Starting our own Company?

In the dark and cold Norwegian winter of 2002/2003, Harri and I hung out quite a lot together with our girlfriends. Both our happiness with the job at Trolltech decreased as we saw our ideas not resonating with Trolltech’s management. In addition to that, after a few years in Norway, the long winters made life not so enjoyable.

We started discussing ideas of a tool to automate testing of the applications written using Qt. At this time, the topic of automated software testing wasn’t that hot, esp. when it came to testing GUI applications. There mostly existed only a single tool for this in the market, Mercury WinRunner, which worked only with native Windows and Unix Motif applications, but not with Qt.

Through our work at Trolltech, we know how many professional and also mission critical Qt applications were out there, and there must be a market for a tool to do automated testing of these applications.

The more we brainstormed about the ideas, the clearer our ideas got that we’d like to found a company and build this tool.

While we earned good money and both had enough savings to allow us to work with no income for several months, we wanted to ensure that we can start earning money early on to support building the company.

This was in early 2003, just after the Dot.Com bubble burst. So we didn’t even think of any outside funding or approaching investors. From the start, we wanted to bootstrap the company and grow it organically. One reason was certainly that software companies didn’t really get investments at this time, but we also didn’t want to repeat what we saw in the later years at Trolltech.

While we were determined to go forward with our plan, we still needed to get clarification on a few important points. First of all, we needed a name for the company. Secondly, we needed to decide where to found the company. 

Then we needed to find a viable business model to start generating some revenue early on while still being able to spend most of the time on developing our product. This also meant designing the product so we could soon come up with a useful MVP to start generating revenue with product license sales. 

And finally, we needed to decide on the timing to quit our jobs at Trolltech.

Initially, our plan was to move to Silicon Valley. We both have been there a few times to represent Trolltech at trade shows or to visit customers. We inhaled the vibe and really wanted to start our company there.

But without an American on board, founding a company in the US seemed like a major hassle. So we quickly discarded that plan.

Plan B was to move to Hamburg, Germany and initially live with Harri’s parents. This would give us time to look for apartments for ourselves and an office. After some back and forth, this plan seemed reasonable so we had one more item figured out.

Coming up with a company name was fun, but also challenging and not that easy. We didn’t want anything too lame, so it should be some kind of fantasy name. Of course, we also needed to ensure that we could get a .com and .de internet domain for the name and that searching for the name on google brought no meaningful results to ensure, once live, it would be easy to be found.

Esp. to meet these requirements, we figured to best create a name by combining words. So we put up a list of nature-related and tech-related words and started combining them randomly.

One of the name candidates we really liked was EggPlant. But unfortunately, this name was already taken although we couldn’t really find out more about that business. Funnily enough, EggPlant would become one of our main competitors a few years later.

At some point the combination of frog and logic created froglogic which we instantly liked. A quick check, and we purchased the internet domains.

What we didn’t think about then was that pronouncing froglogic with no vowel between the consonants g and l made it hard to pronounce for people of certain native langauges, such as German. Our German customers often referred to us as frologic which is why we later also purchased the frologic.com and frologic.de domains.

After the company name was set, we debated how to call the product we were going to build. We didn’t want to limit ourselves to one product for the future, so a product name same as company name wouldn’t work.

Initially we started with a pretty lame name, F(rog)L(ogic) Tester. While self-explanatory, we wanted to find something better which people would remember. Over a long dinner we bounced ideas around and finally came up with Squish as the name for our testing tool. The thinking behind it was that a test tool squishes bugs (i.e. eliminates software errors).

Ready to get started, we quit our jobs at Trolltech and started organizing the move to Germany. Mid of June 2003, together with our belongings, we took the ferry from Oslo to Kiel and drove to Hamburg where Harri’s parents accommodated us and we could set up a first, preliminary office.