S&P Global–IHS Markit post-merger: One giant leap for data analysis?

On Christmas Day in 2021, NASA launched the James Webb Telescope, which is the successor to the Hubble Space Telescope. The aim of the optical telescope is to see infrared light from the time of the universe’s “birth” just after the Big Bang. The project has been described as “one of the most ambitious engineering initiatives ever attempted.”

While not quite on the magnitude of space exploration and understanding, Adam Kansler, president of S&P Global Market Intelligence, likens his company’s data endeavor as one that borrows inspiration from that of the James Webb Telescope.

“It’s an enormous telescope—and we’ve built enormous telescopes and there are telescopes we’ve put in space before. But we haven’t put something quite like this in space,” he says. “What you’re now able to see and what you’re able to understand about the universe is incredible; I think we all wait each day to see what new thing we’ll discover and be able to understand. That’s been our journey of discovery over the last year as we put these two businesses together.”

Those two businesses are, of course, S&P Global and IHS Markit. Just over 13 months ago, the $44 billion deal closed. Questions permeated the data space following the announcement around what would be created by the two data companies coming together. (IHS Markit, itself, resulted from a 2016 merger valued at more than $13 billion between US-based analytics company IHS and UK-based data and trading tools provider Markit.)

Initial predications saw S&P’s well-known equities index benchmarks being combined with Markit’s fixed-income indexes and other fixed-income data; S&P’s Cusip identifiers with Markit’s Reference Entity Database; and the wealth of legacy IHS information assets with S&P’s data, analytics, and data distribution mechanisms. Not all of that has panned out. In 2021, the European Commission stipulated that S&P had to divest Cusip due to the merger and it was sold to FactSet for close to $2 billon at the end of 2021.

Kansler, who served as the president of IHS Markit’s financial services division prior to the merger, says S&P Global Market Intelligence now encompasses over 18,000 people and some 300 products. Under the umbrella of Market Intelligence is proprietary data and content, workflow solutions for managing a portfolio, credit and risk information derived from the ratings business and other historical credit information, climate risk tools, and daily pricing and valuations across asset classes.

The hope for the newly-enhanced S&P Global entity is that by combining such a massive and diverse set of data, the company will be able to build an analytics engine that can allow users to “explore” that data to yield “new discoveries”.

‘Highly complementary’

Prior to the merger, the datasets of each company weren’t necessarily duplicates of each other, says Octavio Marenzi, CEO and founder of capital markets consulting firm Opimas. “IHS Markit did not have a credit rating agency and they did have some stuff on the commodities trading side,” he says. “If you look under the umbrella of IHS Markit and S&P, there are a lot of businesses that have just carried on operating the way they had before the merger.”

Kansler echoes that. “I don’t think I can think of a single case where we said, ‘Okay, we now have two different copies of the same data,’” he says. “The nature of the businesses were highly complementary and it wasn’t that one was collecting data in one place and the other forgot to do it, or something like that. It was the nature of the customers they were serving and the extent that a dataset that one had was necessary and the other was outsourcing that, sometimes to each other or to competitors.”

One example of that is the S&P Capital IQ Pro desktop, where economic and country risk datasets were previously outsourced from a third party, but are now brought over from IHS Markit.

Kansler classifies IHS Markit data as longer historical datasets primarily used in risk management such as pricing, valuations and reference data, and datasets used directly in trading and operations of large financial institutions. The prior iteration of S&P Global had datasets more focused around research on investment banking and strategic research. “It’s a combination of those kinds of datasets that I now think offer a real opportunity with the use of technology to give particularly interesting insights,” he says.

Take, for example, the changing credit spreads of a market where there may be concerns over the creditworthiness of a particular issuer. Combining credit rating information with economic and regional forecasting, news, and information about a firm in different regions can create a fuller picture from which to gain insights—not unlike what the Hubble and Webb telescopes can do together to visualize the universe.

Kansler says about two dozen product ideas are currently in production or being built, while some have already been released. One of those is the supply chain console, which is in beta form. Users can look at a company’s entire supply chain and its various risks, including country risk for the various places they operate. It also includes product type breakdown for companies and their competitors, alternative sources, and maritime and shipping information.

Piecing together information for a more complete picture of markets isn’t new. Combining traditional data with alternative data is becoming more commonplace as alternative datasets continue to grow. Geolocation and ESG data are just some of the areas that fit into the widening bucket.

To Kansler, a traditional dataset would be a time-series dataset of prices, a specific set of numbers, or a reference dataset—something that has been pulled from static data about a particular bond or security. Alternatively, alternative data is not the fundamental time-series data, or daily produced data that is used in a workflow, but is added to it. “Alternative data could be textual data in an article; it could be a forecast produced related to an industry that is slightly outside of the traditional data that you might look at,” he says.

The technology that S&P will use to connect datasets is already in place and its development predates the IHS Markit merger. In 2018, S&P Global bought Kensho, an AI startup. At the time, S&P said its aim was to become a world leader in AI development with the acquisition focused on the talent and knowledge from Kensho’s engineers and the ability to infuse its technology stack with natural language processing (NLP), machine learning (ML), and other subsets of AI. In the period following the acquisition, Kensho’s engineers looked at use cases across S&P where its expertise could be applied to automating tasks.

In 2021, S&P launched a document viewer as part of Capital IQ Pro. The viewer uses Kensho’s proprietary named-entity recognition and disambiguation (NERD) technology, a machine-learning system that links textual data to structured data sources. In this case, Capital IQ users can link unstructured information from regulatory filings, earnings calls transcripts, press releases and the like with the structured information in the massive S&P database.

For analysts in capital markets, finding unstructured data in text or audio is just the beginning of the research process. The idea behind the document viewer is that analysts can enrich the data they find by linking it to past analyst coverage, pricing data, and news about companies or individuals associated with those companies. While having a lot of data is important, providing context around that information is key.

“AI is maybe a short-term abbreviation for the magic that figures out where and when a link should be understood,” Kansler says. “You can have a word appear in a document, but without knowing the context of what that word is, you don’t really know if that document is relevant or not.”

Cloudy future

AI isn’t the only tech on S&P’s roadmap; cloudy skies are also ahead for the data provider. In February, S&P announced a multi-year strategic collaboration agreement with AWS. Kansler says S&P also utilizes Microsoft Azure and Google Cloud, but AWS will serve as its primary cloud provider. The announcement included plans to move data applications like Capital IQ to the cloud by 2025. Currently, three out of four applications are in the cloud.

“Moving all of that to the cloud does two things: It first modernizes the technology stack, but it also allows us to accelerate the pace of innovation and development of those platforms,” he says. “When you move to the cloud, it’s not just taking the existing system and sticking it in the cloud; you actually get the opportunity to rewrite and rebuild it, in a way. So, you can modularize it a bit; you can rebuild the application in ways that allow you to fix or replace; or enhance or add in new capabilities without having to run it through the whole stack.”

Essentially, while combining S&P’s data with that of IHS Markit’s is potentially groundbreaking, you need the infrastructure and tools to provide the context around that massive stockpile of information.

Or, think of it like this: While the Hubble Telescope provided amazing images, James Webb’s infrared images of the same nebulas are dramatically clearer. Both provided data, but you can get more detailed information using better, more robust technology. So combining S&P’s data with IHS Markit’s is just one piece of the puzzle—now they need to build that telescope that provides clearer images.

The image accompanying this article is from NASA’s James Webb Space Telescope. ​It shows the Tarantula Nebula star-forming region.