The number one reason products fail is that nobody wants them. Learn how to identify real demand, spot pull vs push signals, and build something people will actually pay for.
Y Combinator, arguably the most successful startup accelerator in history, has a three-word motto painted on their wall: Make something people want.
Not something clever. Not something technically impressive. Not something you personally find interesting. Something people want. It sounds obvious, but it is the single most violated rule in entrepreneurship — and vibe coders are especially at risk because building is fun, and validation is not.
You can code an incredible product in a weekend. Beautiful UI, clean architecture, clever algorithms. And it can still fail completely if nobody wants it. This post is about how to make sure you are building something the market is pulling out of your hands, not something you are pushing on an indifferent audience.
Push vs Pull — The Demand Litmus Test
There are two kinds of product dynamics, and they feel completely different.
Push feels like this: you are constantly explaining why your product is useful. You send cold emails and get no replies. You post on social media and get polite likes but no signups. People say “that’s interesting” but never come back. You offer free trials and people do not even bother to set up their account. You are pushing a boulder uphill.
Pull feels like this: you mention what you are building and people ask where they can sign up. Beta users email you asking when the next feature is coming. People share your product without you asking them to. You have a waitlist. Customers complain that your product is too slow or too limited — not that it is unnecessary. You have trouble keeping up with demand.
Pull does not mean your product is perfect. Pull means the problem is real and painful enough that people actively seek solutions. Your product can be ugly, buggy, and incomplete — and if it solves a painful problem, people will use it anyway.
If you are six months in and everything still feels like push, it is time to seriously reconsider the problem you are solving. As training.kalzumeus.com puts it: customers should already know they have the problem and be attempting to solve it. If they do not know they have the problem, you are either going to need an impossibly ambitious marketing campaign, or you are delusional.
The “Hair on Fire” Test
Imagine your customer’s head is on fire. They do not need to be convinced that they have a problem. They know. They are desperate for a solution. Any solution. A bucket of water, a fire blanket, jumping in a lake — they will take anything.
That is the level of pain you want to solve.
Now, most problems are not literally life-or-death. But the best solo businesses solve problems that feel urgent and recurring:
- “I waste six hours every week doing this manually.”
- “I lost a client because I missed a deadline that a tool could have caught.”
- “I keep making the same expensive mistake because I cannot see this data.”
- “My current solution breaks constantly and the company that makes it has terrible support.”
These are hair-on-fire problems. The person knows they have the problem. They are already spending time or money working around it. They will pay to make it go away.
Contrast this with problems that are “nice to have”:
- “It would be cool if I could see this visualisation.”
- “I guess this is slightly annoying sometimes.”
- “I do not have this problem often but when I do it is mildly inconvenient.”
Nice-to-have problems create products that people praise but never pay for. Hair-on-fire problems create products that people complain about and still use every day — because the alternative is worse.
Demand Signals You Can Spot Before Writing a Single Line of Code
You do not need to build first and hope for demand. Demand leaves footprints everywhere. You just need to know where to look.
Forum and community complaints. Go to Reddit, Hacker News, Twitter, indie hacker communities, and niche forums where your target audience hangs out. Search for phrases like “I wish there was,” “is there a way to,” “I hate how,” or “what do you use for.” These are people articulating unmet needs in their own words.
Competitor reviews. Find competing products on G2, Capterra, Product Hunt, or app stores. Read the negative reviews — specifically the 2-star and 3-star ones. These are people who wanted the product to work but it failed them in specific ways. Those specific failures are your opportunity.
Existing workarounds. When people build their own solutions — spreadsheets, Zapier workflows, manual checklists, duct-taped scripts — it means the problem is real enough that they are investing time to solve it. If you can replace their cobbled-together workaround with something smoother, they will likely pay for the convenience.
Search volume. Use Google Trends, Ahrefs, or even Google’s autocomplete to see how many people search for solutions to the problem you are considering. High search volume for problem-related keywords suggests real demand.
Willingness to pay (the most important signal). People who say they would use your product are not validating demand. People who pay you money in advance of the product existing are validating demand. Pre-orders, deposits, annual subscriptions before launch — these are the strongest signals.
Why Smart Solutions to Non-Problems Always Fail
This is the hardest lesson for technical founders. You see an inefficiency. You think “I could automate that with a clever algorithm.” You build it. You show it to people. They say “huh, neat” and never use it.
The inefficiency was real but not painful. The person experiencing it did not care enough to change their behaviour. Changing behaviour is one of the hardest things to achieve in business. People stick with what they know even when better options exist — unless the pain of staying is greater than the effort of switching.
That is why products solving “slightly better” problems struggle. “Slightly better” is not enough to overcome switching inertia. You need to be dramatically better, meaningfully cheaper, or solve something the current tools cannot solve at all.
As a vibe coder, your instinct is to build elegant solutions. That is a strength — but only after you have confirmed that the problem is worth solving elegantly. Elegance applied to a non-problem is the most expensive kind of waste: it costs you months of your life.
Your Action Item
Find Three Demand Signals This Week. Choose the problem your product solves (or the problem you are thinking about solving). Go to three places where your target audience discusses their work — Reddit, Twitter, a Slack community, a forum. Search for the problem using the phrases listed in Concept 3. Screenshot or save every instance you find. If you can find ten or more independent people describing the problem without prompting, the demand exists. If you struggle to find even five, seriously reconsider whether the market is real.
CTA Tip: The best products are aspirins, not vitamins. Vitamins are nice to take. Aspirins cure headaches people already have. Build aspirin.
—