Build Profitable Apps (that actually make money) in 2026

What's going on, guys? This is Oliver, formerly from Response AI and a few other software tools that I've since exited um and now running a few different software tools and running roseell.dev. Um I want to address a really important uh problem uh on the channel. So, customer acquisition, how to actually build a software that gets customers and then get people to pay for the software you've built. because a lot of people have commented, look, can you do some videos on marketing or customer acquisition or, you know, just getting clients for the tools that you build? Because I do a lot of videos on how to build them. Um, but if you go back, there's a lot of videos on my channel about getting customers, but obviously since then, I've got a lot of, you know, new fans and subscribers, that kind of thing. So, code is easy now. AI uh makes things easy. Infrastructure is cheap. Backends, front end, servers, whatever it may be. What remains hard, guys, is getting customers, right? So, I'm going to be covering first principles, playbooks, and then case studies from other founders who have actually gotten their first couple of customers or first couple of thousand customers. Right? So, most solar founders fail before they ever need acquisition, right? They build something that no one wants. So, every tactic in this guide assumes one non-negotiable, which is that demand has been validated with real conversations, real money, not just interest. But I'll explain how to do that. So what real validation looks like across the best bootstrap stories uh you know because we're assuming that we're solo founders um validation has consistent markers right prospects feel an acute pain without the product or they use a competitor uh so they're already paying for a competitor or they've talked about you know how painful it is to not have a solution they've hacked together stop gap solutions so scripts or spreadsheets or automations uh sort of duct taped versions of what you can build and they say versions of If you build this, I'll pay or is anyone building this? Some do. Even when the product is early or imperfect, they will buy from you. Right? So, for example, when I built Response AI, which scaled to well over 500 customers, paying customers to validate my idea, I just DM'd cold email agencies and sales freelancers asking if they used video prospecting, right? And what tools they relied on. The responses from those people showed that they were doing video, but they hated the time cost of manually doing like Loom videos and that kind of thing. Or some had even sort of like hashed together Python scripts or ways of automating this process, but it was really really like tedious. Okay, now that's real demand, right? So people were already doing the thing I could offer them with my software, but they were doing it in a mishmash painful way. So step one is to talk to 20 to 30 people in your niche, right? And I don't want you guys to skip over this or think it's boring or think it's cliche, right? You just need to DM a load of people, right? Talk to people. You don't have to get on calls with them. You don't have to like, you know, do anything. You just have to ask them questions. So, um, if they're tweeting about it, people these days will post or tweet or post on Reddit like, I'm having these problems. That's absolutely fine. Um, but you can also obviously pay for this. So, I I typically say to someone, "Look, I'll give you $20 or $30 or $10, whatever it may be, or $50 to get on a call with me or to just have a conversation with me." They don't need to get on a on an actual video call with you or a phone call. You can just ask them for their time or ask them to fill out forms about your product, right? So, ask about workflow. So, how are you doing this solution today? So with response AI, I said, "How are you guys currently building, you know, um video uh and like automated outreach, right?" And they said, "Um, we're doing this or that or we've mishmashed it together or we're using automation. What's annoying or slow about that? And have you tried solving it?" Right? So you just want to try and get a clear understanding of what they're doing and whether uh what they're doing is painful. If what they're doing is working and it's fine, then don't bother. Right? Now a good validation sequence um that you can use perplexity for or Gemini or chat GBT is this um formatted way of actually getting the AI to give you ideas. Right? So do intense deep research on social media tools. Let's assume assuming we're making one, right? I'm building one as a solo founder. I need you to find me hundreds of relevant reviews and negative reviews about these products and return them to me so I can guide my next development process. I need you to do deep research on the key features. Not complaints like support denance me or I don't like the color of this or don't like this button. I mean actual complaints or things that people love about the product so I can copy the key elements of successful apps and build my own. So obviously what it does is it tells you all of the different competitors that you have for that specific idea and then it tells you and finds all of the different reviews. So people saying I don't like this or this is my favorite thing. Right? Some of the best ways to build software, guys, is to find things that people love about a certain software. So, what people love about Linear, a really, really great um sort of like product management app, is how fast and simple it is. Okay? So, it's almost better to listen to what people love than what people hate about certain products. Now, like I said, look for evidence of DIY solutions, right? There's always people on Twitter giving away these templates for um automations like n etc. Right? So scripts, templates, air tables, zappia, um you know, n templates, hacked Google Sheets, whatever it may be, some variation of we wrote our own thing ones. So this is the strongest signal that people will pay for a managed ser uh you managed solution. Another way of doing it is to go on X and just type in something like is anyone building I've seen maybe three or four tweets today that says is anyone building this or that, right? I saw one that said is anyone building a tool to uh transcribe podcasts? Right? And there's three or four people commenting, you know, following this or yeah, interested too, right? That's like five people who want the thing, right? Whether they would buy it or pay for it, you don't know, right? But the whole point is that you can find people in these treasure troves online that are asking for solutions. And again, just to be serious, guys, with the NAN automations, for example, look, straight up plug-and-play systems for lead genen content creation, CRM updates, Slack, whatever it may be, right? Every single one of those, hundreds of people replied and commented on that thing, right? that means that they need those N810 automations, right? But it doesn't necessarily mean that they need to use N8 for that. If you say, look, I know you're using N8, but we have this offer um that helps you do the AI workflows or do the content creation um in a pretty dashboard, people will people will, you know, pay pay you for that guys. Okay, so ship [snorts] bare bones the product fast MVP in days only the core workflow that creates value. So if you're doing a social media scheduleuler, you have to be able to schedule the posts. If you're building a CRM, you have to be able to save and update leads. Um, if you're building a product management tool or a Tik Tok reel creator, you need to get down to the core workflow that creates the value. Okay? And on boarding is structured to get users to the key win as soon as possible. So, for a Tik Tok reel generator, for a YouTube transcription tool, um, for a CRM, um, onboarding your structure to get those users to that specific goal. So, searchable.com guys, it's this new software. Um, I don't personally use it. I just completely sort of like uh deep dove their entire onboarding flow. Searchable is a tool that helps you, you know, helps your brand uh get ranked on like chat GBT and stuff, right? So, if someone asks for best supplements for um, you know, men in their 50s, the searchable is helping you rank for that in chat GBT. So, your product gets recommended, right? Um, basically on the on boarding, it tells you to input your website, it tells you to input some competitors, it tells you to input keywords, and then it makes this beautiful onboarding flow and then tells you exactly what to do next and how to rank on chat GBT. And I recommend that you guys go through that searchable, it's free, go through that searchable.com onboarding to see what a powerful onboarding pro process is like. Now, the [snorts] barebones MVP, right? So common across the examples first version solves exactly one painful workflow end to end. Okay. No dark mode, no you know random stuff, vanity features, colors, whatever it may be. UI can be uglyish as long as it reliably delivers the promised result. Okay, Response AI's initial version, guys, was record one video with your webcam. You upload a CSV of leads and it generates those personalized videos with the website overlays. This is a long time ago now, guys, when these tools were very very mish mash and very sort of like rubbish. Okay, no fancy dashboards, no complex segmentation or UX or design or or billing portals. It just said, "Do you want to subscribe?" Yes or no? And then obviously it went from there. These days, it's much easier to build. And I built Response AI pre chat GBT and pre-coding era. Right now, AR accelerated development. Okay, so we're going to talk about how to get actual customers soon. But the modern stacks um would be Nex.js JS cursor lovable clawed code backend info I always use is superbase you deploy with versella netlify and payments with stripe right so many founders literally screenshot competitor dashboards and tell AI tools to build me something like this and then iterate from there now let's get our first customers guys okay so the self-fulfilling lead magnet this is how I built responseai it's how I built getseer.com it's how I built tricom.com it's how I'm building new software now A particularly effective pattern that I used is the self-fulfilling lead magnet on X and LinkedIn and Facebook. It is a free asset that only your ideal customer cares about. For example, with response AI, three cold email scripts that convert. Only people doing cold email will opt into this. Hence the self-fulfilling aspect of it. Example for an onboarding software user on boarding tearown checklist or list of best app on boarding experiences. All right. From there, what they're going to do is you're going to create a genuinely useful asset for them. You gate it behind an email sign up or a like and comment, you know, expost or a like and comment LinkedIn posts. Those goes those go absolutely insane, guys. They get hundreds of comments. Promote via Twitter or LinkedIn or small ads or communities, whatever it may be. Anyone who opts in for that is pre-qualified for your core product because if someone says, "I'd like to see this PDF on the best user on boarding experiences." They probably care about on boarding guys. If the free thing is good, they think the paid thing must be better. Now, the next aspect, so obviously I would recommend posting one to three of these per week. For Response AI, I had one that I posted almost every day. I said, "Like and comment for this and you'll get this free list or you get this free template or this free whatever." The maddest one I ever did, it got me 150 um opt-ins was a a offer to write their cold email scripts for them. Okay, which was crazy. I was spent maybe two days writing the scripts and then inside the document I said if you want to get better results with cold email by Response AI. Okay. So that's that's the basic sort of like psychology of it guys. You're giving something away for free. You are very aggressively um you know pitching the product and plugging your product. The whole point is that you get people to opt in and get people to you know notice what you're doing. Anyone who opts in is pre-qualified and if the free thing is good they might buy the expensive thing. Right? So founder sales and scrappy channels. So, um, all of the tools that I'm about to, um, reference, I'm not affiliated with, um, but most of them are free, right, which is the most important part. So, patterns seen across founders who get their first couple of clients, right? Or first couple of users. Engage publicly. So, reply to posts. This is maybe a waste of time, guys, to be honest. Like, this is why I've popped this here. I never like talk to people on Twitter or talk to people on X unless it's something interesting, right? I certainly don't do it to try and pitch. I will just raw send DMs on Twitter or X, you know, whatever it may be or Reddit or Facebook conversions of say a 100 messages. You might get two or three um you know free trials or that kind of thing. Saw you posting about X. I'm building a tool specifically to solve that. Would you be open to showing me your current workflow or whatever it may be? You can either pay for calls or you can just ask for them to sign up to a free trial. Here is the exact flow I use right for every single software. I sign up to XR reacher which is free. You can sign up to any of the free you know DM tools but this is one of the only free ones that does like volume. Be careful about getting banned or no but not banned. Be careful about getting sort of restricted right. So make sure that you're posting on Twitter regularly and that you you know have followers and that kind of thing. Being premium for $8 a month does help as well. Scrape followers with XR reacher of specific influencers in your niche. filter by building or founder or built in bio DM while also engaging on X. Right? So on top of that manual Reddit posts using Claude post two times per week on relevant subreddits. Anyone who comments or you know or upvotes and comments and says this is great DM them. Free trial of hey reach on LinkedIn to validate with cold LinkedIn DMs. If interest then upgrade to the paid versions of hey reach because it's quite expensive. cold email. So, I'm building a tool called paper schedule at the moment and I've bought dozens of domains. I think I bought 20 domains. The exact flow I use for cold email is buy cheap.info domains with pork bun that forward back to your main domain. Set up 10 Google Workspace accounts for the domains. A free account with Apollo.io. Find a perfect list using technographics because Apollo.io has amazing filtering, you know, capabilities. Then use Ample Leads, a a sort of it's kind of like um it's just a website that helps you scrape Apollo, right? You scrape that for cheap and then you upload to a sender like instantly or smart lead, right? So depending on how many, you know, emails you want to send, the cost for this monthly would be say $ 20 to $50, maybe more. So that's for someone who has a bit of extra cash. Now I want to talk to you guys about real solo founder playbooks, right? And sorry my my nose is itchy. Now the core idea got to alternate weeks of building and marketing. So so John Yongfuk um built Banner Bear which is 50K plus monthly recurring revenue. Right? It's amazing. Now [snorts] here is exactly how John did it. Right? So free utility tools. The first thing he did was he created like little tools um that you know kind of funneled people back to um his original asset which was obviously Banner Bear. So free tools, generating certificates, banner size calculators, OG image generators, and all of these things acted as SEO magnets, right? So have a little think guys about what free tools you could build for your website because you can code them in seconds, right? With um with cursor or with lovable, whatever it may be, and just pop them on the website. An example that I did with response AI, I built a tool that allowed you to um find the you know verify an email. So just verify one email of a specific email. So it just says verify this email. that got a lot of um traffic because people were coming to the website to verify emails one at a time which isn't ideal but then obviously it would say underneath it are you looking to verify thousands of emails and send emails with response AI and that's how you get initial users who want to use a simple tool to use your full tool right from there then John that's obviously the free tools tactic which is insanely powerful guys a lot of people do this technical tutorials on ffmpeg or a puppeteer or web hooks, whatever it may be. They rank in Google. They attract devs who then discover John's product. Weekly alternation between coding and then marketing and no mixing to avoid this context, you know, thrash. So, I'm going to market one week, I'm going to code another, right? If you guys are better with like, you know, um uh what's it called? Uh like managing different tasks, multitasking, then you can code and market, but I typically need to code and build the product and then go to market. transparent metrics and updates on Twitter, that kind of thing. So, we lost this customer, we gained this customer. Um, you know, the here's how we got more customers this week, that kind of thing. Why it works for solo founders. So, you've got compounding free tools and articles that bring traffic long term, right? And people are using stuff that you build. So, that builds trust. Oh, if I'm using John's stuff for free, then maybe I should use his paid stuff. Low marginal cost. a one-time build for those tools, then it's just occasional updates to the tools or releases, that kind of thing. It works especially well for dev facing tools and APIs and stuff that developers are going to use. And how you copy it is you build two to three genuinely useful free tools that are tightly related to your products, you know, job that it does. Okay, so with response AI, because it was a cold email tool, I built a cold email generator. I built a cold email finder, a cold email validator, verify, whatever it may be. That's three tools that people are going to come back and use regularly and it ranks on Google as well. Now, case me, right? [laughter] Response AI, CEO, comb, capern, a thousand combined customers with response AI at six figures before I exited it. Right. The core idea scale video content and DMs and written content giveaways to be everywhere. Okay, so 50 DMs on X per day, five posts on X per day. So I schedule these two to three months in advance. I have one rainy Sunday where I just schedule a load of tweets, right? Two self-fulfilling lead magnet giveaways per week. So DM all those people who interact, who tweet or comment or like, whatever it may be. Cold email 200 to 500 per day with Pork Bun and Google Workspace. Then if you go back guys on my YouTube channel, you'll find a few videos of me um talking about Response AI. I'm I'm literally pitching Response AI, right? And this is before I got into the coding side of the channel. Niche adjacent YouTube videos, two to three per week on a fresh channel. Okay, so basically videos informative about how to build software or how to, you know, manage leads or how to build content or how to create things, whatever it may be, whatever your tool is, you need to create content around that. and then very very softly um you know post the link inside the description and plug it inside the video like by the way guys if you want to do X then download this tool or or you know sign up for this tool no SEO or paid ads I don't really do that um until I get a decent you know wedge of money in the bank with response I made sure that we were at say 5k per month in revenue before we started spending money on ads it's really important that you do that because you might you know run up paid paid ads and get no responses from it or you know or or signups but you have spent a lot of money. Now Arvid Carl, right, I love this guy. Really, really cool guy. He built Feedback Panda, bootstrapped it to seven figures, exited it. He has a book uh called Bootstrap Founder. Um and he is amazing, right? The tactic was niche content and personal branding. So specifically, don't clock out because it's really important. He wrote a blog and a newsletter focused on SAS bootstrapping, that kind of thing. and he published a book, right? So, the bootstrapped founder book and it shared insights on Twitter and LinkedIn um and how to build a community and how to build a software, right? He used content to attract and convert software customers as founders, sorry, SAS founders as customers and he bootstrapped Feedback Panda to 20KMR. Right? So, the whole point here is that you become some kind of little personal brand around a specific topic. So for me, I made sure that I was a personal brand around cold email before I built Response AI. Nathan Barry, right, eight figures, multiple eight figures, literally worth millions and millions. He built ConvertKit, right? So he created a free course on email marketing to attract leads. So consider guys that roseell.dev, right? Rosewell.dev is my course and I teach people how to build software. Now, I won't do this because I don't know how to build it, right? But imagine if I released a competitor to cursor and said, "For $20 a month, you can code, right?" I wish I could build that because it would be insane because everyone who bought my book um would probably look or check out the coding tool that I built because I'm teaching people how to code and then I'm showing them how to do it with my tool. You see how that works, right? So, Nathan Barry created a free course on email marketing. Then he used the sequences to nurture leads and convert them into paying customers for a tool about email marketing and he grew convert kit to 300 million you know three million AR but it's way more now right so offer highv value content around the tool and then let the product sell itself now here's some final sort of words on pricing and unit e e e e e e e e e e e e e e e e e e e econ economics, right? So don't underell across all these founders. One pattern is consistent, right? No one scaled on $7 a month plans. Typical shapes entry tiers in the 29 to $49 a month range and then serious tiers 99 to199 annual plans with 40 to 50% equivalent discounts to frontload cash and reduce churn. Low prices attract side project tourists. High support low retention massive overhead and issues. High enough prices filter for serious users. Make foundled support economically viable because you've got to be paid enough for your time. And they make affiliate partner economics work. So, if you want to affiliate with someone on a $7 product, it's not going to make any sense. Now, here is the 30-day execution plan, guys. To tie everything together, here's a concrete road map, right? So, ignore that 12week. Actually, that's a typo. Week one, validation and ICP. Talk to a few users on DMs and calls. Document their wording around the problem. Identify any DIY hacks that they use with NA10 or Zapier, etc. Lock in a narrow ICP and a simple positioning statement. We help roll that company do job without pain. Week two. Use AI assisted tools like cursor, superbase, stripe, whatever it may be to ship a barebones MVP. Create a self-fulfilling lead magnet checklist, script pack, mini tool, whatever it may be. Put up a landing page with the value proposition, lead magnet opt-in, clear description of the target user, whatever it may be. And use data fast for analytics to check how it's going. Start posting, building public content, and sharing the lead magnet around with within week three. I think before that, in fact, you'll have your first paying customers. So, launch quietly to the lead magnet list and DM contacts. Offer a discounted early adopter or lifetime plan to five to 10 people. Run structured calls with them to observe usage and refine on boarding. Get your brother or your sister or your mom or your friends to sign up to the software and see if they can figure it out themselves even as non-technical people and you instrument the product to measure time to aha and retention. So time to aha meaning how fast can someone get value from the tool. Then on week four choose one primary acquisition channel. So for B2B it might be cold email, LinkedIn. For dev tools it might be technical content, GitHub, that kind of thing. For creator tools it might be YouTube and affiliates. for broad SAS it might be content product hunt launching on different uh websites etc right so that is my guide guys any problems at all give me a shout any questions I cover a lot of this in the roseorld.dev of course and I'll see you guys soon and take care.