The following is a guest article from Gene Ekster, who has been involved with alternative data on both the buy and sell side. This article is the first of a series.
The rapid uptake of alternative data research by both fundamental and quantitative institutional investors is shifting the investment landscape and creating innovative sources of alpha generation. The field is still in its early phases of development, yet depending on the resources and risk tolerance of a fund, there exist multiple approaches to participate in the paradigm.
Regardless of the path taken, the process to extract benefits from alternative data can be challenging, yet with the right tools and strategy, a fund can mitigate costs while creating an enduring competitive advantage.
Alternative data research
Alternative data operations contrast with the established investment research traditions of relying mostly on information that’s derived from the workings of the financial system itself with sources typically accessed through a Bloomberg feed or an analogous system.
Alternative datasets are often less readily accessible and less structured than existing established commercial sources. Examples of such data include point of sale transactions, web site usage, obscure city hall records and other information which contain return-generating insights.
In today’s alternative data market, a fund can participate in this paradigm via a combination of:
1) Directly licensing third party datasets and growing an internal R&D practice to mine those datasets.
2) Licensing ready-to-use analysis from intermediary providers.
3) Establishing an internal data generation capability such as running a web harvesting operation or primary research.
In practice, the mixture of sources is assembled over time and often follows a logical sequence. For instance, licensing ready-made datasets gives funds the lowest cost and most rapidly accessible exposure to alternative data.
If adding an internal R&D team is sensible for the shop, it could build the team by leveraging the fund’s prior experience of consuming the ready-made option. Arguably building a team is the most expensive, but also the most comprehensive method of deriving sustainable value from alternative data.
Building an in-house alternative data capability
An increasing number of funds across the investment spectrum are building internal alternative data groups, sometimes referred to as R&D. Internal demand and a long term vision to have a data driven competitive advantage are motivating the funds to invest into building out their alternative data competencies. These groups generate value by acquiring unique datasets and developing them into internal products.
A full build-out of an R&D group is a considerable capital investment that requires up to a year of ramp up before it starts producing valuable insights. This kind of infrastructure has a different payoff cadence from other research spends such as expert networks which deliver a faster upfront return. Only certain funds are willing or able to invest into a dedicated R&D group, but doing so can provide a substantial lead in the ability to consume alternative data, thus the advantages of scale play a key role.
As with any technology focused endeavor, the team and the infrastructure can make the difference between an exercise in frustration and a humming alternative data machine. Luckily, a build-out of an R&D group can be a scalable undertaking with a minimum staff of just two or three people growing to 20+ person enterprise.
Team build-outs can be considered internal startups, and like a startup they need upfront capital to build infrastructure, talent pool and product development, then similarly, growth is best assessed by emphasizing product output metrics at first and via bottom line ROI metrics (P&L impact vs. cost) in the later stages.
Alternative data is poised to revolutionize the research process for both quantitative and fundamental investors. It can be thought of as an arms race or simply a competitive advantage; either way, once adopted by a portion of the market, the rest of the market needs to also adopt to stay competitive.
In the next article we’ll examine the role of the R&D team in creating a successful alternative data research operation.