Published by dowlingj on

What is it with European Data Platform companies?

TLDR; We build them up and sell them off (cheap). How do we stop repeating this mistake?

Europe has a long history in Enterprise computing. This article describes the worrying trend of deep-tech European IT companies selling up early, what causes it, and some suggestions on how to fix it. This article was motivated by the acquisition of Data Artisans by Alibaba (congtrats guys, btw!). The problem is not at the individual level of their single company, but a systemic issue for Europe.

The Stories of Alexa and Flink

Alexa is the now famous voice-activated AI product sold by Amazon. But the technology was originally built by Evi Technolgies, a Cambridge-based company in the UK, that came out of research. It was a deep-tech product that needed significant investment to make it a commercial success. From insider gossip, I heard that with money running low they could not convince VC backers to further fund the product. But the technology had huge potential, so the big American companies came sniffing. Google put in an offer that the founders tried to get them to increase, but Google walked away. When Google walked, they were forced to accept a low-ball £26m deal from Amazon.

Data Artisans build the world’s best streaming analytics platform, Apache Flink. Their company is an Enterprise software company, reselling the platform to Enterprise customers. They were the largest data platform startup in Europe, but just sold to Alibaba (not an Enterprise software company!) for €90m. Could they have built a bigger company and become Europe’s darling? Maybe, maybe not. But they had the technology to do so.

Other European technology platforms that have since been acquired or moved to North America include MySQL, JBoss, Elastic, Nokia.

Crossing the Cultural Chasm

Part of the reason European companies exit early is the founders. What european cultures have mostly in common is that they value quality-of-life over material assets. There is not the same drive to become a billionaire. If you can exit your company, and afford to buy a nice house, a summer house, and maybe a sailing boat, you will be considered ‘rich’. This means that when your company is on the cusp of becoming a global player, the business development people will come in and an make offer for your company. And your support network, of family and friends, will encourage you to accept. And in many cases, we do.

This cultural chasm is hard to cross. I think the best way for Europe to get over it is to help redefine success as not a monetary exit value. Did you build a company that contributed back to ecosystem it came from? Did you create sustainable jobs and a sustainable ecosystem? Did you inspire others to follow your lead? In some cases exiting early is necessary and maybe even good, but in Europe it happens too often and is culturally expected, not just accepted.

Why Data Platform Companies Matter

Data platform companies matter because they are at the bottom layer of Enterprise computing stacks. Every application builds on properties and features provided by the data platform. The operating system, the filesystem, the database, the resource management platform – the choice of these systems has huge knock-on effects on the quality of the applications built on top of them. Ecosystems form around data platforms – all the way up to consumer market applications. And ecosystems means clusters of competence, where talent can be found and funded, and startups can thrive.

Fragmented Venture Capital Markets

One charge levelled at Europe that partly explains the lack of global enterprise technology behemoths (SAP accepted) is the lack of deep VC markets. In the US in 2017, $84.2 billion was raised in VC money for IT, compared to €8.8 billion in Europe. That is almost an order of magnitude more VC money in IT.

However, two new initiatives have been very successful in helping to bootstrap technology companies in Europe (from an albeit low base):  the H2020 SME instrument and the European Investment Fund (EIF). The H2020 SME instrument is a way of getting no-strings-attached financing up to €2m to develop a product or market. This funding is extremely important in (southern and eastern european) countries with underdeveloped venture capital markets. Spain have been the biggest winners in the SME instrument program, to date. The winners have, however, been selected by expert evaluators hired by the EU Commission. Naturally enough, the VC community in Europe think they should be picking the winners (not the so-called ‘experts’), so they have got their own tranche of money through the EIF. Northern Europe, with better developed VC funding opportunities, have gained relatively more from the EIF.

The number of VC funds that are backed by the EIF represented 41% of total investments in Europe in 2014. In 2018, the EIF started a new VC fund-of-funds that aims to raise 2.1 billion euro. Anecdotally, these new VC funds appear to be taking greater risks in investing in early-stage technology companies – our deeptech company, Logical Clocks AB, raised our first seed round with two EIF-backed VC funds – Inventure and Frontline.vc. Most VCs we talked with prior to them only wanted to talk about our revenue, not our technology (which most of them didn’t understand). Still, for later stage startups, the larger amounts of VC money are still mostly raised in the USA.

So, how do we get over the “hump”, where European technology companies either get acquired or moved to the USA? The 10-fold VC funding gap is starting to close, but slowly. Is it just a case of having more later stage funds? Partly yes, but we also need to close the cultural chasm, where founders are happy to settle. VCs can help by more encouraging their investments to do the hard work on building the business, not selling out.

Summary

With Data Artisans being sold to Alibaba, Europe lost their leading Enterprise data startup. This is not the end. We, at Logical Clocks, with the world’s fastest hierarchical filesystem and a Data/AI platform are ready to take up the slack. But, this acquisition is a sign that Europe needs change if it is to establish itself as a global player in IT again

Finally, we can say to our customers and users, we will not sell out early. We are in it for the long run, and we are doing our best to build a long-term sustainable Data and AI platform, not just for Europe but for the globe. We will also continue to support Flink as a service on the Hopsworks platform.

References

Follow us on Twitter

Star us on Github



Categories: Flink