Interview: Bert Svab, Liftago. Why hackathons matter and what is it like to lead BI at an ambitious, transportation startup at 25.

Bert Svab, Aimtec Hackathon

PB: It’s great having a chance to virtually sit down with you, Bert. Would you like to start by introducing yourself and what you do?

BS: Sure. I’m only in my mid-twenties and I currently work as a Business Intelligence Analyst at Liftago, a transportation startup in Prague, Czech Republic. I am also a member of Google Developer Group and I organise hackathons and industry events in my native town of Pilsen. I have almost accidentally studied at a local electrotechnical university, but I quickly realised that I didn’t want to become a straight-cut programmer or electrical engineer.


PB: Tell us about Liftago, the company you currently work for.

BS: Liftago is based in Prague, Czech Republic. We are a startup with a long-term vision to create an open marketplace for urban transportation and replace the need to own cars into city centers with an on-demand mobility. Take a look at Google Play! Liftago is currently the best-rated ride-hailing app around here. It provides passengers with a choice of nearby drivers with rating, price and other real-time information. To keep in touch with us check out our Blog or follow our Instagram profile.


PB: How does the C-Suite at Liftago look at data? Is it seen as a strategic opportunity or more of a box to tick on a buzzword list?

BS: The attitude towards data was one of the reasons why I decided to join. The CEO is an ex-engineer, so there is great importance put on data analytics and I’d say the entire company really adopts data-driven mindset, whether we are talking about product development or marketing.


PB: What does the data team look like?

BS: Well, you are talking to it right now. BI is my responsibility and Liftago overall has a team of 27. My first team member is an intern from Poland who studies quantitative finance and she is super-skilled in data modeling in R, while at this point the biggest chunk of my work is the ETL and BI. Our cooperation works perfectly and, for example, we have built a brand new dataflow for marketing incentivisation within two weeks. We are considering to add another full-time team member soon but we also work with consulting company Trologic.


PB: Outsourcing in BI, using agencies, consulting companies and micro teams is a big topic these days…

BS: I really like working with our consulting team. Trologic brings different angles; they are often able to spot things and issues insiders might miss and it is also really important to have your hypothesis validated from outside.


PB: What is the best thing about working with Liftago from analyst’s point of view?

BS: What I like about Liftago the most is openness. We collaborate with local open-data initiative, organise and attend hackathons where we are trying to join public transportation data with ours, to help users move around as smooth as possible.


We recently opened up one of our data sets and challenged the tech community to work with it in different ways by tagging #jedemeData. The outcome was amazing and you can see the results here.

Pavliscak Matus

Open data by Liftago and Power BI – Matus Pavliscak


Paradox of choice by Martin Polacko

PB: What are the most critical components for success in a data-driven startup?
BS: The way I look at things, there are two important success factors for data analytics. Time and availability. For example on the availability, we use MongoDB as our primary database for all our data, but because of Mongo’s level of difficulty, only a handful of people in the team were enabled to work with the data directly. This has changed with Keboola as our ETL framework in picture where reports can be created simply with SQL so it is a lot easier to work with for a much broader team.


PB: How do you guys share insights with people who are not ‘data savvy’, the people on the front line of your business, the drivers?

BS: We take this really seriously. For every new driver we hold orientations where we share really interesting tips and tricks we get from the data. Of course we don’t try to overwhelm them with charts and make sure we deliver these insights in an understandable way.


PB: Let’s rewind a bit before all the cool data magic touch you do now. How did you get started, what was your first job?

BS: My first job straight out of university was really different. AIMTEC, the company I joined, has been in the market for 20 years now so you could say it is more traditional when it comes to IT services. What I tried to do when I started there was really to evangelise the ‘data-driven’ mindset I believe people should have, not just among the development team but also amongst the business users and consultants. I wanted us to get rid of Excel, start using more effective tools, automate processes and set up standardized framework for measurement of our KPIs across the board. In every occasion I tried to do some “disruptions”  – in fact,  innovations in general were the main subject of my job.  It wasn’t easy but that is the only right way, so I’m glad it has been started.


PB: That’s not easy. People are generally opposed to change, and disruptive approach like this isn’t always appreciated by everyone.

BS: It wasn’t easy at the beginning and it is a long-term effort, but the great thing about data is that you can convince people with results. I always seeked innovation and have always been attending community events and meet-ups and hackathons. That’s a great way to connect with people, to keep learning new things and get constantly inspired.


PB: How did you get involved in organising data hackathon on your own?

BS: Well, I’m really active in this community and I had the chance to take part in many others, including the Data Festival which Keboola team organised, and I decided I needed to do something to keep this community going in my hometown. My previous company was really supportive, especially HR who saw it as a great opportunity for employer branding and recruitment. Solutions we built were typically around warehouse management so we decided to spin the hackathon around IOT which, together with augmented reality, was really an interesting topic for my company at the time.


PB: How did the first event go?

BS: It was really a nice success and despite some initial pushback, everyone in the company eventually loved it. We had, at any given moment, about 70 people actively working on the solutions or in other workshops for non-programmers which were part of this event.


PB: What did you enjoy about it the most?

BS: It is really the opportunity to connect people from both sides: the engineers and programmers on one side and the business people on the other. I don’t really feel I belong 100% to either of these groups and I think a dialogue between those two is absolutely critical for the future success of our field.


PB: This is a really interesting point. We see very few people ‘in the middle’. What brought you there, what is your technical knowledge, and why do you define yourself as the mix between the two?

BS: I never saw myself as a programmer before university, but I was classically self-trained from the book. I learned the usual HTML, CSS or PHP. Later in uni I took some classes in C, Java and MATLAB. I really like mathematics so MATLAB was something I really enjoyed. Much later, at one of the community events, I was really impressed by some outputs from analysis in R and I really saw that as something I wanted to focus on. R is my favourite but I also frequently use SQL. On the other hand, I’m also much interested in economics, project management, and business in general.


PB: You are 25, literally at the start of your career, in your second job, but this entire industry keeps changing so fast. What do you want to focus on in terms of learning and your own growth?

BS: Rather than to say that I’d like to focus on a new language or learn some new tool, I’d like to keep growing my skillset more horizontally. Like with everything else, practice makes masters and when I have a chance to build similar project, any analysis in R, ETL or reporting schema in BI tool for the 5th time, it will be much easier to avoid some novice mistakes and the entire process will be much smoother. Also, I believe communication is one of the most important areas of data analytics, so I think that’s something I want to keep working on.