A View on the Information Age
A relatively easy task at the time, as London only had a handful of really tall buildings, which is no longer the case. More cranes meant an increase in construction and greater investment in large real-estate projects, which generally was an indicator for economic growth. So for many of us, cranes provided a physical data point of economic sentiment.
My how things have changed! Today, we have access to satellite and weather information that tells us, for example, where cargo ships are heading or how well crops are doing before the harvest. Location-based information now tracks how individuals shop; and, of course, we can follow clicks, likes, tweets and downloads revealing what people are doing and thinking pretty much all the time. We have so much data coming from so many sources, it’s mind-boggling to think about how much information is out there for the taking.
And yet, the financial services industry struggles to tame all this data and turn it into useful insights. The paradox of the information age is that we produce so much of it and yet we’re unable to consume it all. The way that asset managers have managed data in the past, i.e. reviewing financial statements or parsing through earnings calls is much like crane counting — but very different from how they’ll need do it in the future. The approach taken by most analysts simply isn’t scalable for the information age.
Part of the problem is that the information needed for decision-making – whether those decisions involve selecting investments, launching products or creating new services – isn’t as accessible as it needs to be. We’re swimming in data streams while we wait for data lakes to be built. Couple this with the sheer amount of unstructured data being created —the kind that doesn’t fit neatly in a spreadsheet — and it’s no wonder the information age has many financial institutions scrambling to keep up. We’re capturing everything, but we aren’t yet using it effectively. At least, not all of us.
New data isn’t just long (rows on a spreadsheet), it’s also wide (columns of information). A small alternative fund manager, for instance, could look at alternative data sets like news stories, Twitter feeds, satellite images and so forth. In fact, they’re probably already assessing 100-150 alternative data suppliers with 2,000-2,500 specific data sets (location-based, alternative credits, satellite, etc.). By my estimate, it could take at least eight hours just to look at one data set. To look at all of them would take a decade!
The paradox of the information age is that we produce so much data and yet we’re unable to consume it all.
As the Harvard Business Review commented1, “Despite massive investments in technology to store, analyze, report, and visualize data, employees are still spending untold hours interpreting analyses and manually reporting the results.” And while data scientists can help make sense of some of the information an institution is sitting on, building a workforce entirely made up of data scientists simply isn’t a scalable solution.
But on the horizon is something that could change everything – asset intelligence, or AI.
Today, some organizations are using advanced natural language processing platforms, powered by AI, to automatically write performance reports for mutual funds in mere seconds, something that used to take a team of analysts hours, if not days, to produce. As the aforementioned Harvard Business Review article pointed out, “The [AI-]backed reporting incorporates disparate data sources and performs real-time analysis on portfolio performance. These systems examine the facts, determine which of these facts are most notable, and output these facts as readable narrative text.” What’s even more amazing is that the system can deliver that report in whatever form makes the most sense to the end-user, be it a bulleted list of key findings or a detailed summary. And because the data system is managed and organized by AI, not a human, it’s as close to real-time analysis as we can get. This, in theory, eliminates the need to re-create reports, as the system is constantly updating.
AI, also known as cognitive computing, includes machine learning and natural language processing and has the capability to transform our businesses and operating models. In my view, AI is just at the beginning of its true potential. It will eventually bed down into enterprise solutions that – similar to Oracle databases and Excel spreadsheets in the past – will be whirring in the background without notice. Every year brings us closer and closer to a real-time solution for data management and consumption.
At State Street, we’re making our own foray into this brave new world. We’ve developed a media sentiment analytics indicator to provide asset managers with new ways to help forecast risk and performance. We’ve created novel methodologies that measure turbulence, systemic and tail risk. We have an investor behavior indicator that looks at concentrations in certain asset classes as a gauge of risk. And we’ve developed an almost real-time indicator on price levels – where web crawlers search the internet to understand pricing for common goods. All of these developments are based on large data sets, advanced technical capabilities and partnerships with leading researchers.
So, while I will always look fondly back on the days when we would count cranes in the London skyline I am excited by just how much opportunity comes with the Information Age. As our technological prowess grows, our understanding of the world will only deepen. And with knowledge, comes power.
1. Frankel, S. (2015, May 22). Data Scientists Don't Scale. Retrieved February 20, 2018, from https://hbr.org/2015/05/data-scientists-dont-scale
David Pagliaro is senior vice president and head of State Street Global Exchange in EMEA since June 2017. David has over 20 years’ experience with developing and implementing strategies for growth, driving commercial and product development and leading partnership activities on both a regional and global basis. Last year he spent three months studying archaeology in Rome.