Validating Ideas

Let's take a look at a couple of real-world examples from the Rails World.

Dmitriy Zaporozhets built it in 2011 because his company needed a self-hosted Git solution. He validated the need with exactly one customer - his own team. Which is a common story: Build the tool that you need and use. You can definitely do that, especially if you see business or fun value. Chances are, others will get that value as well and you will have a product. 37Signals built Basecamp and a bunch of other products and even Rails primarily for themselves.

A Theo guy here, goes as far to say that idea validation and marketing before you have the product is bogus (X link). Obviously, he is trying to be polarizing, but he has a point. That is, it's a working strategy. At the same time, his X thread completely ignores the reality of the other side, namely that an uncountable amount of builders have built something that they came to while doing various ways of marketing and sharing about their ideas.

And here's the kicker with the GitLab story: Dmitry immediately open-sourced GitLab and listened to feedback from other developers facing the same problem. Today it's worth billions.

Meanwhile, Pieter Levels built Nomad List by first creating a simple spreadsheet of cities for digital nomads. No Rails, no database, just a Google Sheet. When hundreds of people started contributing and asking for features, only then did he code it up. First month revenue: $1,000. Then, PhotoAI? He tweeted "would you pay $10 for AI headshots?" before writing a single line of code. 100 people said yes, so he built it in a weekend and made $100,000 in the first month.
Start with conversations, not code. When Amy Hoy was building Freckle (now Noko Time Tracking), she didn't build features she thought freelancers needed. She hung out in freelancer forums for months, documenting every complaint about time tracking and invoicing. She called this "Safari research", observing developers and designers in their natural habitat. I would call it: "Externally Provided Feature Board". The result? A time tracker that actually solved the problems people complained about daily, not the imaginary problems she assumed they had.

Want a counter-example? Contrast all that with Everpix, the photo storage startup that spent years building perfect photo algorithms nobody asked for. They had 55,000 users but died because they never validated if people would actually pay $49/year for photo storage.

In short, your validation vs building ratio should match your your product-risk vs market-risk.

When to validate/market 0% and build 100%?

Do you need the tool for yourself badly? You can probably opt into just building and then presenting it to the world, hoping for the best! Hopefully your project doesn't take too long. You take on the risk of the market not needing it, because you need it anyway, that being for business purposes, personal workflow improvements and entertainment, or a portfolio project. Also, is the product risk very high, so that you aren't sure if you can build it at all? These are bonus points.

When to validate/market 80% and build 20%?

For most builders and ideas, the market risk is the highest. You don't have your Ideal Customer Profile down yet and your offer isn't clear-cut. You struggle with your market fit while you know that your is mostly about submitting some forms and doing a few background tasks. You should go out into the world and talk to potential customers and do all the other cool marketing stunts.

If you're like the Tuple founders (remote pair programming for developers), you might need more building upfront to prove the technical feasibility. They still started by gauging interest with a simple landing page that got 500 signups before they wrote production code.

When to validate 50% and build 50%? 

Further down into your project journey, you will naturally increase the ratio of your building time to account for building more features and fixing bugs. And you will still need to keep talking to customers. A lot. What do they like? What do they hate? Where did they find you? Why do they stick? As BLACKGIRL BOOKAPP calls out her successful formula: Keep building. Keep talking to customers. Keep costs low. Repeat.

---

This is just a tiny overview and our own take for you to understand what it will take for you in terms of idea validation specific to your situation. You will need to dig deeper into the marketing side of things. It requires to build knowledge and get creative. The best resources for that are:

- The Mom Test - Rob Fitzgerald
- 140 SaaS Marketing Opportunities - Corey Haynes
​
​---
​
​Next up, now that you have talked to enough people to know that your idea is legit or you are already building anyway, you will want a landing page to build a list and start tailoring your offer.