View all articles

Harnessing artificial intelligence and natural language processing to transform active portfolio management

Download a PDF of this article

It wouldn’t be controversial to say modern computing has changed active portfolio management for the better, and innovations in artificial intelligence (AI) and natural language processing (NLP) are no exception. While these new technologies promise to transform the fundamental research process in the months and years ahead, there’s a lot of hype, and perhaps some misinformation, about what these technologies do, why investors need them, and how they will impact portfolio management. The short answer is AI and NLP will empower active managers to be more efficient and effective in a market increasingly flooded by programmatic and passive fund strategies.

Buy-side analysts and fund managers have access to unlimited datasets and research content, and it seems new datasets that impact the investment process come to market daily. There’s only one problem. All this information comes in different formats, different cadences and from countless different sources.

Too much information, not enough insights

The risk of information overload and ‘drinking from a fire hose’ is all too real as there’s simply too much information to parse through efficiently to ensure all the actionable insights have been seen. In fact, in our recent 2021 Investor Survey: Research Management Trends in the Year of a Pandemic, we asked asset managers how the explosion in investment data has impacted their research process, and we got an interesting and understandable set of responses. While 72 percent reported more data has made it easier to identify alpha-generating ideas, 58 percent also said it has made it harder to separate a true signal from all the noise. So, more data doesn’t always equal better insights.

At Mackey, we’ve always been committed to providing users with a way to seamlessly aggregate, curate, and build a cohesive view of their fundamental investment research. Our systems are developed to ensure the right data sources and actionable insights are assembled into a unified view so portfolio managers and analysts can devote more of their time to their investors’ mandate — outperforming the broader market.

Looking to the future of fundamental research

The ability to deliver returns that consistently outperform will, in part, require active managers to deploy smarter tools to maintain a competitive edge. Analysts and portfolio managers will need systems that alert them to investment insights they would have otherwise missed, rather than being tasked with finding them through traditional, manual processes. The volume of available data is simply outpacing our ability to consume it, whether in bulk via a terminal model or through curated views in a research management system (RMS). So now those tools must evolve to better support the fundamental investment research process.

This is where AI and NLP come into play. But what exactly is NLP, how will it be applied, and what does that mean for the future of active portfolio management?

Unfortunately, NLP is a frequently misused buzzword. It’s become far too easy to take the promise of NLP at face value, without deeper scrutiny, as is the case with many emerging technologies.

In the simplest terms, NLP makes it possible for computers to read text, hear speech, interpret it, measure sentiment, and determine which parts are important.

Many technologies touting NLP capabilities, however, are little more than glorified Boolean operators, merely algebraic toolkits with no intelligence built in, artificial or otherwise. While flexible search and fast consumption is absolutely a vital part of the investment workflow, it is not NLP.

How true NLP will empower the active manager

When leveraged properly, NLP helps investors to see the right insights for their investment thesis at the right time. At Mackey, giving our users a way to capture, share, and act on new research insights seamlessly and in real time has always been our DNA, and investment teams on our platform expect us to stay ahead of the curve.

That’s why we recently released our NLP 2.0 engine on the Mackey platform. With this release, we took the time to deliver more than just a neat sales hook (though our sales team, no doubt, loves to walk prospects through the NLP demo). Our NLP model provides real value to customers on day one with contextual search, relevancy scores, and smart summaries. More importantly, though, that value grows over time as the model matures with client use. Ultimately, we see this technology providing each client organization with their own unique set of sentiment factors based on an aggregate view of their investment research content.

Introducing Mackey NLP 2.0

With the roll out of NLP 2.0on MackeyRMS, users on the platform are automatically recommended relevant insights within their research database. Users can also manipulate full datasets by relevance and proprietary metadata in real time across both their proprietary intellectual property in the RMS and external investment data sources.

This enables analysts to cut through the noise and sort by relevance on an unlimited number of factors, curated based on the customer’s own data. Combine that with the insider events, corporate filings, and disclosures data from our InsiderScore and inFilings platforms, and investors will have the most powerful and efficient way to see, share, and act on new insights through a single dashboard view.

NLP 2.0 also empowers busy portfolio managers to instantly retrieve the most relevant and actionable data points within their research. They can also scan NLP-generated note summaries to dramatically reduce the time they spend evaluating the impact of a lengthy document on their investment strategy and portfolio holdings. Trend analysis and sentiment factors are just a couple additional areas where NLP can and will be leveraged to provide our users maximum insights with minimum efforts.

Fundamental investors spend a great deal of time and money producing and sourcing research that gives them a competitive edge. They increasingly need their research management technology to help them see the right insights at the right time to better understand the impact on their investment strategy.

NLP 2.0 is an important update for our platform and our customers. It also marks another step in our journey to deliver powerful new data, analytics, and software capabilities that empower our clients to see more insights, save more time, and do more of what they do best — make better investment decisions.

Will Keuper

Will Keuper is responsible for product development at MackeyRMS and oversees the firm’s business across the EMEA region. He has more than a decade of experience within the financial industry and has been with MackeyRMS since 2012. Will holds a BA in Economics from Harvard University.

More of our stories from

Harnessing artificial intelligence and natural language processing to transform active portfolio management

Will Keuper, VP, product at MackeyRMS, examines the growth of AI- and NLP-powered investment technology and discusses its potential to transform active portfolio management.

Has the pandemic created a competitive disadvantage for asset managers dependent on legacy research management systems?

Andrew Robson, CEO at MackeyRMS and InsiderScore, examines how the events of the last 18 months may have widened the competitive gap between investment teams using modern technology and those still reliant on legacy systems.

Three key questions investment teams must answer before a merger

2021 looks poised to become the year of the megamerger for the global investment management industry, and nowhere is the market more active right now than within Australia’s superannuation industry. Our latest article examines three questions investment teams must answer to ensure a smooth transition.

Dispelling the myth that upgrading your research management system will be too painful

Research management systems have for too long been the application in the corner of the buy-side front office that overpromises and underdelivers. Investment management professionals understand that they need technology to help manage their investment research, but most legacy systems are antiquated, slow, and inefficient. Worse yet, these systems are often tolerated too long out because of the belief that upgrading will be too painful.

More proof the old way of managing investment research is no longer viable

Dedicated, modern and mobile-friendly research management systems are no longer a nice to have for investment management teams. They're quickly becoming a must have.


All Topics