July 31, 2018 - Paul Lashmet | Industry Solutions

Keys to a Successful Alternative Data Strategy

Asset management firms increasingly leverage alternative data to enhance their investment strategies and gain an informational edge over the rest of the market.  It is becoming the normal course of business to use new types of data in alternative ways and soon “alternative data” will simply be referred to as “data.” The ability for one firm to distinguish themselves from another depends on why they selected certain data sets over others and how they enable their subject matter experts to put alternative data to use.

In our previous post, based on the report, “Put Alternative Data to Use,” by Greenwich Associates and commissioned by Arcadia Data, we considered the types of alternative data used and desired by the buy-side community, how they measure return on investment, and how deeply they analyze it.

In this post, we expand on that research to understand the approaches used by hedge funds and asset managers to select data sources and how their people use it.  Finally, we respond to reported roadblocks by describing the keys to a successful alternative data strategy.

What Comes First, the Data or the Strategy?

Each survey participant was asked to tell us whether the data or the strategy came first.  The three responses they had to choose from were:

  1. We have a strategy first and then look for data to support, debunk, or enrich it.
  2. New types of data inspire new strategy ideas.
  3. It’s a simultaneous process:  Data informs ideas, which lead us to new data.

The results are illustrated below.  If you consider response #3 to be a be a mix of #1 and #2, asset managers and hedge funds have opposite approaches to the initial steps of selecting alternative data sources.  Asset managers predominantly have a strategy first while hedge funds lean towards both.

Asset manager strategy

This makes sense as asset managers have long-term investment strategies in place.  Their approach is to select data sources that will optimize the fundamental analysis on existing strategies to get an edge over other, less creative managers.  Hedge funds, on the other hand, take a more exploratory approach to identify unique angles that could support a new investment strategy.

Targeting Data Assets

In both approaches above, data is evaluated by testing what fits where and interpreting the results but how does the buy-side acquire the data?  To understand the targeting process we asked, “How do you find out about new alternative data sources?” for which they ranked their top two choices of these five responses:

  1. A direct search to fill a specific need.
  2. A generic search looking for something new and interesting.
  3. Recommendation from colleagues.
  4. Recommendations from other investment firms.
  5. Direct calls from provider salespeople.

Both hedge fund and asset managers predominately do a direct search to fulfill a specific need and hedge funds also rely on referrals from colleagues more so than asset managers do.

Find new alternative data

Coming in dead last for both asset managers and hedge funds is “Direct calls from provider salespeople.”  This shows that direct sales by data providers are not the most effective approach. There are two reasons for this:

  1. One provider’s data set may only be valuable to the buyer in the context of other data sets when patterns start to emerge. Without that context, it is harder for a direct sales call to articulate a relevant value proposition. 
  2. The reasons why a data set would be valuable is proprietary.

Data providers would benefit from a “try and buy” delivery platform like a data marketplace that enables buyers to plug into a variety of data types to figure out what works best for them.  In an earlier post, “Monetizing Data for the Alternative Data Economy,” I describe how and why it would work.

A Self-Service Approach

To better understand what the buy side does with the data after they find it, we asked, “How do you analyze your alternative data?” to which they had several responses to pick from:Analyzing alternative data

  1. Data visualization tools.
  2. Analytics tools provided by the data provider.
  3. Microsoft Excel.
  4. Modeling library/analytics platform.
  5. Market data desktop.
  6. Internal proprietary system.
  7. Execution Management System (EMS).
  8. Order Management System (OMS).
  9. Another team does it for me.

The results illustrated in the nearby graph show that the respondents, all of which work in the front office, take a self-service approach to analyzing alternative data.  

These points support that observation:

  • As a group, data visualization tools, which are typically self-service in nature, is the top pick, especially with asset managers.
  • Microsoft Excel, the quintessential self-service spreadsheet mastered by analysts and portfolio managers everywhere is the top pick for hedge funds.
  • Coming in dead last is, “Another team does it for me,” further proving the self-service point.

Collaboration Across Teams

So far we’ve described the initial drivers of the alternative data selection process, how data is acquired, and the self-service nature of working with it.  Below we show how alternative data is worked across the organization.

We asked the respondents, “When analyzing alternative data, do you tend to work independently or with a team?”

As stated in the report, “Two-thirds of buy-side firms using alternative data—which includes nearly 80% of hedge funds—take a team-based approach to analyzing alternative data.”

To add more color to the process, we have listed the responses to this question, “How do you share information?”

The responses show that collaboration is an integral part of the process but done manually by sharing spreadsheets and documents via email and shared drives.  

An opportunity exists to accelerate time to insight, gaining an edge on the competition by optimizing the collaboration process.

Keys to a Successful Alternative Data Strategy

We end this post with the responses to this question, “What obstacles do you have that prevent your use of alternative data?”

The responses from which to pick from were:Roadblocks to using alternative data

  • Prohibitively high fees.
  • Can’t find the right sources for what I need.
  • Internal procurement processes are too cumbersome/slow.
  • Data not compatible with data analysis systems used.
  • Management not convinced of data’s value.
  • Human capital needed for integration not available.
  • Lack of time needed to evaluate data.
  • Difficulty understanding/working with datasets that are not customized for your specific use.
  • There are currently no real obstacles.

The smallest obstacle to putting alternative data to use is human expertise.  As a group, the biggest roadblocks have to do with cost and technical implementation.  Making it easier for you and your teams to acquire, analyze, and collaborate with each other will improve the economics of your strategies.

The list below describes key aspects of a successful alternative data strategy.  Download this solution brief for an overview of the data architecture and business intelligence capabilities needed to fulfill them.

  • Let subject matter experts drive the process — not IT. Portfolio managers, quants, traders, analysts, and their teams are best equipped to derive meaning (and thus value) from new types of data. They understand the historical relevance of a topic and quickly discern nuanced disruptions that could flag an opportunity.
  • Leverage all data, big or small. The subject matter experts must have the ability to explore and evaluate any type of data (structured, unstructured, real-time, or historical). Creativity is key to spotting and implementing new opportunities, so they need to be able to play with it.
  • Enable collaboration. The efficiency and efficacy of your team’s work increases through collaboration, leveraging the collective intellect of colleagues who can expand on an idea and investors who provide direct market feedback.
  • Ensure timeliness and data integrity. The informational advantage gained using alternative data diminishes as those sources become generally understood and adopted. To accelerate an informed decision-making process and stay ahead of the game, it is critical to ensure the integrity of the underlying data and to optimize the models that use it.

Related Posts