Jindřich Josefus and Martin Čejka. Two friends who have been around IT since childhood. During that time, a decent community of "kids" has formed around them. And so, when they were looking for the next steps to take, out of dozens of ideas, they chose the one that, according to them, is the middle ground between monetization and the level of innovation - a public opinion poll. And so the startup Foxily was born.

The manager of the business incubator, Jan Wedlich, talks about you as a possible mover who will change quantitative research. Can you explain to us what the revolution consists of for you?

Obtaining public opinion data or market research is demanding on human resources, it is also expensive, time-consuming and negatively perceived - just imagine how an unknown number calls you, and after answering the phone, you are asked what your favorite radio station is. We propose a virtual assistant using artificial intelligence methods. While a paid call center employee initiates a single call with an uncertain outcome, a virtual assistant conducts hundreds of parallel conversations in a social network environment that potential responders use almost continuously. The virtual assistant is friendly, rewards respondents, uses elements of gamification. The respondent is also satisfied with the fact that he can write when it suits him - for example, on the way from work. If we then want to know what the public thinks about, say, yesterday's events in Ukraine, we can come up with results within a few hours, and that is a huge difference compared to weeks of polling. A research entity can better utilize its resources, receive more orders, reduce the demands on its hardware and software equipment, and draw many other benefits from the automation of the inquiry process.

So do we understand correctly that you can solve the interviewer problem for research agencies? Or do you want to become one of such agencies?

In this case, it is not yet clear. We did not come to crush the competition thanks to technological knowledge, nor to take away the work of call center employees, of which there is, on the other hand, a great shortage. In order to become a research agency, we lack the knowledge of constructing the right questions, analytical processing of data and their visualization. It is certainly not unattainable, but for now we will become partners of research agencies, when we will be able to eliminate their problems with data collection and improve their position in the market.

How can you use a chatbot to monitor the quality of human responses? Don't they just click through and you get irrelevant information?

We believe it's just the opposite. If you put up an online questionnaire, I think a lot of users under a fake account will fill it out as quickly as possible, regardless of the veracity. It will also attract only a certain group of people, just as only a certain group of people will not hang up the phone calling a call center. In our case, the respondent is motivated, not "annoyed". For the quality and speed of filling, his credibility rises, and thus the reward for the conversation. I can also imagine a high-ranking manager who chats with a virtual assistant for long periods of time, and sometimes goes out to dinner for the rewards given. Of course, there are methods that can often detect false claims. We want to implement them.

Your huge advantage is that you have the entire development team in-house. Did you build this team for the given project or did you "just" redirect yourself from another field?

We have been in the IT world since high school. Meanwhile, a large network of capable programmers has formed around us. When we sat down and started coming up with innovative ideas, we created a really big sheet. We sorted it - the idea of a virtual assistant as a pollster was such a golden mean between possible monetization and the rate of innovation.

You are part of a business incubator - how did you find out about it?

I can't remember that quickly now. We have known about it basically since its inception and have followed it from afar. When we decided to start our own company in the Ústí Region, where both founders are from, we did not hesitate to submit an application.

Could you currently evaluate the positives as well as the negatives of the business incubator?

The first positive that one perceives immediately upon arrival is the great atmosphere - the entrepreneurial spirit. But he will get much more there - consulting with experts can take a budding entrepreneur over many obstacles. We would also like to highlight the role of Honza Wedlich - he is the face of the incubator, has insight into many fields, and seems to us to be the most qualified person in such a place.

And what about the negative? I could imagine a permanent space for incubated companies, where they could meet and help each other, where they could bring their first customers, where they could paste their first business plan and have it in front of their eyes. I would appreciate it if the workshops worked directly with incubated projects and not on a general level, this is of course very time-consuming. It's not the fault of the choice of workshops, it's just that entrepreneurship is a much more complex activity than the startup culture presents it.

What are your plans now? What awaits you now?

We were forced to slow down development. As a company, we have to make a living, we are implementing processes, accepting orders for software development, we have new employees, we are trying to build a place on the market. We also thought about an alternative way of funding sources, i.e. finding an investor or using subsidy titles. It's still an open question for us. We have managed to implement MVP, we know that it is technologically passable. Programming work on a larger scale, marketing activities, legal services are needed. We consider establishing the right business partnerships to be an important task. Development should start again in mid-June, when part of our capacity and budget will be freed up.

Thank you for your time and we wish you the best of luck.