From a personal software project to a full-time startup!

Today, I want to share with you my personal look back on 2015, in particular, how I transformed my personal software project - a home cloud platform named Datagnan - into a full-time startup this year, which lessons I’ve learned on this path, and which aspects are covered by next year’s roadmap.

At the beginning of 2015

Until the end of June 2015, I was working part-time as a Post-doc for the Karlsruhe Institute of Technology (KIT). Part-time, because I wanted to have enough free time for my personal research on Datagnan while also earning enough money to cover my monthly expenses and having a health insurance. While this was a very comfortable situation, I decided in May 2015 to discontinue my employment contract and focus completely on Datagnan. I registered a company on June 6th, did all the paper work, established a financial bookkeeping system (which I decided to do myself since I do not have many invoices or expenses yet, and also have basic experience from the past as an independent software developer), and set up a contract for myself as an “executive director”. It may sound great to name yourself an “executive director” of a company, but to me it isn’t. Nothing has changed compared to before, only that you have one more responsibility (at least on paper) and obligations that you did not have before. And sometimes, to be honest, especially if you work in your office room at home not yet earning any money with your idea, the fact that you officially run a company feels kind of unreal.

Nevertheless, I was highly motivated and enjoyed having enough time to work on Datagnan. I spent my days and weekends developing components and writing one line of code after the other. And when it was foreseeable that a working prototype is close, I set up a webpage for Datagnan, created marketing material, and eventually participated in a local startup competition in order to get feedback from people outside of my social circle (naturally, friends tend to be less critical with your work and rather focus on how brave it is to start a project or company and leave paid work). The participation in the competition was a great experience. Although Datagnan qualified first in the online voting, other startups (mostly B2B-oriented ones that are at a further stage of development) were selected as winners by a committee of startup experts. The result showed me that, unless your product or innovation is released or plain simple, it is still perceived as an idea only by others, and that it is difficult for others to understand the innovation of your product if they haven’t experienced it yet.

Today, communicating the idea and innovation behind Datagnan to others is the biggest challenge. Not because I am bad at explaining, but due to the fact that people have different associations when I am talking about personal home cloud storage. And depending on the association, I have to explain in different ways why Datagnan is different to what they associated. My next blog post will focus solely on that topic.

Lessons learned in 2015

Working on an end-user product takes much more time than I anticipated, and dealing with all the legal, financial and marketing aspects is a big time consumer as well. Although I discontinued my part-time job at KIT in order to work full-time on Datagnan, I regularly have the feeling that I am only as fast as before. But this is not true. Instead of spending time on my previous part-time job, I spend time on new tasks I have to deal with now. Tasks that are necessary and equally important as the software development tasks. I noticed that, as a computer scientist, I tend to consider “work” only as the time during which I make progress on the Datagnan software — as it is the end-user product and the reason why I started. I took some time that this is wrong.

Finding the right balance between “solid software engineering” and “make it work” is a continuous challenge. Datagnan will provide an alternative to cloud-based solutions, and user privacy is a primary objective. Hence, I want to avoid bad practices from the very beginning. All datagnan software components are developed using a test-driven approach, i.e. after designing the component API I write test cases and only then develop the component. All components are built and tested every night on different operating systems and platforms (Linux, Windows, Mac). To ensure that the components work together as envisioned, additional integration and stress tests are used. Of course, it takes more time to employ and adhere to such an engineering process — compared to making it just work — but I believe that it will be beneficial in the long run. Dan Luu recently wrote about this balance in his great blog post How Completely Messed Up Practices Become Normal.

Running a startup is a big roller coaster — in every aspect. After registering the company, setting up all required legal documents, kick-starting a financial bookkeeping system, etc., I felt very positive and my motivation was above 100%. But there are also times during which motivation drops below 100%. Especially when I meet with established but skeptical parties in the local startup community. My biggest challenge then is to distill valuable feedback from all comments I receive, to “ignore” everything else, and to keep a positive attitude — which is not easy if the feedback comes from a party that knows the startup universe much better and for much longer than I do. A strategy that helps most in these situations: I let go of my fears and allow Datagnan and myself to fail (read Permission to fail by Michelle Wetzler if you want to dig deeper into that psychological topic).

The roadmap for next year

2016 is definitely the year in which Datagnan will see its light of day. First, Datagnan will be released for the Linux-based desktops as well as for Linux-based servers (e.g. to run it on your own NAS) in order to demonstrate how Datagnan works. What will follow is the release of the mobile version for the Android ecosystem and a period during which the focus will be on gaining user traction. To be able to scale beyond a few hundred users, it is further necessary to increase the resources of Datagnan — from a financial point of view and in terms of manpower.

There’s a lot to do next year. But for now, I will call it a day and wish you all a happy new year.

Originally published at

Show your support

Clapping shows how much you appreciated Jens Mittag’s story.