Deep tech is the future, but not all deep tech is created equal. The most successful companies over the next decade will be those who know when to move.
There is no doubt that “deep tech” is one of the most talked about topics in technology. We constantly hear how deep tech is having a profound impact on important segments of the economy such as autonomous vehicles, robotics, smart homes and cities, medical devices, cleantech, fintech, and energy efficiency. Almost every facet of modern society and the economy looks ripe for disruption.
But before going too far with deep tech, it is instructive to define what the phrase actually means, as there is quite a bit of confusion around its nature, where and how it can be applied, and the general scope and practicality of purportedly deep technologies.
How Should we Think About the Future of Technology?
In the words of futurist Horace Dediu, “Those who predict the future we call futurists. Those who know when the future will happen we call billionaires”. The important message of this quote is the impact that timing has on the way technology matures. In fact, timing is probably one of the most critical components of any investment decision. But it is only one of several important dimensions that we use to inform our investment decisions.
What is Deep Tech?
Deep tech can span across many technological areas and can impact diverse business applications. On the technological front, these can include innovations in processing and computing architecture, advances in semiconductors and electronic systems, new kinds of batteries and electronics, machine vision and speech technologies, general artificial intelligence and machine learning, haptics, and many more.
Jared Friedman, Partner at YCombinator, defines a deep tech (or “hard tech”) company as one that will take a lot of time and money to build its first product, a product which it may not actually be possible to build. This stands in contrast to a traditional technology startup, which can typically get a product or service off the ground and into the hands of users with little more than an MVP and some audience capture techniques aided by Facebook and Instagram.
Another way to think about the difference between traditional technologies and deep tech is in terms of their attendant risks. Traditional technology businesses tend to be most focused on market risk, whereas deep tech businesses tend to be more focussed on engineering risk.
A traditional technology company will try to exploit an existing technology stack by using it to open up a new market, running the risk of failure if its business model or product-market fit aren’t right. That’s market risk.
A deep tech company, on the other hand, will not be as focused on business model innovation — at least in the first instance — but will rather concentrate on proving that a particular piece of new technology or a combination of new technologies is even feasible. These activities all attract significant engineering risk.
This is not to say that deep tech companies do not have market risk concerns, merely that they are irrelevant if the company is unable to address the deeper engineering risks associated with their technologies. It is simply not worth even attempting to address a market if the underlying technology does not — or perhaps even cannot — work.
This engineering risk distinction is useful because it broadens the definition of deep tech beyond any specific technology, and frames it in terms of an existential risk of failure. Because the risks of failure are so high, it follows that there are outsized returns available to those that can execute successfully.
With this more nuanced definition of deep tech, it is important for entrepreneurs and investors to understand several critical investment factors: What specific investments should I make? How does timing affect those investments? And: What frictions and/or lubricants might make or break those investment decisions over time?
The Deep Tech Divide: Pioneering vs. Visionary
The first thing to do is break up the domain of Deep Tech into several categories. These will allow you to concentrate on the elements that are most relevant and, vitally, possible to realise.
Visionary Tech is Deep Tech that almost no-one can do yet and, even if it is doable, it is limited to the likes of the Big9 or even nation states. This is tech that has the potential to change the world, but it isn’t going to happen this year. Often, there are genuine computer science problems that remain unsolved with this type of technology. Type 5 fully autonomous vehicles would be a good example of this.
Pioneering Tech is deep tech that is exploitable right now (or at least imminently) by entrepreneurs. It is available to operate and scale with components that might not be off-the-shelf, but that can at least be assembled with a suitably skilled engineering team. Most, if not all of the basic computer science problems have been solved. The tech is novel and new, but it works, and importantly, there is the infrastructure to support it. “Autonomous objects” that run around pre-defined routes, like that pioneered by Ocado in the UK, would be a good example of this kind of technology.
Commodity Tech has passed through its deep tech phase and matured to the point where just about everyone can do it. The raw tech is fully worked out, infrastructure exists to deploy and operationalise it, and there are industry-standard examples of it working at scale. “Lane departure technology” in high-end sedans is an example.
Exhausted Tech has been more or less exhausted for innovation. Diesel-fuelled internal combustion vehicles might be a good example of this kind of tech in the mobility space. Cathode ray black-and-white televisions might be another example in the media space.
A company looking to have an immediate impact and lead a market might choose to become a tech pioneer, applying scalable deep technologies to create new business models and markets, while perhaps deploying more commodity technologies to existing products so as to stay up-to-date with consumer needs.
What Deep Tech is Ready Today?
The visionary tech of today will relatively quickly end up as the pioneering tech of tomorrow. It is therefore important to constantly scan the horizon for Pioneering technologies that are about to wash up onto what we refer to as the deep tech shoreline.
The deep tech shoreline represents the sweet spot where pioneering deep tech is ready for production rollout and scaling. At DV, we are very aware of the full suite of visionary deep technologies that are emerging, but whereas the opportunities to exploit these is generally the purview of academia, The Big9, or state actors that have the resources to do the necessary hardcore scientific research, pioneering deep tech is available today to the world’s brightest entrepreneurs and the most forward-thinking corporates. These are precisely the people that we love to work with across at DV, and also the area in which the majority of technology companies will want to operate.
Identifying the Right Deep Tech
The next thing we do is divide up the deep tech landscape into a series of technology corridors. This is a simple idea that will allow you to group broad swathes of the technology landscape for the express purpose of focusing on the ones that matter the most.
Right now, for example, machine intelligence, IoT, decentralization (blockchain and crypto), and alternative realities are at or approaching market readiness at various levels of maturity. We regularly see companies and products using them for practical purposes, and we have been involved in creating several innovative pioneering deep tech ventures for corporate partners during 2019.
The Right Time for Deep Tech Innovation
Once you’ve understood the technology it makes most sense to deploy, it’s still necessary to develop an understanding of how technology matures in order to execute at the right time.
After spending significant time working with corporate partners and entrepreneurs to launch new businesses, we have noticed that technology innovations tend to move through a fairly common pattern.
Whilst this pattern is well known to successful entrepreneurs, it is not always well understood by corporates. The reasons for this misunderstanding are due in part to the innovation antibody response, but misaligned incentives, unrealistic expectations, willful ignorance, and cultural fear of failure also play a big part.
The first phase of disruption occurs with the advent of new technology. The technologies (or, more often, confluences of different pieces of enabling technology) allow entrepreneurs to try out new business models and new customer experiences, and to create new products and services that sit on top of those underlying innovations.
While there is a lot of focus on the first phase, opportunities with raw technologies on their own are typically just the start. Based on our experience, we have learned to sequence technology maturation into four phases:
When a new technology arrives, opportunities emerge for new types of startup and new businesses whose primary opportunity area is to innovate with or on top of the new technology itself. Once businesses begin to exploit the technology, opportunities emerge for new types of infrastructure to support these new technologies and interactions between users at scale. Once new infrastructure allows new technology to operate at scale, new types of marketplace and business operating model emerge to facilitate interactions between producers and consumers of the new technology. Finally, once the tech, infrastructure, and operating models bed down, opportunities emerge for new types of aggregation and disaggregation to occur on top of the new technology stack.
For example, consider the development of ridesharing. It emerged out of the confluence of a number of different technologies, some new, some old: 3G and 4G , GPS, touchscreens, CPU power optimization, and a revolution in user experience. These elements then combined to enable the modern smartphone.
As soon as the utility of a smartphone-type device like the original iPhone became apparent as users flocked to own one, telcos took this as a clear signal to invest in supporting infrastructure: Ubiquitous wireless broadband. Smartphone app stores launched, solving the problem of mass distribution of apps to end-users and thereby unlocking new marketplaces and business operating models.
At the end of this sequence, we see the emergence of ridesharing companies like Uber and Lift who connect end-users seeking mobility solutions on one side of the market, and private drivers using their own vehicles on the supply side. Although not the first example of such a network based on smartphone technologies, it is one of the clearest examples of the kind of demand aggregation network that this technology stack unlocks.
It might seem as if Uber and Lyft appeared on the scene out of nowhere, but their success, or even existence, would not have been possible without these previous steps.
This pattern of innovations washing through the economy, from the exploitation of raw technology at one end, through to supporting infrastructure, marketplaces and operating models, and then on to increasingly more sophisticated aggregations and disaggregations at the other end, happens repeatedly. As a technology matures, it triggers potentially large value destruction and creation opportunities, redistributing existing profit pools to the participants able to best exploit the tech and the new infrastructures, marketplaces, operating models, and aggregations and disaggregations that build on top.
So how can companies or entrepreneurs learn from this?
If a technology is immature, then it doesn’t make sense (yet!) to look for opportunities with marketplaces or business operating models. Instead, entrepreneurs should be looking for opportunities to develop the raw technology first. Then, as use cases emerge to use the new technology at scale, entrepreneurs should look for infrastructure plays. When the infrastructure for larger scale deployment is in place, entrepreneurs can look for marketplace, operating model, and new aggregation/disaggregation opportunities.
The lesson for innovators and entrepreneurs is to make sure that they are aware of where in this maturity chain a particular technology stack sits, and make investments accordingly.
Understanding Lubricants and Frictions
The final factor worth considering is the effects of lubricants and frictions on the progress of technology.
As we have seen, tech tends to go through a predictable set of phases. Most technologies start off as a raw technology in a lab, academic or industry context, and then mature over time to either widespread adoption and use or failure and irrelevance. Determining whether a technology will succeed or fail is critically dependent on how it is either accelerated by lubricants or hampered by frictions.
Technologies can have their adoption accelerated by coincidence with complementary technologies or business models, societal or political context, or sometimes just dumb luck. Ultimately, timing is one of the most critical factors for success. Alternatively, technologies can have their adoption hampered by incumbent technologies or business models, leap-frogging technologies, societal or political context, or failure to live up to expectations. Sometimes, failure may be completely exogenous. In some cases, both frictions and lubricants may work as opposing forces on the same technology.
The bottom line is that understanding how frictions and lubricants can hamper or accelerate a technology is fundamental to working out when it makes sense to invest. Get the timing wrong, and the deepest most disruptive technology will fail to deliver customer value.
Not All Deep Tech Is Created Equal
The key to leveraging the power of deep tech is cultivating this awareness in order to correctly identify the sweet spot. Being truly innovative and achieving success takes a degree of calculated risk, but by understanding what deep tech is — and what it is not — makes getting the timing right is a lot more likely.
It is clear that deep tech will play a big part in all of our futures, but it is not all created equal. The most successful companies over the next decade will be those who understand how technology matures over time — and know when to move to take advantage.