – Step 6-7: Create scroll-stopping content 10:15 – Steps 8–9: Double down on organic and run paid ads on winners 11:11 – Step 10: Capture emails from day one 11:47 – Steps 11–13: Manually perform the workflow and document every step 13:40 – Steps 14–16: Turn mechanical tasks into agent workflows and connect to real tools 14:47 – Step 17: Add orchestration, retries, and verifications 16:32 – Steps 18–19: Store user preferences and launch with high-touch onboarding 18:20 – Steps 20–21: Publish measurable proof and move to per-task pricing 21:21 – Steps 22–23: Outcome pricing and compounding value 22:07 – Steps 24–27: Expand workflows, build switching costs, create case studies 23:25 – Steps 28–30: Hire from the niche, reinvest profits, become the default layer 24:08 – Closing thoughts Key Points Start in a specific sub-niche, not a broad market — that is where sustainable cash flow lives, not VC competition. The future of SaaS starts as a service business: manually performing the workflow is how I learn what to automate. Media is a core business function, not an afterthought — content creation runs in parallel with product development from day one. Mechanical tasks are AI's strongest suit; separating judgment tasks from mechanical tasks is the key architectural decision. Per-task and outcome-based pricing is replacing per-seat models, and indie builders have a structural advantage in making that shift. Orchestration — coordinating agents, validating outputs, and resolving issues — is the new interface layer and the highest-value position to own. The #1 tool to find startup ideas/trends - https://www.ideabrowser.com LCA helps Fortune 500s and fast-growing startups build their future - from Warner Music to Fortnite to Dropbox. We turn 'what if' into reality with AI, apps, and next-gen products https://latecheckout.agency/ The Vibe Marketer - Resources for people into vibe marketing/marketing with AI: https://www.thevibemarketer.com/ FIND ME ON SOCIAL X/Twitter: https://twitter.com/gregisenberg Instagram: https://instagram.com/gregisenberg/ LinkedIn: https://www.linkedin.com/in/gisenberg/

Speaker names marked ? are inferred from speaking time and the show’s known hosts — they are not voice-verified.

• 0:00Greg Isenberg?

  I don't know what to tell you. This is the greatest time ever to building SaaS. I mean, no other time in history has it been this cheap to be able to build SaaS, build audiences around it, get it to people who have billions of billions of dollars of spending power who want to buy your product. Now, I have a playbook, 30-step playbook for where the future of SaaS is going. And, you know, why are you going to trust me if you're new here? Well, I've been an advisor to some of the biggest software companies on the planet. It's an advisor to TikTok, advisor to Reddit. I've built and sold three venture-backed companies. I've been in the arena, as Tramath says. And today, I'm going to actually just tell you, here's 30 steps for how you could be thinking about building a software as a service company now.

• 0:54Greg Isenberg?

  I'm not going to give you the silver platter. I'm not going to give you the, here's the one idea that you should do so that you can actually, you know, build it. But what I am going to do is here's the framework for it all. I'm going to walk you through, in this episode, I'm going to walk you through each of the frameworks and the steps of the frameworks for how you can go and build with AI a software as a service startup. So let's get right into it. The first thing is, you're going to want to go ahead and start with a sub-niche inside a big market. You're going to have to find a big market. And a big market could be something like, I mean, we're talking huge here, you know, finance. So what's a sub-niche? Well, you know, FIRE, financial independence, retire early, which is a movement within

• 1:39Greg Isenberg?

  the finance movement for Gen Zs, might be the sub-niche in it. So, you know, you can use AI, you can use tools like idea browser.com to find some of these niches. But you want to go, and you want to go into these sub-niches. Like, don't make the, don't make a mistake of trying to build something for a huge market, because that's where the venture boys are playing. This is for the people who want to build a cash-flowing startup that, you know, maybe they want, could raise venture later, but, you know, right now, it's about building cash-flowing startup, you know, $100,000 a month, a million dollars a month type thing. So. That's what you're gonna want to do first.

• 2:22Greg Isenberg?

  Second, you're gonna want to map this sub niche's workflow end to end. So I'll give you an example of, you know, just a local roofing company. So the owner's daily workflow end to end. And by the way, you can use AI to go and help you figure out these roofing. Sorry, the this, these workflows. So step one, check new leads, website form, Google Ads, Facebook messages. Step two, respond to homeowner and qualify the job. Step three, schedule a site visit. Step four, drive to the house. Step five, take photos. Step six, estimate materials. Step six, write and send a quote, follow up, negotiate. It doesn't matter what the exact steps are, right?

• 3:03Greg Isenberg?

  The thing here that matters is you have to map the workflow end to end. You can either do this manually, i.e. you're, you know, calling up. In this example, you're calling up owners. You're trying to ask them, you know, how do I, you know, what do you do in your day to day basis? Or you might be an owner of a local roofing company, or you might work in a local roofing company. So you might be that person yourself. So you know the workflow. So in your head, trapped in your head, you have the insights in order to actually go and do it. So and then the third option, obviously, is you can use AI. You can go to, you know, Manus and ask it to, you know, figure it out for you.

• 3:48Greg Isenberg?

  You can go to Claude Code. You can go to, you know, if you're using OpenClaw, you can use ChatGPT. So figure that out. Those are the three ways of doing it. Super important to have a dialed workflow end to end. Then you're gonna want to identify where money changes hands. Because where money changes hands, i.e., you know, in this example, negotiate scope or price, collect deposit, order materials for suppliers. That's where you're gonna be able to create some like wedge in software and sort of take a piece of that pie. So that's an important piece here. You can highlight that.

• 4:28Greg Isenberg?

  You could be doing this on like a FigJam. You can be doing this on an Excalidraw. You can be doing this on pen and paper where you're writing out these workflows. Google Docs. It doesn't matter, but highlighting where money changed hands is really important. Step four is spotting repetitive mechanical steps. Why do you want to do that? Well, you want to do that because that's where there's an opportunity to do some sort of, you know, agent or automation. So you're going to want to spot those and then, you know, again, take note of what that is. So for example, checking new leads, website form, Google Ads, Facebook messages, phone calls, that might take five minutes every day or 10 minutes every day. And if you can save people in this owner 10 minutes every day, what's that worth to him?

• 5:12Greg Isenberg?

  Well, if that person is making $250,000 a year, $500,000 a year, a million dollars a year, you can literally quantify that and it becomes thousands of dollars. And then, so if you can say, hey, I'm going to save you 50 hours of time this year, 100 hours of time this year, 150 hours of time this year, what's that worth to you? And if you're, you know, chart, if that person's time is worth $400 an hour, then you can go and just be like, well, I'm going to go ahead and save you all that time and you're going to be able to make more money. So that's step four. Step five is you're going to quantify the cost of those steps. So I kind of just talked a little bit about that. Quick break to invite you to something. Now, this isn't an ad.

• 5:56Greg Isenberg?

  I just want to invite you to a free event because I think that you're going to get a lot out of it. I wanted to take one hour of time where we just talk about building businesses in the age AI. People say SaaS is dying. I actually believe the quite opposite. I think that SaaS is just evolving. I think right now is an incredible time to be building software startups that help you craft your dream life. And for all those reasons, I said, let's just book one hour of time. It's going to be 11 a.m. March 12th. That's a Thursday where we can go and lock in and just talk about building businesses in the age AI. I'll include a link in the description and the show notes to join and I can't wait to see you there.

• 6:37Greg Isenberg?

  Step six is create scroll stopping content around that workflow. You're going to not just want to create a product, my friends. No, no, no, no, no. You're going to want to create media. You're gonna go on ahead and, you know, create an Instagram, create a TikTok, create an X, pick one channel that you're gonna focus on and build an audience around that. It's easier said than done. Create scroll stopping content. Like, I know what you're thinking, like, easy, you know, that easy for you to say, or that sounds easy, you know, easy, but it really isn't. And you're right. But the truth is, you couldn't use AI to go and research,

• 7:18Greg Isenberg?

  ask it, you know, in this niche, how are people going viral, right?

• 7:24Greg Isenberg?

  Give it, you can even ask, you know, a Manus, a Claude Code, a ChatGPT, like, you know, what, just like content ideas. What are some content ideas that would go viral in this niche? And ask it to do, create a calendar for you. So you can use it as, you know, an AI CMO. And, you know, the best, the best thing you can do is, in my opinion, use something like Manus, or use something like Claude Code,

• 7:57Greg Isenberg?

  and, and create some automations around making, making a repetitive task around, you know, getting content ideas, giving script ideas, and, and creating content. You can use AI to create some of that content. You know, I saw, you know, I'm starting to see AI videos that are actually like, you can, you can barely tell that they're AI videos. You can use AI videos. You can either do it yourself. You can record it yourself. You can do faceless content. But the point here is, and, and, oh, and you can also use something like an Open Claw that actually just does this all repetitive for you, Claude Cowork.

• 8:38Greg Isenberg?

  I could do a whole episode on this number six. People let me know if that's interesting. But the point here is that you're going to use some AI automation to come up with ideas for you, research for you, scripts for you, ideas, and you're gonna push it to the limit. You're not just gonna be like, give me content ideas. No, you're gonna push it to the limit and ask it to give you non-obvious ideas, consistently providing a context, and then doing, schedule a task, so you're getting this on a daily basis. You're gonna wanna study step seven. Study posts that get saves, replies, DMs.

• 9:15Greg Isenberg?

  Sounds obvious, but a lot of people don't do it. You know, I recently have been posting on Instagram over the last six, seven, eight months. And I look at, I can like pull it up here. Like I look at, uh, pull it up here, like the insights, uh, view insights. And I could see right here, um, in my latest video, it's got 2,700 likes, 5,000 bookmarks. And I could just see, uh, basically how it, how it compares to other videos. And you, you develop this intuition around your content. You develop your intuition content. And by the way, why do we care about content?

• 9:59Greg Isenberg?

  We care about content because yes, you're going to build an audience and then hopefully you can sell this tool that you're building. We don't want to forget, we're building a SaaS tool. But the, the other piece of it is when you create content, that could be used as ads, right? So you're going to want to double down on organic angles that convert, but you're also going to want to run step nine, paid ads on the proven organic winners. Because it, once you've shown that you can prove organically that you're getting some traction, there's a good chance that those organic pieces of content will be able to, uh, work as standalone ads with some good ROI.

• 10:44Greg Isenberg?

  Now, I'm not saying every single organic piece of content that goes viral, or when I say viral, but I mean, I mean viral within your niche. So that could even be 60 likes, 80 likes, 120 likes. Depends how big your niche is or small your niche is. Point here is that, uh, there's a good chance that if it works on organic, it will work on paid. And that's, that's going to be huge for you. Mistake a lot of people make, uh, step 10, capture emails from day one. Your email list is your, it's like your foundation, you know? When, when sales are down, you can just go and email that, that, that list and be like,

• 11:28Greg Isenberg?

  you know, we're doing a discount or we're doing an event or, uh, You know, because you can't rely on social either ads or even your organic that it's gonna work, right? So having that list of people who are interested in your brand, in what you're doing is going to be key. Step 11, manually perform the workflow. So you know, now that we've mapped out this workflow, you're gonna wanna manually do it. Now, why do you wanna manually do it? Manually is gonna help you, you know, basically learn about how to actually perform this piece of work.

• 12:11Greg Isenberg?

  So when I say manually, what I mean is, a lot of these future of SaaS businesses are actually gonna start off as service businesses with human beings at the core of it. It's actually gonna start as being you, the person to actually go and fulfill the task. And I know that might not be sexy, and I know that that might, you know, people look at that and be like, well, that's not what I signed up for. I signed up to create this software, that software business that is gonna work 24/7, is gonna have high enterprise value, that allowed me to quit my job, is gonna, you know, hopefully sell for a lot of money, you know, one day. Well, yes, that could be one reason, you know, why you wanna do this.

• 12:54Greg Isenberg?

  But the point here is that you need to really, really be dialed around how to actually perform the outcome, right? So, you know, maybe you've done this in a past life, like, you know, for example, in the owner of the roofing company, maybe you've, you were that owner in the past, so you know how to do it, but you're gonna wanna manually perform it probably from the start. You're gonna wanna, step 12, document every step precisely. And step 13, you're gonna wanna separate judgment from mechanical tasks. This is a really important step. AI is incredible, incredible at mechanical tasks. And in judgment, sometimes it's good and sometimes it's bad.

• 13:37Greg Isenberg?

  So you're gonna wanna start with mechanical tasks, which leads me to step 14, which is you're gonna wanna turn your mechanical tasks into agent workflows. That's the opportunity here. If you can create those agent workflows, automate a lot of that stuff, that's where a lot of the value is gonna be. Step 15, you're gonna wanna design agents to complete the full tasks.

• 14:00Greg Isenberg?

  Step 16, connect agents to real tools. So this is using things like MCP. But the idea here is your agents are going to need skills and tools to actually go and perform these tasks. So that's why you might need to have it hook up to email, have it hook up to Slack, have it hook up to CRM, have it hook up to Stripe. You can obviously imagine in the example I gave with the roofing company where you'd have things like, you know, email integration like over here, you know, Facebook integration, you know, Stripe, et cetera. It's going to need access to that. In the beginning, you're going to do this manually, but eventually the agents will have

• 14:45Greg Isenberg?

  access to these tools. Step 17, add orchestration, retries and verifications. I actually saw a really good tweet by my good friend Scott Belsky on orchestration. Let me see if I can pull it up. First of all, Scott, legend, used to run product at Adobe, one of the best, you know, seed investors I know, invested in some of the biggest, you know, consumer companies that, you know, you've, you probably use. And he says, "The orchestration layer is the new interface layer. As we spend our day coordinating agent workflows in a model agnostic fashion, local and cloud,

• 15:27Greg Isenberg?

  and validating outputs, human in the loop and resolving issues, the ultimate layer to own is where coordination takes place." So I think he's absolutely right. And I see some replies here. We've entered the conductors era. So a lot of talk right now on this orchestration layer, orchestrating agents. And that's kind of what you're going to be doing when, when you're building your, your new software, you're going to be orchestrating the agents. The outcomes and the resolutions are the end goal.

• 16:08Greg Isenberg?

  And I think this is just, yeah, a new paradigm, a new way of thinking around building software. And yeah, here's another good reply. "If coordination becomes the choke point, whoever owns that layer owns the flow." So I cannot agree more. I think this is like where it's going.

• 16:26Greg Isenberg?

  So yeah, going back to here, you're going to step 17, you're going to want to add that orchestration, retrys and verifications layer. Step 18, you're going to want to store user preferences. So as people are using, using the product, that's going to be a little bit of your mote, the user preferences, how they're using it. And then the memory, long-term memory into your, you know, agent SaaS is going to be an important piece as well. You're going to want to step 19, launch narrow with high-touch onboarding. So,

• 17:00Greg Isenberg?

  you know, it's very tempting to do onboarding, like a signup flow and stuff like that. That's very short, sweet, but you want to really get to know the customers, get as much data as possible as you, as, you know, when it, obviously, you're not going to be able to like,

• 17:17Greg Isenberg?

  ask them a thousand questions. So you have to basically figure out like, what is the right amount of asking in terms of onboarding to make the experience good, but also to create a mote for you. Because what I don't want you to do is create something that has basically a very, like, you're not getting the right data into the product, or you're expecting it later. And that also, you know, doesn't make the experience good for the customer, could be better. And/or competitors come and just knock you out of your stocks. By the way, while we're, while we're building the product, we're creating, just, just so we're

• 17:58Greg Isenberg?

  on the same page, we're creating content like what we talked about before, we're trying, we're creating, you know, putting up one piece of content minimum every single day. We're trying ads on the ones that work and we're trying to really understand and build that,

• 18:14Greg Isenberg?

  that foundation of a media company at the core of this.

• 18:20Greg Isenberg?

  Step 20, publish measurable proof. So, you know, going back to like, remember we talked about how you're saving time for, for the owner of the business. Well, what is that in terms of revenue hours saved? Step 21, you know, this, step 20, 21 is move pricing for per seat to per task. And this is one of the reasons why a lot of SaaS companies in the public markets are down 30, 40, 50% from all time highs is they're scared that not only are people going to be able to vibe code or, or, you know, use AI to create competitors to, to that, to whatever product it is, Salesforce, whatever.

• 19:04Greg Isenberg?

  But it's also, the per seat model is sort of losing its allure and people just want per tasks in AI age. Do this thing for me, like, complete this workflow and this workflow to me is worth $200 every time you do it. And that's, that's where it's going. And you, as the person building this future of SaaS, you will be able to compete better with the, against the existing products in your space for two reasons. One is, well, for a few reasons. One is, we've gone more, now we've, we've gone more sub niche, right?

• 19:44Greg Isenberg?

  So you can compete against the big venture backed people because you're a bit more bespoke. You are building media at the core of what you're doing. So you have this unfair advantage around not only an audience, but you're building, you know, email newsletter and you're also doubling down on paid ads. You, you're, you're able to basically do better ads than a lot of the competition.

• 20:14Greg Isenberg?

  And then you're doing, you know, using AI and agents to actually fulfill the work. And that's better than having humans and you'll be able to compete on. It's better because, you know, if, of course, AI hallucinates, right? But if you, if you do it right and you do it in a small way and you don't overload with context and you're doing checks and stuff like that, you will be able to get incredible results. So I think because of that, and also it's going to cost less, your cost structure is obviously less. You don't need to raise millions of dollars if you don't want to.

• 20:57Greg Isenberg?

  You don't need huge teams if you want to have people. You'll be able to be more cost effective and you'll be able to move from a per-seat basis to a per-task. You don't need to start by per-task if you don't want to. You can actually start by per-seat if it's easier or whatever. But my point here is eventually you're going to want to make your way to a per-task basis. And that's going to shift you, step 22, to outcome pricing. That is the goal here. Step 23, as your value compounds, as you go and add more workflows, as you go and get, do a better job at adding workflows, as you build your brand so you have more trust,

• 21:40Greg Isenberg?

  you can increase pricing as that value compounds. Once you've finished this workflow, say you've actually gone and just, you've done it. Like this, you've built this really good AI agent software that goes and does this better than anyone. And it works and you feel like you probably can't get more LTV out of it, given the product offering. That's when you can go and explore basically increasing into different adjacent workflows, adding different adjacent workflows. People here also can think about, do I want to buy another company? Do I want to build it myself? You know, a lot of time here, you know.

• 22:24Greg Isenberg?

  You might not want to buy something. You might just be like, AI is so easy to create the product now. You might just want to build the product yourself. But this is going to lead you to step 25, orchestrate multiple agents across the lifecycle. Step 26, build switching costs through data and memory. We talked a little bit about that before, but you're going to always want to layer on more data, more memory. Now you have multiple agents across the lifecycle. You're probably in a few different workflows. Step 27, turn those power users into public case studies. So by this time, you probably have these power users, hopefully some that are well-known in the sub-niche, and turn those into public

• 23:07Greg Isenberg?

  case studies. And when I say that, don't just create articles, tweets, that sort of thing. I'm talking about like, full-on, send a camera crew, or go yourself and film these people, and showcase that, and then put paid spend around that as well. Step 28, hire operators from inside the niche. So at this point, you're probably making money, profit. Go and reinvest that in hiring operators. Step 29, reinvest profits into distribution and product depth. You're now probably in an adjacent workflow. But you can also just

• 23:49Greg Isenberg?

  go deeper, and then also start spending a lot more money on content, a lot more money on paid ads. But when you get to step 30, the goal here is you've become the default execution layer for that sub-niche. I don't know how this doesn't fire you up. If you've made it to the end of this, I hope are seeing this and being like, yes. SaaS is being disrupted right now. You know, people say SaaS is dead. Like, I don't know if SaaS is dead. It's not, it's not dead. It's evolving. And this is what it's evolving towards. I've handed you the playbook and the framework on a silver platter. I cannot wait to see what you build.

• 24:35Greg Isenberg?

  I am totally rooting for you. I'm so excited for what, you know, what you do with this, what you end up building, and how it ends up creating impact in whatever niche you, you know, you create and whatever communities you you pick, and in your own family and personal lives. My name is Greg Isenberg. This is the Startup Ideas Podcast. If you enjoy this, give it a like, comment, and I'll do, you know, keep doing this, do more. I have so much fun just sharing as much sauce as possible. And I thank you for, for being here with me, for spending your time with me.

• 25:13Greg Isenberg?

  And I won't let you down. I will just continue firing sauce into your ears, if you're listening on Spotify and Apple, and into your, into your corneas, your irises, your eyes, if you're on YouTube. This has been awesome. Future SaaS. That's this episode of the pod.