We interviewed Melior.AI CEO and CTO Martin and Jose Marcos Rodriguez at Web Summit last year to ask them what young startups need to know before development, which department should get the biggest slice of the budget, and how they managed to get their first client with only an idea.
This is a transcript of an interview recorded at Web Summit, November 2018 in Lisbon. You can read the interview below, watch it, or listen.
Hello Martin and Jose! Tell a few words about your project and what you do.
Jose: We’ve developed AI-based chatbot technology. We have a family of chatbots, mainly for e-commerce. We have an enterprise solution and a pay-as-you-go solution that you can configure to whatever need you have – any language, any functionality – it’s language-independent and domain-independent. That’s the beauty of our technology.
So it works with any language I want? English, Russian, Portuguese?
Jose: That’s right! Completely language-independent. There’s absolutely nothing stopping you from using a chatbot in your language. Or multiple languages, if you want!
Martin: And I will add something to this! What got us started is really the idea that most companies do not have enough data to train chatbots with, and to train their AI with. So we started wondering – how do we democratize it? How do we find a solution that can be applied to businesses that do not have enough historical data?
Would you say the AI element is more useful for big corporations or small businesses?
Martin: Oh, for both! Our application is typically for companies that require interaction with a customer. Whether you’re a small or big company, if you have customers who ask you questions – we can help you.
The most important thing is to validate your hypothesis. Don’t jump right to development without doing market research.
And how old is your company?
Martin: We’ve been around for just over a year.
Are you cash-positive at this point?
Martin: Definitely not. And it’s not going to happen for a while. Right now we are burning cash. We bootstrapped and now we are running a tokensale!
So what are your plans for this year? What are you going to do to come back [to Web Summit] next year in Beta?
Martin: I think we’ll be at Growth next year. [Laughs] Think we’ll skip one. By next year, you will see our technology used a lot more, a lot more companies visibly using our service, and our team will be larger as well. So we’ll have more stories to tell about our clients!
You said you’re practically burning money right now, what do you spend the most on? Marketing? Sales?
Martin: R&D! R&D definitely! Research and development – that’s where our money is going right now.
We’re trying to collect as many good advice from projects like yours as we can, so I’ve got to ask: what is the #1 advice you’ve ever been given that you’ve actually used in your own project?
Martin: Oh, I’d love to answer this first. Validate, validate, and validate! Before you build it – ask people if they will use it. Show it to your potential clients, ask them for feedback, and they will very quickly tell you if you’re on the right path or not.
And how exactly did you organize this process? How did you validate your idea?
Martin: Oh that’s a good story…
Jose: Yeah, a really good one.
Martin: We were all involved in this area of customer support and AI and whatnot, and we saw that companies do not have enough historical data and that it’s not the correct route. So we sat down with a VC, a pretty famous VC. And we said to this gentleman, look, we don’t want your money. But we want to run this idea by you, tell us what you think. And we walked out from this meeting… when all we had was an idea, we walked out with a first client on board. That meeting instantly gave us a customer! [Laughs]
Jose: And I agree completely! The most important thing is to validate your hypothesis. Don’t jump right to development without doing market research.
Don’t ever think that you are the smartest out there and nobody’s thought of your idea before. See what’s out there – there might be other solutions, completely different ones, solving the same problem already.
Would you pass that advice you’ve received to other startups or do you have one of your own as well?
Martin: Definitely validate, but also don’t ever think that you are the smartest out there and nobody’s thought of your idea before. See what’s out there – there might be other solutions, completely different ones, solving the same problem already. And ask your customers first.
Jose: I would say… Look, but don’t copy. Innovation is very important. Small companies and young people have a lot of great ideas that they need to follow. Innovate! There’s still a lot to do.
This interview has been edited and condensed for clarity.