We forgot to ask "Should I?"

Stefan Latuske — Thu, 11 Jun 2026 10:01:49 +0000

As a solo developer, AI is a tool that has greatly amplified my ability to bring ideas to life. Before that, every idea I had was just put into a file called "Project Ideas," never to be touched again and eventually forgotten.

With AI, we now have a way to bring these ideas to life very quickly, but a lot of the time we forget to ask: "Should we?"

The data clearly shows that people are creating more apps and tools than ever before. But if you look a little closer, you'll also see another trend on the horizon: "ZERO Users!"

Most people have forgotten to do the most important step before building something: validating whether there is any demand for what they are building.

How to validate an Idea

Trust me, I've been there. You have a great idea, you're super excited to build it, so you just get started. You sink a few days into the idea and later do a Google search, only to find out there are already lots of similar products. You get super demotivated and just drop the idea.

Turns out this is not how it has to be. Instead, you can follow some simple steps to figure out whether your idea has actual demand and should be pursued further.

The easiest first step is always a quick Google search to see if something similar already exists. It doesn't matter whether you find something or not. This is simply to figure out your next step.

Oh no, I found a Competitor!

Calm down! This is actually great news because it means someone else has already done the work for you and found a way to turn the idea into a business.

A lot of people would stop here and look for a new idea, but there are several reasons why you should still consider moving forward:

The idea is already proven to work as a business.
You can learn from your competitors.
The market is big enough to support more than one solution. (Come on, there are 8 billion people on this planet. There's enough pie for everyone.)

If you still want something original, feel free to validate another idea. But if you want to move forward with this one, your next steps should be:

Competitor analysis (AI can help).
Figure out your product's "moat."
- MOAT: A unique advantage your product offers that makes it difficult for competitors to copy. This can be knowledge, location, connections, etc. (Coded features do not count!)
Run some low-cost experiments, for example with Facebook Ads, to get commitments from people to use your product. -Commitments ≠ signups. You need something that creates a barrier, such as a pre-sale or a deposit. If people are willing to enter their credit card details without receiving anything yet, you know there is demand.
Deliver on what you promised (the actual hard part).

Great nobody thought of this yet!

Hold your horses! Just because it doesn't exist right now doesn't mean nobody has thought of the idea before. It could also mean the idea has already failed or that there is simply no demand for it.

One thing is certain, though: the idea is probably original enough for you to be a pioneer in this field.

This is very exciting, but also much riskier than pursuing a proven idea, so make sure to do your research and validation thoroughly before committing to build it.

But where do you actually start? There is no competitor to analyze.

Well, you'll need to do a little trial and error. The first step is to clearly define the problem your business solves. For example:

A mobile car detailing business solves the problem of busy car owners not having the time to drive to a car wash and wait for their vehicle to be cleaned. Instead, the service comes to the customer's home or workplace. Once you've identified the problem, you can test demand by talking to potential customers and offering the service on a small scale before investing heavily.

Once you've figured out who you need to target, you need to start gathering feedback on the idea. This should be as low-cost and as fast as possible.
Facebook Ads, Reddit threads, or just some good old face-to-face conversations are the prefered method for this. Don't be afraid to experiment with your validation tests. Anything low-cost is worth trying.

But be careful here: don't ask your friends or family because they are more likely to tell you your ideas are great. You want people who don't know you so they can give you the honest and harsh feedback you need.
This step can be deflating because, more often than not, people don't care about your little idea. Don't be discouraged by that. It might simply mean you're looking in the wrong direction and asking the wrong people.

Now that you've gathered some data, you need to analyze it to figure out what works and what doesn't. Taking the mobile car detailing example from earlier, you might have assumed that people would value the service because they're busy but still want a clean car.

Your data might confirm this, or it might show that most people don't care much about a clean car, while dog owners want their cars detailed more frequently because their dogs shed too much hair inside. This would suggest that you should pivot and focus your advertising on those people instead.
Whatever your data shows should also be validated with additional experiments.

Once it's clear that the demand for your idea is big enough, start building!

Closing Thoughts

Before you spend days or weeks building, spend a few hours validating. Talk to people. Run small experiments. Collect data. Challenge your own assumptions. Every hour spent validating can save you dozens of hours (and $$$) building the wrong thing.

And if your idea turns out to be bad? Great. You just saved yourself a lot of time and learned something valuable. Validation isn't about proving that your idea is good, it's about finding the truth as quickly and cheaply as possible!

DEV Community: Stefan Latuske

We forgot to ask "Should I?"

How to validate an Idea

Oh no, I found a Competitor!

Great nobody thought of this yet!

Closing Thoughts