• wblogo
  • wblogo
  • wblogo

Using AI Tools To Make Venture Capital Investing Faster, More Efficient

Tom Burroughes, Group Editor , May 3, 2021


Venture capitalists have been heavily involved in backing AI and related technologies that are changing the face of business, but perhaps they ought to use these ideas in their own firms more. We speak to a San Francisco-based VC shop which claims to be doing precisely that.

Crunching the numbers
Data is making a big difference, Makkawi said. In the case of the Private Alpha Fund, the algorithmically-driven system scores more than 500 from 8,000 companies and ends up investing in 30 to 40 of the top scoring firms. (In the case of the Private Tech30 Fund, it had a target of 30 to 40 ventures; it ultimately invested in a diversified portfolio of 38 ventures.)

The Private Alpha Fund, which was launched last June and is already invested in 13 firms, is even now showing double-digit returns on invested capital. Back-testing of the EQUIAM Systematic Ranking , which it employs, shows that this fund beat top-quartile VC funds as a category with less volatility and less correlation to public markets. The Private Alpha Fund charges a 2 per cent annual management fee and 20 per cent on carried interest - a fairly standard fee. The fund has a target size of $100 million. 

A big challenge in conventional VC investing is a wide dispersion of performance between the top and bottom quartile funds. Based on an average for the whole sector, the risk-adjusted return results for 75 per cent of funds aren’t very attractive, he said. 

One reason why it is now much easier for VC investors to shift in and out of their investments is the vigorous secondaries market. At present, EQUIAM reckons that the secondary market is worth up to $70 billion in the US alone. It also predicts that the venture capital direct secondaries market is expected to grow by more than 8x over the next 10 years.

What end of the field does EQUIAM play in? Makkawi said the business looks at late-stage and growth companies, such as Series B+ companies with market values greater than $250 million, as they are typically well established, with both revenues and addressable markets.  

“In addition, our algorithms only select companies with sufficient high-grade data quality. In practice, this means companies that have already completed three funding rounds (seed, A and B) and therefore have generated the requisite data. Our process eliminates companies that don’t meet our data-quality criteria,” he said.

Latest Comment and Analysis

Latest News