Event Panel Review: Data Science to Improve Decision Outcomes
A panel discussion at the Northern Trust Capital Markets Summit
Robust data analytics solutions are more accessible and cost effective than many believe
By Paul Fahey, Head of Investment Data Science Solutions at Northern Trust, Joshua Rand, Managing Director at Essentia Analytics and Greg McCall, Co-Founder at Equity Data Science (EDS)
Using data science and analytics to improve decision making has grown significantly over the past decade in many industries. But even as technology continues to improve and make data more accessible, leveraging data science and analysis can still be seen as expensive and time consuming. In this Q/A panel we discuss how data analytics are more accessible and cost effective than many believe.
Data management and analysis is a major hurdle facing asset managers today. How does EDS solve this issue?
Greg McCall: Fund managers create, consume and make decisions based on data that is, most often, poorly managed – meaning it does not sit in a centralized database, where it can be organized to be readily accessible and available for analysis.
How did your backgrounds and experiences lead you to create the EDS and Essentia Analytics solution sets?
Greg McCall: My background is like that of our asset manager clients, so I have walked many miles in their shoes. I was a fundamental portfolio manager and analyst following the technology sector for more than 25 years. During those days I was always data driven, trying to better understand where we succeeded and failed and how we could be more productive. Yet I was consistently coming up short at finding an effective team solution (beyond excel) to our challenges to drive continuous improvement in our investment process. Obstacles I came up against included things like ineffective use of our intelligence (due to siloed data/content and teams), non-existent measurement or feedback and limited collaboration tools, beyond excel.
Regarding how I leveraged my experiences in creating solutions, I often quote Andy Grove, the former CEO of Intel, who says it best: “What we have learned from decades of rapid development of information technology is that the key is relentless focus on ‘better, faster, cheaper’ – in everything,” Grove said. “The best results are achieved through the cooperative efforts of different disciplines, all aimed at the same objective.”
Equity Data Science (EDS) follows this formula. We have a strong focus on creating better, faster, less expensive solutions. In addition, EDS was launched by a multi-disciplinary group of software developers, data scientists and fundamental investors, each with different perspectives yet with a single goal of improving accessibility to investment data analytics.
Beyond being inefficient, how does the analog process that managers currently use make it difficult to fully access valuable data (i.e. market, vendor, their own research)?
Greg McCall: With an analog process, managers reduce their probability of successful outcomes, not because their investment process is flawed, but because of the difficulty in optimizing the available information and intelligence – even if we solely just look within their own walls and forget the outside world. If we further assume that we live in a more competitive world with more data and with less time-to-decision, the current analog process breaks down even faster.
The keys to successful investment management, and to all enterprises, are improving productivity, continuous innovation, competitive differentiation and maximizing intelligence. The most successful funds today have been leveraging technology (digital transformation) to become more accurate (data driven) for more than a decade. EDS makes that digital investment process easier to implement for everybody, big and small. The “equity” in EDS is about equitable access to data science for all firms, regardless of size.
How do managers enable continuous improvement? For example, how do they leverage the datasets that are generated by observing their skill and behaviors?
Greg McCall: Driving improvement in any organization requires continuous self-reflection and feedback, identifying where you succeed and where the blind spots are. EDS has made that much easier with a powerful, configurable and cost-effective investment process management platform.
The EDS mission is to provide the best possible technology to achieve three goals – building, operating and sustaining a repeatable investment process. If we do this successfully, any and every piece of information that can be useful in the decision-making investment process is optimized ̶ both before and after an investment is made.
A few examples of how EDS helps managers leverage data to drive continuous improvement:
- Prioritizing research: The most valuable asset an investor has is his or her time. EDS helps them maximize their time by automating the research and analysis process.
- Determining the accuracy of their financial models and forecasts: All investors live by the success/failure of their performance and EDS helps identify which inputs into their investment process, and which stocks, sectors, team members, and datasets, are helping them succeed and where the blind spots are.
- Lastly, fundamental investing is both qualitative (talking to people) and quantitative (numbers and math). EDS is the only platform that brings them both together in one system and across the entire investment lifecycle.
Why do managers need to evolve to survive and how can Essentia help them? What are some of the challenges to engagement and how do you discuss these with managers?
Clare Flynn Levy: At the end of the day, active fund management has been disrupted, albeit very slowly, by index funds. That’s because the average active fund manager hasn’t been outperforming index funds, net of fees. As a result, we’re seeing lots of consolidation in the industry and the number of active fund managers is decreasing.
In order to improve the odds of surviving, all active fund managers need to raise their game and take advantage of a data-driven feedback loop on the quality of their decision making. Logical as that is, every organization has some people who are very resistant to actually doing it. That’s natural but it’s a great example of how emotionally-driven decision making can hold back performance. Fortunately more and more managers are starting to realize that it’s worth pushing through the fear. Meanwhile, the next generation of managers, who are now rising up in their organizations, are very comfortable with data-driven feedback so they take to the idea naturally.
At the same time, asset allocators are noticing certain managers using behavioral analysis to differentiate in the market, which is leading them to ask other managers how they are mitigating their own behavioral biases. As a result of all of this, we expect behavioral analytics to become a standard part of the way the industry measures itself in the future.
In summary, what should every asset manager know about data science solutions and how does it factor into a whole office strategy?
Paul Fahey: Data science solves a number of critical challenges facing fund managers. First, it can help streamline and harmonize multiple linear workstreams into one dynamic platform and bring transparency to what’s often hidden in spreadsheets or inside the mind of a portfolio manager ̶ unlocking a manager’s unique dataset, their own decision making. The ability to codify a process helps distribution efforts and provides investors with the clarity and evidence they desire when making allocation decisions. It can also remove key person risks (replacing people with process) and help with succession.
Second, it can help to improve alpha generation by enhancing portfolio construction and position sizing, ensuring that not only is there an objective way to rank conviction but that such conviction matches with position sizing in the portfolio itself. Too often we find fund managers generating significant alpha through their highest conviction ideas to only see this alpha eroded due to lower conviction positions underperforming.
Finally, data science makes it easier for investors and regulators to review past decisions and perform due diligence. It also improves an investor’s ability to identify the investment skill of the manager which should lead to better outcomes over time.
Now all these challenges can be solved with variable cost fintech solutions like EDS and Essentia. Our partnerships with these firms form a key aspect of the Northern Trust Whole Office strategy, which facilitates access to new technologies and capabilities that address all our clients’ challenges, not just those focused on the back and middle office. For Northern Trust, Whole Office goes beyond data and into problem solving and decision making.