Everyone I know is building an AI app right now.
Roman built a job board.
A plain one. It scrapes listings straight from company websites and puts them behind a paywall.
Last month it did $28,000.
His job board business is called Career Hound, he's completely non-technical, and he launched it two years ago with one goal: make $1,000 each for him and his co-founder. It's now at $25K+ a month with 11.7 million social views in the last 30 days.
Gus from Starter Story sat down with him and pulled the whole playbook apart. I watched it twice. Here's the full breakdown, with every number Roman shared.
The Numbers Behind Career Hound
Roman opened his dashboards on camera. Here's what a boring job board at scale actually looks like:
- $28,000 total revenue last month
- $12K MRR (the gap comes from a mix of weekly, monthly, and yearly plans)
- 8,000+ users, 779 active subscribers
- Pricing: $7/week, $20/month, or $50/year
- Average lifetime value: $50 per user
- Churn: 31%
- 11.7 million social views in the last 30 days
Two things jump out at me.
First, the hard paywall. No free plan, no freemium funnel, nothing. You want the jobs, you pay. Most founders I talk to in the club are terrified of doing this. Roman just... did it.
Second, that 31% churn. That number would make most SaaS founders throw up. But Roman is selling a job search product. People find a job and leave, which means the product did its job before they cancelled.
The lesson: churn benchmarks only matter inside your category. A $50 LTV with 11.7M free monthly views is a money printer, ugly churn and all.
Why a Boring Job Board Beat a Thousand AI Apps
You've seen a job board before. You've probably used five.
So why does this one make $25K a month while most die at zero?
Roman's answer surprised me: "What makes me stand out has nothing to do with the product. It all has to do with my ability to spike people's emotions through my content."
He studied psychology in university. Then he sold alarm systems door-to-door for five summers, eventually managing sales teams. No coding background at all. He calls being non-technical his secret sauce, because it forced him to get elite at the one thing most technical founders skip.
Distribution.
The product is simple on purpose. Easy to explain, easy to understand, people get it in one sentence. If you can't pitch your product in the first two seconds of a TikTok, no amount of features will save you.
I've made this exact argument before in my piece on boring business ideas that make money. Proven category + better marketing beats novel category + no marketing. Every time.
Timeless Content vs Timely Content
This is the core of Roman's marketing system, and it's the part I'd steal first.
He splits all content into two types:
→ Timely content: whatever's trending right now. Works this week, dead next week. Build on it and you're on a hamster wheel, chasing trends forever.
→ Timeless content: the questions your audience will always have. In his niche, people will always want to find companies hiring remotely, worldwide. That desire never expires.
So his winning format is stupid simple: a hook calling out the dream ("companies hiring remote, worldwide"), a list of companies offering it, and a closer sending people to Career Hound.
He's been running variations of that same format for two years. He repeats what converts until it stops working, and two years in, it hasn't stopped.
Meanwhile the timely stuff (he showed an example format everyone's copying right now) burns out fast because viewers learn to scroll past it. They already know it's an ad.
Find your niche's timeless questions. Answer them in a repeatable format. Run it until the wheels fall off.
Content Thrifting: The Research Habit Behind 11.7M Views
How do you find those timeless formats? Roman has a name for it: content thrifting.
He spends hours scrolling content in his niche across Instagram, Reddit, and TikTok. Video and text, never just one format. Every time something grabs his attention, he saves it. If it grabbed his attention, there's something in it.
His logic is brutal and correct: "To spot something that's not obvious to other people, you need to do more research than others. If you scroll enough, you'll inevitably become aware of things others aren't aware of in your niche."
Most founders guess what content will work. Roman has scrolled so much he's barely guessing anymore. The patterns repeat, and the patterns reveal the timeless topics.
That's the unglamorous truth behind 11.7 million views in 30 days. It comes from a volume of research most people won't sit through, not from virality hacks.
The $20 Creator System (With Gray Hair on Purpose)
Roman doesn't make most of this content himself. He runs a network of UGC creators, and he deliberately recruits older ones.
His reasoning: if you're looking for job search advice, why would you listen to a 19-year-old? Gray hair and wrinkles read as wisdom and credibility, on a subconscious level, to his exact audience. He matched the messenger to the market.
The economics are wild:
- He pays no more than $20 per video in ~90% of cases
- He finds creators mostly by DMing people already posting for fun: "Hey, do you want to do some paid content work?"
- Each creator gets a sheet of formats with instructions so simple that people in their 50s who've never made content can execute them
His quality bar for the instructions is one question: "Have I left any room for confusion, or is it stupid simple?"
Zero research required from the creator. Open the laptop, follow the sheet, film, done. That's why he can pay $20 instead of $200. He's not buying creativity, he's buying execution of a format he already proved.
I broke down a similar approach in the UGC creator system story if you want a second angle on running creator networks at scale.
How to Find Your Own Boring Winner
Roman's advice for anyone starting from zero is refreshingly concrete.
Step 1: Source ideas from money trails, not brainstorms. Sign up on SideShift (a creator marketplace) and look at which companies are hiring creators. Companies paying creators have money, which means they're making money. Or use Sensor Tower to find apps already getting downloads.
Step 2: Build a similar product in that proven space. Pick one you actually enjoy. Then use the incumbent's content as inspiration for your own marketing.
Step 3: Let the proven space de-risk your learning. This is the underrated part. If you struggle to get paying users in a market where someone else already wins, you know it's not the product. It's your execution. That clarity is worth a lot.
Step 4: Run marketing experiments daily and chase failed results. Every failed experiment eliminates what doesn't work and gets you closer to what does.
Roman ties it back to his door-to-door days: "We used to tell ourselves, I get closer and closer to my sale with every door I knock. You get closer and closer to your first dollar with every marketing experiment you do."
Five summers of slammed doors will teach you things a growth course never will.
The Stack: Lean Enough to Stay Profitable
For a site doing $28K a month, the tech stack is almost comically lean:
- Short URL for tracking social views
- PostHog for marketing experiments
- WhatsApp for creator communication
- Microsoft Clarity for watching site visitors
- Hetzner for the server
- Postmark for email campaigns
- Google Workspace and Cloudflare for the basics
No expensive analytics suite, no custom creator portal. He manages a whole creator network through WhatsApp and a spreadsheet of formats.
When your LTV is $50, keeping tools this cheap is what makes the math work.
What I'm Taking From This One
Roman's parting advice: "Once you make your first money online, it becomes really obvious that you can do it. If you're running marketing experiments daily and extracting insights from every failed one, it's actually inevitable that you'll create a profitable business. It's not rocket science. It just takes consistent work over a couple of years."
Inevitable is a strong word. Coming from a guy who spent five summers getting doors slammed in his face, I'll allow it.
The product didn't win this. The research did, executed through a repeatable format and a bench of $20 creators following instruction sheets.
Two years. From "let's make $1,000 each" to a real business.
FAQ
How much money can a job board business make?
Career Hound made $28,000 last month with $12K in MRR, 8,000+ users, and 779 active subscribers. Pricing runs $7/week, $20/month, or $50/year behind a hard paywall with no free plan. Job boards monetize either side: charging job seekers (Roman's model) or charging companies to post listings.
Do you need to code to start a job board?
No. Roman is completely non-technical. His co-founder built the original scraper that pulls jobs directly from company websites, and Roman runs the marketing side. He argues being non-technical was an advantage because it forced him to master distribution instead of hiding in the product.
How do you market a job board without a big budget?
Roman's system: identify timeless content formats through hours of niche research (he calls it content thrifting), then pay UGC creators around $20 per video to execute those formats from a simple instruction sheet. That system generated 11.7 million social views in 30 days.
Is 31% churn bad for a subscription business?
For most SaaS, yes. For a job search product, it's expected: users cancel because they found a job, which means the product delivered. With a $50 average LTV and near-free acquisition through organic content, the unit economics still work. Judge churn against your category, not a generic benchmark.
Steal Playbooks Like This Every Week
Roman's story is exactly the kind I chase on the Profitable Founder Podcast: real founders opening real dashboards and explaining what actually worked.
If you're building toward $100K MRR and want the tactics founders actually used (not the sanitized LinkedIn version), come listen.
