Traditional innovation scouting (yes, it’s an oxymoron)
Is there a problem with corporate venturing today? Oh no. Corporate venturing is good, as well as being necessary for our world’s evolution and for the resolution of massive issues. As a critical step in the innovation lifecycle, collaboration between small agile innovators and giant industrial strengths is the nuclear fusion that can spark a new life. In fact, the problem is how those industrial strengths scout for innovations.
Think about it. In our world that has been pervaded by technology and data, how is it possible that an activity mobilising zillions of dollars still relies on… opportunity and manual search?
These days, no one would understand running a search on any given topic without asking Google first. Right. However, when it comes to specific sectoral searches — like in HR (head hunting) or in corporate venturing (startup scouting) — it’s rare to find those who consider it logical to have some form of technology helping with that. It is simply not there yet. Of course, you can subscribe to a startup database, and you will have your Altavista to scout innovations. But where on earth is artificial intelligence in those domains? It’s like these sectors are 20 years behind the curve.
For the times they are a-changin’
You better start swimmin’ or you’ll sink like a stone, for the times they are a-changin’. In the current decade there is no excuse for running startup scouting activities as though the world had not changed. In recent years globalisation became a reality, data volume exploded, data science made huge progress, the information age invaded our lives and real-time is now the norm. In the twenties of this century, access to information is universal, immediate and filtered — and so should startup scouting be.
However, most of today’s corporate venturing connections are still made through opportunity, events, long research projects or hard-crafted processes (hackathons, corporate incubators, etc).
Corporate venturing in general, and startup scouting in particular, are still businesses built on personal relationships. Today, AI is there to help, and yet it is underutilised in those fields. As a result, the reach remains close (or even poor, if you are far away from such ecosystems) and the quality of projects is merely average because they are often choices made by default. The potential of data science is underestimated by most companies — they prefer investing millions of dollars in internal R&D even though a cool, younger company somewhere might well have done the hard work before, with faster and better success.
Companies are not investment funds
VCs are looking for financial returns. Plenty of data is available for that, generated from tools like Crunchbase, Pitchbook and other CB Insights. Those platforms are used by investment professionals, and also by corporate development teams in large industrial companies. But what do they need when searching for relevant startups? They need insights about startups’ actual activity, technology, products, features, models and real-life concerns. In comparison with all this, financials come second. However, in startup databases, there is generally little non-financial information. Besides, when you focus on figures and financials, the odds of getting outdated data are much higher: startups are born, evolve and die quickly, with each life step likely to change the data. Staying up to date is very difficult.
Currently, the world counts 472 million entrepreneurs. A new business opens every three seconds. For companies, that makes it very difficult, tedious and time-consuming to spot the right startups to work with. In big corporations, extended teams can be dedicated to spotting and analysing startup files. Meanwhile, in midsized companies this work is often done by a single individual who has this task on top of their other duties. It’s either very expensive or it’s inefficient. But either way, it’s not optimal.
When it’s not done on a self-service or opportunistic basis, companies use the following three means for their startup scouting activities:
Whichever of these methods is used, opportunities are being missed. New technologies like machine learning, NLP and cutting-edge data science techniques are almost non-existent in that business. In the startup area (which is increasingly chaotic, worldwide, too fertile and noisy), a new approach to scouting startups would involve personalisation, context and networks that can be tapped with artificial intelligence. That’s what Novable is doing.