Luck and it's in a domain I'm indirectly in. A friend of a friend said:"It would be nice if X existed." At first, I didn't want to do it. After a few days I thought that it would be interesting to do some PHP coding (this was in 2006) and tried a few things. I posted the prototype in a forum related to the niche, the first 100 users came, they gave suggestions and I kept improving it and it kept growing.

Only after 5 years did I turn it into a business. Until then, it was something like a project for me as a student.

So, there really was no formal process. I did a few things right, I learned a lot along the way, but there's nothing to copy or to model. It was mostly luck and an opportunity I took.

Edit: I once read: "The perfect time to start a company that is successful in 2013 was 2007" or something in that sense. This is what happened. I took this opportunity. I would do things differently today, but to be honest, I have the same hard time as everybody else to come up with a new business and a viable idea. Because everything changed since 2006. So what I did right back then wouldn't work now, process-wise.

Still, I did a few things right I guess that are still valid in 2013. Mostly: I concentrated on important things, I do customer support myself, I listen to people, I'm friendly, I help people solve their problems. I care about them, about their privacy and do everything to protect it. I write emails that take me 30 minutes if it's necessary. Why? Because I only get a handful of emails per day. It's not scalable but then, it doesn't have to be. I do some SEO and rank well for my niche keywords. About 50-70% is word of mouth.

What I don't do: I don't have a blog, I didn't integrate any social network crap, I don't use Google Analytics or any other fancy third-party application that must be included via JS. I use Piwik that I check occasionally (sometimes every week, sometimes only every few months). I don't have big metrics. My most imporant metrics are: Money earned per week/month. Amount of active (paid) users. Signups per week.

Edit2: My product has nothing to do with PHP and doesn't target IT people. Just in case this is misinterpreted. I mention PHP only, because people on HN hate PHP that much and sometimes forget that it doesn't matter if it's PHP or anything else, as long as it pays the bills. PHP is fast, PHP is reliable and I get shit done in PHP. That's why I use PHP.

I heard that before and my sibling-poster is correct: You have no idea what my product is, so you can't say anything about the price. If my product were MP3s, 5 bucks per MP3 wouldn't work.

Anyways, your point is not totally invalid. I did a mistake when I changed from free to paid. I started with 99 Cent per month. The goal was to convert as many users as possible. This worked well. Only about 10-20% stopped using my service. However, it made future price increases harder.

Some people would pay more. For some people the service delivers a a lot of value and they'd probably pay 10 bucks a month. However, there are also quite a few people for whom 2 euro is the limit, especially since the long-term costs would explode for them.

And I can't experiment with pricing. The niche is too small to have this go unnoticed. I know, because I tried. Either I increase it globally for everybody or not. To be honest, maybe I didn't find the perfect pricing point yet, I don't know. But it's okay, because I can live quite comfortably.

What a silly statement considering you don't know what the app is actually doing.

Your assumption is correct. I use Paypal, wire transfer (cheap in Germany) and might use Stripe or Paymill soon.

The solution is simple: People can buy only 3 months at minimum. This makes the transaction fee comparably small.

It serves users in a certain niche. They enter some data, it runs some algorithms and outputs the data in a different way. Also, it educates users on the overall niche topic.

Look, what it does, is not that important. You wouldn't learn anything from it about running your own business :)

Edit: Note: Parent changed his post.