I Killed My Startup Idea in One Week (Here's How You Can Too)

There are many articles on how to come up with a startup idea. In a nutshell: find a problem that deserves to be solved better, while people care about it and are ready to pay.

Some suggest starting with a problem you need solved for yourself. When I had my 2 previous attempts, I based them on this approach. I wasn't happy with the type of product I had at the time, and I wasn't happy with education in one specific field I was interested in. So I covered both directions and was profitable, yet not scalable.

Recently I fell in love with another idea, sadly a B2C type. The idea is based on self-awareness and self-reflection. Yes, the topic is tricky, but when you're in love - it's not easy to see all the red flags. That's why it's crucial to do a small test before moving forward.

As a Product Manager, I'm proficient in defining next steps: what idea goes in, what goes out. So after some basic market research, I created a short plan to test the idea: a simple landing page that imitated the core exercise, tested for free with friends and interested people, improved according to feedback, then ran targeted ads on Meta with a call to action to leave an email. The goal was to see how hooked the audience was before building anything real.

Of course, I was also captured by thinking about how many useful exercises I could place in the mobile app, how amazingly ChatGPT could help, how wonderful influencer marketing would work, and how perfectly a referral program would fit.

Good thing I didn't spend much time designing or researching, as all this time would have been wasted.

"create a pic of a founder chasing his dream startup idea"

Common test issues to think about upfront:

1. Usability of the test can be an issue in itself. The first test should be about setting up the right design so users don't get confused at that stage. My users were confused, as the idea is not easy to explain even sitting next to a person.

2. Distant testing without watching the user won't always work; initially, you have tons of questions no fill up form will cover. People want to suggest lots of things, but keep it to the point. Before scaling to a bigger audience, make sure your test is perfectly built, has a clear goal, and is narrowly focused on one simple idea.

Red flags I discovered during the first test:

If the idea is so new that you need to explain it in the design of the test, that's a red flag. Users will get lost before they receive any value. If the idea isn't answering a specific problem people are already trying to solve, red flag. They're not searching for a solution, so getting new users organically is quite hard. If users need to work hard to get to the value, red flag. You want to spoon feed value as fast as possible. Any additional friction leads to drop off, unless the problem is so extreme users are willing to fight for the solution. If your product doesn't deliver measurable value, red flag. Users need to notice and appreciate results: "File uploaded," "message sent," "10k steps reached." Otherwise, they won't feel satisfied and won't retain. If your target audience is quite specific and hard to detect and contact, red flag. I was measuring my target audience by myself and my experience. Don't do that unless you're creating something for yourself only.

After I discovered these findings, I could improve the test and publish it online, or check if the big picture still looked attractive.

To be honest, the red flags were already enough for a no go. Yet I decided to do the big picture analysis. I created a business model and calculated unit economics: potential users I could acquire, potential retention, potential monthly subscription profit, and marketing costs.

Sadly, the results were upsetting. I could never be profitable with paid marketing only. B2C is really hard in these terms. What was a new problem is that referrals based on vague value won't work. It doesn't solve a problem in a measurable way, and it doesn't spoon feed the value. In a nutshell, my idea was to have a tool to work with to get some insights. As with anything like that, it's fun at first and overwhelming the next second.

I'm very happy to have experienced this tiny test project myself. Even though many of these lessons have been written about many times, you only get it under your skin when experienced.

Hope this article helps someone create a good testing plan and not waste time before the idea has proof.