Stop building tools for "freelancers" — it is a death sentence for your SaaS
Most indie hackers are burning their limited runway on a lie.
We tell ourselves we are building for a "broad market" to maximize our chances.
"It is for creators."
"It is for small business owners."
"It is for developers."
In reality, targeting everyone is the fastest way to talk to nobody.
I have seen founders spend six months building a feature-rich dashboard only to realize they are competing with free spreadsheets and $10B companies simultaneously.
They have no leverage because they have no focus.
A real ICP isn't a category. It is a surgical strike.
There is a massive difference between "tax software for freelancers" and "automated tax-compliance for EU-based designers earning over €80k."
The first one is a commodity. The second one is a must-have that people pay premium prices for without blinking.
If you cannot describe the exact moment your customer's hair is on fire, you aren't ready to write a single line of code.
You need to work backward from the pain, not forward from your features.
We built IdeaToLaunch to stop this cycle of building things nobody wants. It is a validation engine that helps you pressure-test your ideas and define your ICP before you waste a year of your life on a ghost town.
I broke down the specific framework we use to move from a vague idea to a data-driven ICP over at IdeaToLaunch.
Check it out: https://ideatolaunch.co
Really like this approach. Turning the common mistake of “building something nobody actually wants” into a product that makes validation the starting point instead of an afterthought is exactly what many indie hackers need. The strong focus on defining a clear ICP and relying on real demand signals rather than assumptions or gut feelings is a huge advantage. This kind of mindset can save founders a lot of time, effort, and wasted resources. Great insight! 🚀
Love this—turning “I built something nobody wanted” into a product that makes validation the first step, not an afterthought, is exactly what most indie hackers need. The focus on real demand signals instead of vibes is such an underrated advantage.
Appreciate that, Aaron. That's exactly the thesis, validation shouldn't be something you do after you've already sunk months into building. Most post-mortems start with "we didn't talk to users early enough." We wanted to make that impossible to skip. The demand signals piece is key, gut feeling is fine for picking a restaurant, not for picking a market.
this hit me hard honestly... i'm literally in this trap right now. been saying frikt is 'for people who have frustrations' which is basically everyone and nobody at the same time lol
the hair on fire thing is the best way i've heard it explained. if you can't picture that exact moment you don't really know who you're building for
going to spend some time this week trying to get way more specific about who my actual user is. thanks for the reality check
Respect the honesty, Karolis. "For people who have frustrations" yeah, that's literally 8 billion people 😄 But the fact that you recognize it puts you ahead of 90% of founders. One exercise that helped us: pick your best existing user and write down exactly what they were doing 5 minutes before they signed up. That moment IS your ICP. Once you nail that, your copy, features, and pricing all snap into focus. Good luck narrowing it down this week. I would love to hear what you land on.
The "EU designers earning €80K+" example proves the point but also creates a different problem - how do you even find those 500 people to validate? Sometimes starting broad and narrowing based on who actually converts works better than guessing a micro-niche.
You don't need to find 500. You need to find 5.
The "how do I even reach them" objection is actually the validation itself. If you can't figure out where 5 EU designers earning €80K+ hang out online, you don't understand the problem well enough to build software for them.
That's the test most people skip.
"Start broad and narrow based on conversions" sounds logical but assumes you have enough traffic to generate statistically meaningful data. Most indie hackers don't. You end up with 40 signups from 15 different use cases and zero clarity on what to build next.
The micro-niche forces you to do the hard work upfront actually understanding the customer before writing code. It's slower to start but dramatically faster to PMF.
This tension is real, but it's actually a feature, not a bug.
The question isn't "broad then narrow" vs. "precision targeting." It's understanding which validation question you're answering at each stage.
**Stage 1: Pain validation** — You don't need to find 500 EU designers earning €80K+. You need to find 5-10 people who experience the exact pain point. At this stage, you're validating that the hair-on-fire moment exists, not that the entire ICP definition is perfect. Those 5-10 conversations tell you if the problem is acute enough to pay for.
**Stage 2: ICP refinement** — This is where "who actually converts" data becomes valuable. You might discover that the €80K threshold was wrong, or that UK designers convert better than EU ones, or that the pain is actually strongest for agencies with 3-10 designers, not individual freelancers. But you only get clean conversion data if you started with a hypothesis sharp enough to attract people with the real pain.
The "start broad and narrow based on conversions" strategy works when you have distribution leverage (existing audience, paid acquisition budget, viral mechanics). For most indie hackers, broad means you're optimizing conversion rates on traffic you can't afford to acquire. You end up with statistically insignificant data because the 50 signups you got came from 12 different problem spaces.
The micro-niche approach front-loads the hardest question: where do these people congregate? But that's the question that reveals whether you actually understand the pain. If you can't name the Slack communities, subreddits, LinkedIn groups, or industry events where EU designers earning €80K+ talk about tax compliance, you don't understand the problem deeply enough to build a solution they'll pay premium prices for.
The real trade-off isn't discovery cost. It's confidence vs. speed. Broad-then-narrow is faster to launch but slower to product-market fit. Precision-then-expand takes longer to find your first 10 customers but converts them at 10x the rate because your positioning is surgical from day one.
This is one of the best breakdowns of the validation spectrum I've seen.
The part about distribution leverage is the key most people miss. "Broad then narrow" is a luxury strategy, it works beautifully if you're sitting on an email list or can spend $5K/mo on ads to generate signal. For a solo founder with $0 marketing budget, it's just expensive guessing.
Your point about naming the Slack communities and subreddits is exactly the litmus test we bake into IdeaToLaunch's validation flow. If you can't answer "where do these people already talk about this problem," you're not ready to build yet.
The confidence vs. speed framing is spot on. I'd add one thing , precision-first also gives you positioning language for free. When your first 10 users all describe the problem the same way, your landing page practically writes itself