Most investors claim to follow the data, which is good investing practice and even better marketing. Besides, no really successful investor is ever going to claim blind luck or gut instinct as their secret sauce.
But letting the data drive oneâs actual investment decision-making is a lot more difficult in practice than it is in theory. After all, thereâs a lot of data out there, Sailesh Ramakrishnan, a partner at early-stage global venture capital firm Rocketship Ventures told Karen Webster in a recent conversation.
He said the world is awash with data all day, every day â from mobile apps, social media, ratings sites of all sorts, etc. â a stream that generates a constantly shifting sea of information for any investment firm.
But Ramakrishnan said that information breaks down into three distinct types. âThere’s a bunch of day-to-day, changing data that come in, things like newspaper articles, employment history, new executives joining, funding announcements and so on,â he told Webster.
On the other end of the spectrum is the static, mostly historical data about a company. And in between is the slow-changing data â quarterly performance results and the like.
âSo there’s a whole continuum of data, and not only do you need different techniques to extract information, you also need different ways of combining these different streams to get an entire image,â Ramakrishnan noted.
And that is what Rocketshipâs algorithm-based investment model was constructed to do. It sets multiple models keyed into different time slices against the startup ecosystem and gears the firmâs investments toward early-stage firms in their earliest investment rounds (generally the seed, A and B rounds).
The model aims to accomplish the same goal of every early-stage investor: to get in on the ground floor with the next amazing company and disburse the funds entrepreneurs need.
Following The Data To The Unexpected
Rocketshipâs models are varied â some compute data every day, some every few weeks and others every few months.
Ramakrishnan said none of these models are perfect, because perfect models donât exist. But theyâre designed to learn and improve over time, filtering data into better guidance and recommendations as to where the firm should be looking to invest.
That doesnât mean the model gets to make decisions on its own. Ramakrishnan said one of the most important realities of working with mathematical modeling is that it has its limits. Reality is full of intangibles that matter very much to a company’s success, but theyâre hard to present mathematically.
âThat is why we have not invested in every company that our algorithms identify,â Ramakrishnan explained. âWe as human partners spend a lot of time trying to understand that âsecret sauceâ that exists within the company, and whether that is a sustainable, resilient element.â
But it does mean that when the data point in a certain direction, the firm knows thatâs the place to start looking â even if itâs not what Rocketship expected to see.
A World Of Opportunities
That was the firmâs experience almost immediately upon launching its first fund five years ago. The plan was to do what nearly every Silicon Valley investment firm was doing at the time.
Rocketship intended to start local with all the opportunities in the Valley, then down the road push out into the country at large and eventually the wider world. But when the firm actually started running its algorithmic models, Rocketship quickly found that its plan was, in a word, wrong.
What the data told the company was that its own backyard was the wrong place to play. The broader world was full of amazing companies without much regard to borders â in India, the European Union or Latin America.
Ramakrishnan said Rocketship was founded by career data scientists, all operating under one golden rule: âNever impose oneâs strategy in conflict with what the data is saying.â
âData offered us these kinds of global opportunities and we followed,â he said. âWe became a global investor pretty much on day one, and were immediately very different from what most other investors were doing.â
Thriving During The Pandemic
Ramakrishnan pointed out that the world of investing is changing all around us, but in ways that play to Rocketshipâs strengths.
In a world where a pandemic has shut down face-to-face meetings, everyone on Earth suddenly has to learn something that Ramakrishnan said his firm has spent the past half-decade working on: investing in firms whose founders youâve never met in-person.
And he added that the investment landscape is still lively in an awful lot of places. For example, firms that enable cloud-shift, FinTechs that enable lending, firms specializing in employee management and neobanks/digital banks are all areas where opportunity is exploding in response to recently skyrocketing demand.
Democratizing Venture Capital
Perhaps even more interestingly, Ramakrishnan said, is that the investing landscape itself is beginning to change as it becomes more globalized and democratized. The balance of power is shifting in ways he believes will ultimately benefit the best, most innovative companies worldwide, without regard to where they were founded.
Ramakrishnan said the next amazing startup might come from Silicon Valley, but it could just as well come from Vietnam, Nigeria, Chile or Colombia. And those firms will come to market better able to build a track record of results without raising capital â which means by the time theyâre talking to potential investors, âthe dynamic has changed,â he noted.
The money will always be extremely important, but the data-driven investing world of the future is about more than that, he said.
âEverybody’s asking investors, âWhat can you do for me?ââ Ramakrishnan said. â[But] itâs not just about if we have the dollars â it’s because of our backgrounds, our data science, our data.â
âWe now have to have those reasons why you should take our money from us versus anybody else who’s offering money to you,â he said. âAnd I think that dynamic â where there is that recognition of the value investors play over and beyond just the dollars â is [an] essential part of this conversation.â