Disruption vs. Enablement: New Products

“Creativity is just connecting things. When you ask creative people how they did something, they feel a little guilty because they didn’t really do it, they just saw something. It seemed obvious to them after a while.”

- Steve Jobs

This is part 5 of a 7 part series. If you haven’t done so already, you can read the first four posts here, here, here and here.

The last post in this series that falls under the “disruption” category is in many ways the most difficult to qualify as ”new products” is a vague denomination which I’ll do my best to classify.

In some industries, it may become apparent that a startup’s best chance for success is to compete with incumbents in a market by creating their own product due to a newfound advantage such that the resulting market dynamic that prevents established players from competing successfully. This market dynamic is most often either a shift in consumer preferences/behavior or technological progress that for the first time makes the previously impossible, possible.

One way to think about opportunities in which the potential exists for new products is when path dependence creates an opportunity that previously wasn’t possible or available. More often than not, everything has to break right to set up these massive opportunities to create valuable, long-lasting businesses.

This is the best place for a startup to sit because they’re creating a wedge where the value isn’t subject to incumbent competition or a Bonferroni correction. This point was made famous by Clay Christensen in his seminal book The Innovator’s Dilemma and if a startup has a moat associated with their model like network effects, economies of scale, or brand then profits can be abundant.

To use one of my favorite examples from history, consider the auto industry in the US and Japan in the mid 1900’s. Early on, Henry Ford and other American car manufacturers dominated the industry due to mass production resulting from assembly line techniques established by Ford that churned out a large number of analog automobiles. This approach worked but because there were so few models (SKU’s), cars were a high-priced, status symbol for wealthy people.

On the other side of the world in Japan, real estate costs for plants were high and therefore the companies were space constrained, forcing them to do smaller batch manufacturing. Workers also all spoke the same language (unlike in the immigrant heavy US working class) and the fact that there wasn’t enough space to accommodate task specialization meant that they would need to be trained to be able to handle all parts of the process. While Japanese manufacturers could create fewer finished cars (inventory), it could release newer, improved, and differentiated products which global customers preferred compared to the standardized US models. The blind spot in the US automobile industry is only apparent in retrospect and took several factors initially independent of one another converging in such a manner that was just right to change the course of an entire industry.

Dominoes are an apt metaphor for certain factors breaking just the right way to create new products. The most prevalent example today is artificial intelligence. Today’s advances in AI are the result of the falling cost of data storage, coupling with digitization such that data is being produced and stored in areas it never had been before, along with an increase in cheaper computing power as predicted by Moore’s Law. These two factors led to 2012’s favorite term “big data”. Only once these three forces converged were the AI techniques first discussed dating as far back as the 1950’s usable in many different areas. The cherry on top is GPU performance. GPU’s developed by companies like NVIDIA were used primarily for gaming applications, however it turns out that GPU’s are incredibly efficient in powering deep learning applications, a fact that wasn’t widely known until just a few years ago (which has had quite the effect on NVIDIA’s stock price).

This domino effect has led to a Cambrian explosion of startups and innovation making previously impossible products like Amazon’s Alexa platform, autonomous vehicles, and intelligent robotics, while deep learning can even be applied to areas like biology . These second order effects are having third and fourth order effects as well such as:

• Bioinformatics. The Human Genome Project was a 13-year effort that began in 1990 to sequence all the base pairs in a human genome to create a map of human DNA. This massive new data set has led to a newfound understanding of how the genome affects the likelihood of developing certain diseases, how we react to different foods, our physical characteristics, amongst many others. However, combining this dataset with the advancements in artificial intelligence mentioned above is where the real magic is happening. We’re seeing incredible progress in the speed, efficiency and cost of processes critical to our health like drug discovery, disease testing, monitoring, and even gene editing.
• Smart Sensors. The incredibly fast growth of the smartphone market has led to companies all over the world competing to deliver the best device with the fastest processing power, best functionality, etc. This has led to very powerful, yet cheap sensors that can be embedded into any device, not just a smartphone. When these sensors are combined with AI; advancements like drones, autonomous vehicles, virtual reality, smart medical devices, satellites, and the Internet of Things become possible.
• Bitcoin and the Blockchain. Unlike previous examples, the blockchain came about because of a convergence of a real-world event a technological advance. The 2008 financial crises led us to the brink of the world financial system collapsing beyond repair. It was then that someone with the name (or pseudonym) Satoshi Nakamoto released a whitepaper detailing his innovation, creating a peer-to-peer electronic cash system that was completely decentralized and free from outside influences that had plagued monetary systems around the world. This real-world macro event may have been the catalyst for creating this new decentralized monetary system but like AI, it was only possible then due to the falling costs of computing power and bandwidth which made it profitable for miners (those that verify transactions on the network) to earn bitcoin for lending these cheap resources to the network.

Another area that’s less technical than the examples mentioned above is e-commerce. Successful startups such as Warby Parker, Casper, and Dollar Shave Club, amongst many others seemingly came out of nowhere to becoming household brand names.

The Warby Parker story is a famous one — a University of Pennsylvania student breaks his glasses and can’t afford to pony up $400 for a new pair so he starts to dig into the industry and finds that a large conglomerate, Luxottica, has a monopoly and therefore sets prices outrageously high to benefit from large profit margins. By building a vertically integrated brand, the Warby Parker founders were able to sell high quality eyeglasses for much more affordable prices. What this story often leaves out is the path dependence that made such industry disruption possible.

A decade ago, starting and scaling a company like Warby Parker would’ve been quite difficult prior to the path paved for them by the Internet’s behemoths. Google and Facebook have a billion-plus users, all of which are generating large amounts of data on their personal preferences (also made possible by the falling cost of storage). This allows companies to test and then target very specific populations of users that are most likely to be interested in their product without having to give a large percentage of their profit margin to a retailer with high foot traffic. By being able to test and then target very specific audiences most likely to convert to purchase while being able to create a brand identity via creative advertising, companies like Warby Parker can sell their product at scale from day 1 without being limited to expensive prime brick and mortar retail space.

The future is just one of many possible outcomes and new product opportunities are found most often when founders are paying close attention to developments in several different areas that once converged, provide a blueprint to create a disruptive, new product.

While it’s just about impossible to create a specific checklist for this post, what history tells us is that founders who create disruptive new products harness one or often more than one development, whether these are technological in nature or not, in order to create a new product entirely in an existing category or a even create a new category.

In order to fend off fast moving competitors and incumbents, and the massive amounts of money that will inevitably come flooding into the space, the most important thing for founders to think about when building companies that introduce a new product is their moat, or defensibility.

Outside of network effects and data network effects, some things to think about here are:

• Understanding previous barriers to entry in the category
• Why the barriers to entry have recently changed or been lifted
• If pricing power exists
• And if a startup can provide the brand, product or service with such a strong value proposition that incumbents can’t compete and other startups can’t easily unseat them

You wouldn’t for example, want to create a better, millennial-focused airline due to the industry’s low barriers to entry, high fixed costs, and relative lacking opportunity to differentiate in a way that can’t be copied. However, creating a supersonic jet and partnering with industry incumbents like Boom is doing — well now that’s a different story entirely that will be interesting to watch play out.