The Path to Enterprise AI
Our investment in Dataiku
The modern data toolkit
Organisations of every size and type are becoming more sophisticated in their use of data, and as investors we continue to see tremendous opportunity for startups in this space. In 2015 we invested in Collibra, which has since grown to become the world’s leading data governance platform, and this year we’ve led rounds into two early-stage players: Quantexa enriches and joins together disparate pieces of data, and quantilope provides a platform for marketers to gather insight from their target customers.
But it’s still not easy to become a data-driven organisation or to realise the vision of ‘Enterprise AI’ — that is, augmenting the work of people across all teams and disciplines with AI for more innovative operations, processes and products. You may have the tools to generate, store, catalogue and enrich data, but you still need to answer specific business questions:
How do I maximise customer lifetime value?
When will this aircraft component fail?
How can I optimise this factory assembly line?
Data science is the tool to answer such questions. But in most organisations, it’s carried out in silos and is poorly integrated with the rest of the business. There are never enough data scientists to go round, and existing data science tools only cater for expert users. This means that projects are handed off to a specialist analytics team, and because stakeholders are unable to collaborate effectively during the project, there’s often a frustrating outcome at the end of it all.
Dataiku: Collaborative data science for all
So we’re thrilled to announce our investment into Dataiku, whose data science platform has set the standard in this domain, in a $101m round led by ICONIQ and supported by existing investors Alven, Battery and FirstMark.
Founded in 2013, and with offices in Paris, New York and London, Dataiku has quickly grown to become a global player in the emerging category of Enterprise AI, and is already recognised by Gartner as a leader in this new domain. The Dataiku platform enables users to design and run a data science project end to end, allowing them to rapidly test and iterate on models, and then to deploy models into live where they can have real impact on the business.
Uniquely, Dataiku enables deep collaboration between stakeholders throughout the project. Data scientists work alongside business analysts, data engineers, developers, statisticians and so on. This makes agile, joined-up working possible and has been fundamental to Dataiku’s philosophy from the outset. The ability to shape and define a new space with such clarity of vision is rare for such a young company.
Dataiku also lowers the barriers to entry for data science, meaning that even ‘citizen data scientists’ — business analysts or marketers, for instance — can be meaningfully involved in the project throughout. Even non-expert users can start learning about data science: at first, they may opt for the ML-powered ‘be-driven’ route, but in time they gain confidence and begin to unlock the power of the platform. It’s hugely exciting to see a company that’s not only built a great product, but that’s also having a positive, meaningful impact on its users.
France: the new hotbed of B2B
Dawn is the largest VC fund dedicated to B2B software and B2B fintech startups across Europe, and Dataiku marks our first investment in France. Evgenia, who leads our coverage of France, has written a great overview of the market which explains why it’s such an exciting place to invest. In particular, she notes that France is a hub for AI and B2B:
Facebook opened its AI research centre in Paris in 2015 (its first outside of the US), and back in 2013 hired French native Yann LeCun to lead its AI research. Rakuten and Google both also have research centres in Paris, and there’s a strong academic background in the field.
Within France, there are now at least 180 AI startups. In keeping with that trend, there were at least 6 early stage exits to the big US tech giants — Amazon, Apple, Alphabet, Microsoft and Facebook — out of the 53 that occurred since 2011.
Dataiku also epitomises a maturing European tech ecosystem in which a successful startup gives rise to the next generation of winners (as most famously happened with PayPal in the US): many of the Dataiku team previously worked at Exalead in the 2000s — including Dataiku CEO Florian Douetteau, who’s even compiled a handy list of the ‘Exalead Mafia’. We’re hugely excited to renew our focus on the French market as we look for the next B2B startup with global ambitions. Dataiku is just the beginning for us in France.