A Q&A with Dataiku co-founder and CEO Florian Douetteau. The Paris-based data science startup announced in late February that it has completed a €3.2 million Series A funding round. Investors include Alven Capital and Serena Capital. It was founded in early-2013 by Douetteau, CCO Marc Batty, Chief Data Scientist Thomas Cabrol and CTO Clément Stenac.
SUB: Please describe Dataiku and your primary innovation.
Douetteau: Dataiku provides a software platform called Data Science Studio (DSS). Dataiku’s software platform aggregates all the steps and tools necessary to get from raw data to production-ready applications. DSS allows both beginners and experts—from business analysts to data scientists—to design, build, and deploy their own data-driven and predictive applications to answer business needs such as, but not limited to, churn prevention, logistic optimization, predictive maintenance, fraud detection.
SUB: Who are your target markets and users?
Douetteau: Our users are data analysts and data scientists alike—experts and beginners. They use DSS in order to quickly prototype new ideas with data. For example, a user could ask: “Can I build an interesting customer segmentation based on the first ten minutes of a customers’ interaction with an application?” If the idea proves worthy, they would use Data Science Studio in order to turn the idea into a running application. In this example, the user could build a production stream that would continuously build user segments from fresh data. The user could then push and use this continuously-built user segment data into a business application—emailing, pricing, etc.
As a result, next generation data teams can circumvent complex IT extractions, transformations, production processes in order to support agile, lightweight, predictive applications. Today, we’ve already had successful experiences in the ecommerce, transportation, and insurance industries, where data teams use DSS in order to build production-ready predictive applications.
SUB: Who do you consider to be your competition, and what differentiates Dataiku from the competition?
Douetteau: Our main competitors are traditional data mining providers such as SAS Institute and IBM. We provide a web-based, collaborative, and visual experience that enables users to easily complete data transformations, statistics, and machine learning. For advanced analytics, this type of user-friendly and accessible experience is too rarely, if at all, available.
SUB: You just announced that you’ve raised €3.2 million in Series A funding. Why was this a particularly good time to raise more funding?
Douetteau: In the first 24 months of Dataiku, we were able to validate our vision that simplified machine learning is the creative leap that will revolutionize the way people build their business applications. When we decided to raise funds, we had reached a stage where we were profitable and needed additional fuel in order to expand our reach beyond France and hasten our product development on an international level.
SUB: How do you plan to use the funds?
Douetteau: With these funds, Dataiku will accelerate its technical and sales development in order to boost its international expansion. The company will open its first U.S. office in New York in the upcoming weeks.
SUB: What was the inspiration behind the idea for Dataiku? Was there an ‘aha’ moment, or was the idea more gradual in developing?
Douetteau: Rather than an ‘aha’ moment, I’d say it was more of a ‘let’s do it’ moment. Everybody talks about data science and lots of people want to enter the data analysis and science field. However, very few applications seem to support data science as it is practiced today. Then, why not build the ideal environment to make data people more productive and creative when answering business problems with data?
SUB: What were the first steps you took in establishing the company?
Douetteau: The first step for us was to build a great founding team. We are the unusual number of four co-founders. Marc is the sales guy, Thomas the data science guy, and Clement the CTO. Data science today is about merging mathematics, tech and business into a single pot. To be relevant, you need to combine these capabilities.
SUB: How did you come up with the name? What is the story or meaning behind it?
Douetteau: ‘Dataiku’ comes from ‘data’ and ‘haiku.’ A haiku is a very small and structured Japanese poem. The essence of haiku poems is to oppose two images or ideas—small solutions to big problems. Furthermore, similarly to building a Haiku poem, we believe data projects should be a structured process, a single flow from start-to-finish for the organizations and people behind them.
SUB: What have been the most significant challenges so far to building the company?
Douetteau: A significant challenge is to avoid being disturbed by rumors and noise; with big data today you have large expectations and large funding. It is important for us to keep our focus on our product and on our users.
SUB: How do you generate revenue or plan to generate revenue?
Douetteau: We generate revenue through annual subscriptions to the Data Science Studio software.
SUB: What are your goals for Dataiku over the next year or so?
Douetteau: Our goal is to make Data Science Studio a well-known and widely-used tool.