A Good Idea Is Worth About 5%
If you stack up everything that makes a digital product succeed — execution, distribution, timing, persistence — the idea itself accounts for maybe 5% of the outcome. The rest is work. Systematic, consistent work where you improve one thing at a time and keep showing up.
MrBeast once said that beginner YouTubers constantly ask him how to build a popular channel. His answer: make 100 videos, improve one thing in each, then come back to me. Nobody comes back. Most because they never hit 10%. Some because they got there and no longer need the advice.
But working hard alone isn’t enough. You can grind for years and get nowhere if you’re not improving. And improving in isolation isn’t enough either. A talented creator locks themselves in a basement, perfects a product nobody asked for, launches it to silence, and doubles down on the wrong thing.
Validation means saying “let me check” as often as possible — not after months of building your dream product. Because nobody else is dreaming about your product. Only you are.
The Validation Loop
The entire process comes down to four steps:
- Build a small thing
- Show it to the world
- Listen to what the world says
- Improve it
If you internalize this, you’re ahead of 90% of digital product creators. But there’s a critical nuance: showing and talking isn’t the same as listening. You’ve met the person who talks passionately about their project but never lets you get a word in. That kind of broadcasting is useless for validation. You need to listen, extract signal, and feed it back into the product.
Start With the Problem, Not the Solution
Your product idea should stem directly from a problem — specifically, from the desire to solve it. Not every problem is worth building for. A good candidate has these traits:
It’s urgent. It hurts right now and can’t be postponed. If your potential customer says they “don’t have time” to try your solution, the truth is the problem isn’t painful enough. It’s easier to sell painkillers than nutrition courses.
People already pay to solve it. This is a strong signal. If competitors exist, that’s good — it means there’s demand. Trying to convince a market to pay for something nobody has ever paid for is brutally hard and usually requires venture-scale funding.
It recurs. Someone who buys a website template will likely buy another. Someone who takes a beginner language course will come back for the advanced one. Recurring problems create repeat customers.
Validate the problem before the solution
The first hypothesis you test is the problem itself, not whether your specific solution works. Skipping this step is asking for trouble.
Before you build anything, you need to understand:
- How do people currently solve this problem?
- How much do they pay for it?
- What do they consider important in a solution?
These aren’t obvious, and your assumptions are probably wrong.
Meet Marta. Learn From Her Mistakes.
Marta: “I have an idea!”
Marta is a freelance productivity consultant. She wants to create her first digital product: a course on task management for everyone. She outlines 8 modules, records 5 lessons. She’s proud. But deep down she’s anxious: “What if nobody needs this?”
Marta is doing it wrong. She started building before testing.
Marta: “I’ll run a poll!”
Marta posts an Instagram story: “Would you buy a productivity course?” 60 people tap YES. 20 write “great idea!”
Marta treats this as validation. Sets the price at 99 EUR. Launch day: 0 purchases. Silence.
“Yes” was polite. It wasn’t a commitment.
Lesson: Declarations are not behavior. A poll is not money.
Talk to the Right People
Your friends and family will tell you everything you want to hear, except the truth. They’re not your customers.
People in Facebook comments who’ll happily take something for free aren’t your customers either — at least not based on a giveaway post.
You’re building a product for a market, not trying to find a market for a product you already built. The most valuable feedback comes from people who pay. If you offer a free tier, be cautious about feedback from free users — implementing their suggestions may not translate to revenue.
Marta: “I’ll collect conversations, not likes”
Marta pivots. She books 12 short calls (15 min) with people who tapped “yes.” Instead of asking “would you buy this?”, she asks: “Tell me about the last time your task management fell apart. What did you do?”
A pattern emerges:
- “I lose 2 hours a week just organizing my to-do list.”
- “I tried templates from the internet but they never fit my workflow.”
- “I gave up and went back to sticky notes.”
Lesson: Marta is extracting real pain and current workarounds — data that shapes a product people actually need.
How to ask
In problem-stage interviews, ask about facts, not opinions. People fabricate opinions but facts don’t lie.
- What’s your relationship with [topic/problem]?
- Do you use multiple products to deal with it?
- What’s the end result you’re looking for?
- How important is solving this to you?
- How much do you currently spend on it?
Ask about a specific past situation. What happened, what they tried, where they looked. Ask about emotions too — how they felt, what made them feel safe about a solution. But don’t manipulate with fear or artificial urgency. Purchases are driven by positive emotions, not panic.
Read “The Mom Test” by Rob Fitzpatrick. It covers everything you need to know about running these interviews.
Standardize your interviews. Ask similar questions every time. Use an AI note-taking tool, then summarize the patterns.
How many interviews? Until you stop hearing new things. You’ll know.
The Smaller the Niche, the Better
Beginning creators constantly want to build “for everyone.” This is a mistake.
A narrow niche is easier to validate because:
- You can clearly define the value they’ll get
- You know exactly where to find them (ads, conferences, communities)
- There are far more people searching for obscure, niche things than you think
Even if your product could serve everyone, you can’t market it that way. “Task management for everyone” competes with Google. “A project dashboard for freelance designers juggling 3-5 clients” — that’s specific, findable, and testable.
Start with one niche. Nail it. Then expand to the next.
Marta: “Niche instead of ‘for everyone’”
Marta sees that “productivity for everyone” is too broad. In her interviews, the sharpest pain belongs to solo creators drowning in tasks and missing deadlines.
She narrows: not “all professionals,” but “freelancers and creators running 3-5 projects at once who keep dropping balls.”
Lesson: Address specific people in a specific situation.
Validate the Market, Not Just the Idea
If no competing products exist for the problem you want to solve, that’s usually a bad sign — not a good one. Competition means demand.
Study your competitors for:
- Language — how do they talk to customers?
- Business model — is there a pattern everyone follows?
- Channels — where do they advertise, what conferences do they attend, what lead magnets do they use?
Beyond that, don’t obsess over competitors. They have the same problems you do.
Marta: “Do they already pay?”
Marta checks: do her target users pay for tools (Todoist, Asana, ClickUp, productivity coaching)? Are there competing products (templates, mini-courses)?
Yes: budgets exist, competitors exist, the need is urgent (missed deadlines = lost money and reputation).
Lesson: Demand already exists — no need to educate the market about the problem.
Pick One Channel
In the validation phase, don’t spread across every platform. Find one channel where your target customers actually are and test there.
Social media is natural for early validation — it’s free and gives fast signal. Paid ads can work too if you want to test purchase intent (and our 500 PLN ad test covers exactly how to do that).
Marta: “Where will I actually reach them?”
Marta has a small newsletter (300 people) and a LinkedIn following (1,500). Instagram is mostly “creative types” — less likely to pay for productivity tools. She picks LinkedIn as her test channel because that’s where people invest in work efficiency.
Lesson: Validate where your customers are, not where you have the most followers.
Build the Smallest Possible Test
Whether it’s software or an educational product, you need to ship fast and iterate. The people you interviewed are your first testers. Give them early access — not as a mass giveaway, but as a value exchange for their time and feedback.
Marta: “MVP, not masterpiece”
Instead of a 99 EUR course, Marta builds an MVP: an ebook + template called “The Freelancer’s Project Dashboard.” One question to answer: will they pay for the result?
The offer is minimal:
- 25-page ebook
- 1 dashboard + 3 views (projects / weekly planner / backlog)
- 10-minute video: “How to set this up in 30 minutes”
Speed over perfection — test at small scale.
Set Your Metrics Before You Launch
Define pass/fail thresholds before you spend anything. Not after.
Marta: “Metrics, not gut feeling”
Marta sets validation thresholds before launch:
- 200 page visits from LinkedIn in 7 days
- 30 waitlist sign-ups
- 10 clicks on “Buy”
- Minimum 5 purchases at 29 EUR (hard signal)
She publishes 2 posts and 1 short story with a client case. Result: 8 purchases (because she’d built the audience and done the interviews first). Validation passes.
Lesson: Clear “yes/no” criteria beat intuition every time.
After the Sale: Keep Validating
Validation doesn’t stop at the first purchase. After someone buys, don’t ask “did you like it?” Ask:
- What was hardest before you bought?
- What did you implement in the first 24 hours?
- What was unnecessary?
- What’s missing to get the result you wanted?
- What specifically did you pay for?
Use the answers to ship a better v1.1. Send the update to existing buyers for free. Collect testimonials in exchange for discounts. This is still validation — just post-launch.
Marta: “Feedback loop”
In v1.1, Marta adds one thing: a 7-step implementation checklist with example setups for 3 work types (copywriter, designer, consultant). She sends it to buyers as a free update.
Lesson: Problem-solution fit in practice — the product grows from data, not assumptions.
The Timeline You Should Expect
Assume that what you want to build today will take roughly 3 years to reach its real form — and it will look completely different from what you’re planning now. That’s fine. It doesn’t mean you won’t earn anything in those 3 years. It means that by running continuous validation at every stage, you’ll eventually reach a place most people never get to — where growth becomes exponential.
If your final product is identical to what you imagined at the start, you’re probably doing something wrong.
The Takeaway
Don’t build in a vacuum. Validate the problem before the solution. Talk to real potential customers, not friends. Pick a niche so small it feels uncomfortable. Set your success criteria before you launch. Ship the smallest thing that tests your core assumption. Listen, improve, repeat. The 500 PLN you spend on a cheap ad test will tell you more than months of building in silence.