 
            Rashid Khasanov suffered a defeat after his first app crashed and burned. But he picked himself back up and built Angel Match to solve a problem that he had faced with that first app.
Today, it's bringing in $37k MRR. And he has three other profitable products to boot.
Here's Rashid on how he did it. 👇
I am a non-technical, bootstrapped founder who's been building SaaS apps for the last six years.
I've always wanted to run my own business. Even as a child, I made kites, built bicycles, bred parrots, and so on to make money. My first app wiped me out completely. I lost almost everything, drained all my savings, maxed out credit cards, and went deep into debt. But I didn't quit. I continued to build small tools on the internet and learned a ton in the process through failures, iterations, small wins, and consistency.
Today, I run angelmatch.io — a fundraising platform that helps startup founders find the right investors, manage outreach, and share updates. It was born out of my personal frustration of struggling to find investors for that first app that failed.
I also run a few other SaaS tools including Investor Hunt, Journalist Hunt, and a social media scheduling tool called Pur Social. I like building SaaS products that solve real painful founder problems, especially database platforms. Building a database product is not easy, but if you get it right, it will have a moat. I see tons of AI wrappers get killed when big LLM players launch new features within their platforms. Curated databases are harder to kill.
Right now, Angel Match is at $37.3K MRR. We dipped from $41K MRR in September after pausing Meta ads for about a month to evaluate how profitable our ads actually are, and also we changed the person who runs ads for us. We will resume ads in November. The goal is to get to $100K MRR. With churn reduction that we are working on and a new ads strategy, I think we have a real shot to scale it.
Investor Hunt is the second app in terms of MRR, and it's at $4557 MRR now. Pur Social is at $491 MRR. And Journalist Hunt, a database of 100,000 journalists for getting media coverage, is at $180 MRR.
Before Angel Match, I teamed up with a few people and built Investor Hunt in 2018. We had cofounder issues, and the app was shut down. The usual story — someone didn't want to share the money, and the others didn't want to do the work. I knew that Investor Hunt could grow, but everyone had to be aligned, and unfortunately, we were not.
So I stepped out and found a remote software developer and built Angel Match from scratch. Soon after launch, it started generating revenue and I never looked back — though I happened to own the investorhunt.co domain, and later restored that as well.
The initial version of Angel Match was very basic. We had tons of bugs. I designed the website and the developer built it. We use Node.js on the front and back end, Digital Ocean for hosting, Figma for design, Stripe for payments, and Klaviyo for marketing.
Aggregating investor data from different platforms into one place was the most difficult part. Over time, we've aggregated 110,000 angels and VCs. We've also built tons of free tools and calculators, a fundraising CRM, email outreach, and other resources.
We operate with a small remote team who help me from design and copywriting to front-end and back-end development.
If I had to start over, I would not start with a shiny, big app that could potentially become a unicorn. Because your probability of success becomes extremely small. You have to rely heavily on factors like fundraising, compliance, team, speed of growth, etc.
Our first app was an SEC-registered robo-advisor, which was very expensive to run without VC money. So if I could go back, I'd start a simple app that solves one problem and I would test if people were willing to pay for it. I would try to build many of these types of small software solutions and put a paywall behind them to find the ones that stick. Then, after finding the product-market fit, I'd focus on SEO early on or find any distribution channel that works.
In other words, test a lot of things and double down on what works. 
I'd be more careful with hiring key people. It's very, very important to find competent people if you want to grow without losing years of your time, money, sleep, and nerves.
We run subscription-based SaaS apps, and we have grown our revenue mainly by working on things that increased our organic traffic, referral traffic, and ads.
We build free tools and launch them on Product Hunt, Indie Hackers, Reddit, Startup directories, and many other platforms.
We publish blog posts consistently.
We schedule posts on social media channels, and the content goes out daily.
We leverage email marketing via onboarding flows, trial nurture sequences, and newsletters.
We ran Meta ads from February 2025 until October 2025, which gave us an average ROAS of 1.6x - not great, but we didn't lose money.
And I tried to build in public many times. Somehow, I quit every time. It's not that easy to stay consistent.
Overall, most of our users come from organic traffic, and we've worked a lot on building this up. SEO takes a lot of time and effort, but eventually, it works on autopilot.
Consider bringing on an SEO expert if you aren't one. Angel Match was stuck at ~$5k MRR for a long time and we didn't know why. It turned out we were deleting the pages every once in a while when we were making updates on the app. So the website couldn't compound in search traffic. This changed when we hired an in-house SEO expert. We've been ranking and growing our web traffic ever since.
As far as growth, a few things that have been particularly advantageous:
Programmatic SEO: If you start any database product, this is a growth hack. Creating structured pages at scale brings us consistent high buyer intent organic traffic. We build this for every database product.
Building small and useful free tools around our main product: Nowadays it has become very easy to build free tools compared to pre-LLM times. We've built tons of free tools and launched each of them everywhere we could. This gave us free, consistent lead magnets, backlinks, referral traffic, organic traffic, and paid users on autopilot. Of course, not every free tool worked well, but it was worth building them and adding to the platform.
Listening to users (as cliché as it may sound): Many of our most useful features came from user complaints. We built the CRM, email outreach, database updates, and many other small features through user feedback. Also, the cancellation flow turned out to be very useful because we've collected tons of reasons and fed them to ChatGPT to get the most common reasons users cancel. Through this feedback, we are improving our onboarding, email follow-up flows with investors, and so on.
Choosing the right niche: Databases are boring. The products we run are not flashy, but they solve boring problems. A lot of founders chase hype, which is great, but many of them can be killed when OpenAI comes out with new features, which has been happening a lot lately.
Working with remote people: I live in the US, and hiring local developers here means I most likely can't start a business. Programmers cost 10x more than the ones you can hire overseas who are as talented and can do the same level of work. For starting a business, this approach was a big advantage for me personally.
The biggest challenge was not to give up for years when I was pushing the boulder uphill and it didn't move as I wanted it to move. As my first cofounder used to say: "Healthy and rational thinking people would have quit a long time ago, but we're still pushing this like crazy people."
He was right, there were lots of people who started with us and then quit. Now they are working in corporate jobs. Not saying that's bad — each person has a different risk appetite and career goals. But I don't regret suffering to do this.
Here's my advice:
Don't chase ideas that sound shiny and cool - I made that mistake back in 2016 building a cool investment app for Millennials. Nobody cares about cool stuff. Chase painful problems. If you find a problem that people are trying to solve with spreadsheets, manual workflows, or tedious tasks, you are already 80% there.
Ship the smallest version of it to deliver value. It doesn't have to be perfect, it just needs to solve that particular problem and you'll find out if someone will pay for it.
Keep talking to users. Don't delegate the customer service tickets to anyone - this will help you understand directly what you should focus on and what problems your users have.
Get your hands dirty. You learn much more from actually doing, but not from listening to podcasts or reading books. Podcasts and books can be helpful, but getting your hands dirty is the best teacher.
Don't quit early and don't go into it for a hobby. Stay consistent and build your app brick by brick. I know a lot of founders who quit too early when they should have kept shipping.
You don't need to be technical or raise VC money to build useful and profitable SaaS products.
While there is a phrase: "If you want to make God laugh, tell him about your plan," I want to run a bootstrapped portfolio of SaaS apps. Hopefully, we can build systems where we'll have tons of apps that run on pretty much auto-pilot.
Andrew Wilkinson's "Never Enough" book inspired me. He runs a publicly traded company called Tiny, which owns tons of profitable software businesses that generate tens of millions, if not hundreds.
To get there, we're working on a few things:
Reducing churn on Angel Match, which is at ~30%. These are the things we are adding:
Onboarding walkthrough and fundraising setup checklist
Email follow-up flows to investors
Demo video
Meta Ads
New onboarding emails and newsletters
Building Autoposts from the ground up (rebranded Pur Social that we acquired). Currently at ~$500 MRR.
Adding another 300k journalists to Journalist Hunt and scaling journalist pages through programmatic SEO. This already is generating ~100 organic clicks daily.
Adding CRM, email outreach, data room, and investor updates to Investor Hunt
If you’re raising capital, you can visit Angel Match. Or follow me on X or LinkedIn — where I build in public on and off.
Leave a Comment
This is such a powerful story. Love how you turned failure into momentum and kept going when most people would’ve quit. Huge respect for the persistence and grit.
absolutely a great