“Software eats the world.” That is the mantra that many tech entrepreneurs wear on their sleeves as they try to build disruptive products to shake up traditional industry processes.
Now these industry-eating entrepreneurs are turning their lens on venture capital, a business centered around human relationships, not procedural efficiency. I attended a mini-conference around the topic this week. The conferenced featured groups like Crunchbase (over 150,000 companies in their database and over 5,000 companies using their API) and AngelList (over 87,000 companies in their database) who each presented their own vision for the future.
It is exciting to see a group of wide-eyed entrepreneurs building new products for an industry that has been an analog industry for many decades.
As with analyzing the impact of technology on any new area, it is difficult to decipher between signals versus noise for what will actually happen. What exactly does Big Data applied to startups mean for venture capital? Too much of the discussion and media coverage to-date have presented this new trend as a zero-sum game — for Big Data to win, venture capitalists must lose in some ways. That, to me, is a false choice.
Instead of “software eats venture capital”, I would suggest that it is more accurate to say that software is nibbling at the edges of the industry. Technology is “disrupting” venture capital the same way it is “disrupting” classrooms, not by replacing the people who engage in investing or teaching, but by making them much more effective and organized.
In order to understand what Big Data analytics and software will do to the venture capital industry, I think it’s important that we dissect exactly where products can impact the venture investing process.
We need some framework to think about this, and here is mine. Like the technology or hardware-software stack, there is an analogous venture investment “stack” with several distinct areas where tech entrepreneurs can hope to add value (and gain paying customers).
Note that this diagram is illustrative. There are many more companies working in the space today and surely many more in the future.
1. Identifying — startups such as Mattermark are tracking website views, funding, employee and hiring, board of directors, social media activity, and creating a single score across the various metrics. Admittedly, this is better for consumer facing startups (though you can get a free trial and decide for yourself).
2. Qualifying + Screening — the line gets blurred a bit between the process of discovering a company and screening them for further due diligence as most of the startups addressing venture capital are in both. Resources such as Mattermark or CrunchBase and others fall into this category. They contain some basic information that investors care about in the initial screening such as capital invested, valuation and size, team and board, etc.
3. Evaluating — the traditional due diligence process is here to stay. Meeting the management team and digging into the details of a company’s financial, business model, etc. is at the core of venture capital. The process, however, can be greatly enhanced by data platforms that track the performance of a company’s product. App Annie and Flurry are two such companies that can act as non-biased sources to confirm a company’s actual engagement and traction (for mobile only).
4. Investing + Processing — AngelList is the big “disruptor” by allowing angel investors to set up their own syndicates, like a micro-VC fund. There are other software tools that help make the paperwork of funding an investment easier, better documented and more organized.
5. Managing + Reporting — coming to the bottom of the venture capital stack and the least sexy (just like hardware) part, software tools are making it easier for investors to track portfolio companies, organize company data and metrics and report their portfolio to limited partners or other constituents.
My framework for thinking about the future is asking the question:
“10 years, 25 years or 100 years from now, will this industry look like it does today?”
In the case of venture capital, having a set of tools to help with each level of the venture investing stack would simply be table stakes in the future, the same way that Bloomberg terminals have become standard and expected in every Wall Street firm.
Ultimately, there will be a number of companies that holds, analysis and tracks venture and startup data and can be connected and communicate with each other. For example, startups funded on AngelList will directly feed that data to Crunchbase and Mattermark or mobile download data from Flurry will be integrated directly with DataFox and become a part of their tracking system.
The most interesting of these will be the top half of the stack where tech entrepreneurs are trying to use data analytics to find the next Facebook or Twitter early on. The Big Data meets venture capital movement has opened a floodgate of new ideas for how to track and use data. At the conference dinner, one entrepreneur was talking about using satellite image of company parking lots to track employee activity and growth!
Of course, investing is all about data, but not all of which fits neatly in a relational database. There is also “data” that is a founder’s energy and ambition, the dedication of other board members, the changing taste of users, all of which can be stronger signals than anything that can be put into a spreadsheet. Big Data is a tool not a panacea.
Like economics, depending on data analytics to predict an entrepreneur’s actions is a dismal science. Observations and data over the long term can form guidelines but not exact rules. Economics is bad at predicting black swans in the same way that data analytics will likely be bad at predicting startup unicorns, but the jury on this next generation of startup data platforms is still out and likely will be for another few years.
Ultimately, Big Data and software will help the real venture investing algorithms known as human beings. These tools will allow investors more time and energy to focus on what they do best — understanding companies and helping entrepreneurs achieve their greater potential.
As one of my engineer friends is fond of saying;
“We shouldn’t let the progress of machine learning be an excuse for the lack of human learning.”
