I have a rule I follow without exception before starting any new project. I will not open my code...
I have a rule I follow without exception before starting any new project.
I will not open my code editor, design a single screen, or write a single line of code until I have answered these five questions with real data — not opinions, not assumptions, not what my friends think.
It took me two failed projects and about 14 months of wasted evenings to create this rule.
This is everything I learned.
Why Most Developers Build the Wrong Thing
We are builders by nature. The moment an idea lands in our heads, we want to open VS Code and start architecting the solution. It feels productive. It feels like progress.
But here is what is actually happening: you are making a bet. You are betting that the problem is real, that people will pay to solve it, that your solution is better than what already exists, and that you can reach your target customer efficiently.
Most developers make all four of those bets simultaneously without checking any of them.
The result? CB Insights analyzed 101 startup failures and found the #1 reason was "no market need" — at 42%. Not technical failure. Not running out of money. Nobody wanted what was built.
I was in that 42% twice.
The 5 Questions That Save You Months
Question 1: Is this problem real and painful enough?
There is a massive difference between a "nice to have" and a "need to have."
A problem is real and painful when:
People are already paying money for an inferior solution to it
People actively complain about existing solutions in public forums
When you describe the problem to strangers, they immediately say "yes, that drives me crazy"
A problem is not real enough when:
Friends say "that sounds useful"
You personally experience it but cannot find others who do
People say they would use it "if it were free"
How to check: Go to Reddit. Search your problem description. Read the threads. How many people are complaining about this? How recent are the posts? If you find hundreds of posts from the last 12 months, the problem is real. If you find three posts from 2019, it probably is not.
Question 2: Who exactly is your buyer — and can you reach them?
"Everyone who has a startup idea" is not a buyer persona. That is a demographic.
A real buyer persona looks like this:
Marcus, 29. Works full-time in product management at a mid-size SaaS company. Running a side project on weekends. Has tried and abandoned two previous startup ideas. Reads Hacker News daily. Active in 3 Slack communities for founders. His biggest fear is quitting his job for a business that fails. He has a budget of $50–200/month for tools that save him time or reduce risk.
When you have a persona this specific, you know exactly where to find Marcus (Hacker News, specific Slack groups), what message will land with him ("reduce risk of failure"), and what price point he will not hesitate on.
How to build this: Find 5–10 people who match your rough idea of the target customer. Read their public posts, their comments, their reviews of competing products. You will see patterns emerge within an hour.
Question 3: Who are your real competitors — and where are they weak?
New founders typically check 2–3 competitors. The real picture requires checking 10–15.
Your real competitors are not just the direct ones. They include:
The manual process people currently use (spreadsheets, Google Docs, consultant calls)
Adjacent tools people hack together to solve the same problem
Free alternatives people tolerate even though they hate them
Where to find competitor weaknesses:
G2 and Capterra reviews filtered to 1-star and 2-star
Reddit threads complaining about specific tools
Twitter/X searches for "[competitor name] problems" or "[competitor name] alternative"
App Store reviews if they have a mobile app
The weaknesses you find in competitor reviews become your product's strongest selling points.
Question 4: Is there a Go-To-Market path that does not require a large budget?
A great product with no distribution path is a hobby, not a business.
Before building, map out specifically:
Where does my target customer discover new tools?
What content or communities do they trust?
What would make them share this with colleagues?
Can I reach the first 100 customers without paid advertising?
If the answer to the last question is no, your customer acquisition cost will kill the business before it grows.
Question 5: What would make an investor say no immediately?
This is the most underrated validation question.
Imagine presenting your idea to a sharp, skeptical VC. What is the first objection they raise? Common ones:
"The market is too small"
"This is a feature, not a product"
"There is already a well-funded competitor doing this"
"Customer acquisition cost will be too high"
"The problem is not painful enough for people to switch"
If you cannot answer these objections with real data, you have not finished validating.
How I Run This Framework in 60 Seconds Now
For my first two projects, this entire process took me 2–3 days per idea. I was manually searching Reddit, reading competitor reviews, building spreadsheets, writing out personas in Google Docs.
After my second failure, I built a tool called Validexio (validexio.com) that automates this entire framework.
You describe your idea in plain text. It runs the complete validation framework using live data — competitor analysis, buyer persona generation, market demand assessment, Go-To-Market strategy, investor objection simulation, and an anti-roadmap (features that sound good but will slow you down).
The full report comes back in under 60 seconds.
There is a free tier. I built it because I wanted every developer with a side project idea to have access to the kind of market research that previously took days of manual work or thousands of dollars in consultant fees.
The Real Point
The validation framework above is not about killing ideas. It is about knowing which ideas deserve your time before you spend months building them.
Your time is your most valuable resource. Use it on ideas that the data supports, not just the ones that feel exciting.
Validate first. Build second. That order matters more than almost anything else in the early stages of a product.
Built something and validated it? Or skipped validation and regretted it? Share your experience in the comments — would love to hear real founder stories.