The Four Types of Startup Opportunities
In our last post, we discussed how data is the new general-purpose technology and that is why at super{set} we form data-driven companies from scratch. But new technologies are a promise, not a sudden phase change. Take electrification. When the first electric motors came out, factory owners took the easy path of simply swapping them in for the old steam engines. The factories still had the line shaft systems that worked like the leather belt-driven ceiling fans you might see at a gentrified hipster bar. This was a good system for transferring the steam power from a heavy centralized boiler, but also wasted a lot of energy running all the machines at once from a single shaft.
Over time, smaller electric motors came onto the market, which meant more efficient uses of energy but also the opportunity to swap in smaller and different types of machines on the assembly line. Industries were designed more efficiently around electrification, unleashing a wave of productivity in manufacturing. The lesson is that innovation takes time to deploy.
The venture ecosystem plays a big part in designing a new world that maximizes the productive value of general-purpose technology. We are still in the early stages of transitioning to a world where every aspect of the enterprise is reimagined around the value-add of data — just as factories in 1900 were reimagined around electrification. Data-driven company opportunities litter the ground, and we look everywhere — across industries, across contexts, across users — to find data-driven secrets we can take advantage of.
In the data ecosystem, there are sources of data and uses for data, and opportunities all along the chain to create value as a tech company. Data-driven companies derive their value to the customer from this journey from source to use — from generating, capturing, orchestrating, pipelining, analyzing, or activating data.
At super{set}, this is where we begin. Across verticals, we ask ourselves:
- What kind of data exists?
- How is the data generated?
- Where is data accessed?
- When is data generated and accessed?
- Who benefits from the existing data?
- Why is this of value to customers and/or users?
Just by asking simple questions through a common framework, we can arrive at a surface-level understanding of an industry, its problems, and where there might be a data-driven solution to those problems. We look at so many of these opportunities, applying our pattern-matching skills honed from years of experience, that we can tell from just a small amount of signal if there is an obvious solution and the kernel of a data-driven company.
Not every solution that is obvious at first will pan out into a successful company. Likewise, not every obscure solution is a dud — some may just take a little longer to nail down the right angle of opportunity. We think of new startup ideas as falling into one of four categories:
- Mac-and-Cheese — obvious (to us) and attractive, these are our layups — we eat mac-and-cheese all day! If only everything were mac-and-cheese.
- Darlings — obvious but unappealing once you scratch the surface. Writers say you should “murder your darlings” — we do the same thing purposefully with darling ideas- there’s no need to be cute or possessive when it comes to bad ideas.
- Frogs — both obscure and unappealing, sometimes a frog really is a frog and not a prince, but you still have to kiss it to find out. We are deliberate about consensually smooching our amphibian friends.
- Soufflé — obscure but ultimately attractive solutions when you dig into them — soufflé is harder to put together but ultimately just as delicious as mac-and-cheese.
Our approach applies to any entrepreneur: sometimes, the first good idea is a rotten apple that you have to discard, however painful that may be to murder your darling. At the same time, don’t immediately write off difficult ideas — that soufflé takes time to bake! That is why our priority in the earliest stage of ideation is to murder our darlings and kiss our frogs.
All of this is hard work. It isn’t enough to merely come up with a data-driven idea, it has to be thoroughly explored and expanded upon. When super{set}’s founders, Tom Chavez and Vivek Vaidya, started Krux after selling their first company to Microsoft, they sat at the kitchen table to thoroughly explore and map out the idea even before writing a single line of code or hiring a single employee. They looked deeply at the pain points and problems potential customers faced. They asked those data-driven questions to determine what the vectors were for a solution. They developed a theory of the customer and who might make a purchase decision. They looked at the competition to see how others had tackled similar problems. Building out Krux was still a long, slow, difficult challenge, but the strategic planning of the solution early on is what led to a bigger, stronger outcome when Krux sold to Salesforce in 2016.
At super{set}, we apply this deliberate approach of early formation discipline at scale. Whether obvious or obscure, we have built a process and an infrastructure to put pen to paper to ascertain where on the map of startup opportunities an idea lies. We call this process the Solution Memo — and we’ll dig into the specifics of it in our next post.
Feeling hungry? You’re not alone — learn what it takes to become a super{set} co-founder here.
