Writing the results
All articles

Using Parsel to extract data from capital markets documents

Built with financial markets in mind

As we have mentioned previously, we initially developed Parsel as a tool for the emerging markets investment research & data platform, Tellimer. Tellimer was looking for a way to automagically pull structured data out of investment research PDF documents, so we first trained Parsel on capital markets & investment documentation.

In the few years since we conceived the project, Parsel's domain of applicability has grown substantially. We've expanded our team and are selling Parsel into entirely different verticals. Over 1,300 organizations spanning industries from industrial goods manufacturers to real estate developers use Parsel to process all manner of different document types.

As we've grown and expanded the types of documents that our algorithms can handle with ease, we remain focused on capital markets documents. And the reason is simple: an enormous quantity of valuable data and insight is trapped in PDF files across the financial services landscape. From regulatory filings for publicly traded companies to bond prospectuses for trillions of dollars worth of new issuances each year, vast reams of market-relevant data remain hidden in unstructured data stores, like PDF and image files.

In this post, I'll demonstrate how Parsel extracts valuable investment and capital markets data from PDF files.

Bond prospectuses, offering circulars, term sheets, and pricing supplements

When a company—or country—wants to raise debt to pay for its operations, it issues a bond. Investors will lend money to the bond issuer in exchange for regular interest payments, according to the terms laid out in the bond's prospectus, term sheet, and pricing supplement. These documents take the form of, you guessed it, PDF files. A $120trillion+ market is, at its core, reliant on data contained in and transmitted via unstructured PDF documents.

Either through our web application or via our API, Parsel makes it trivial to extract and store—at scale—the valuable data contained within these types of documents.

Regulatory and compliance documentation

In most jurisdictions worldwide, institutional financial market participants have regulatory and compliance obligations that they must adhere to. As a result, large compliance teams are devoted to ensuring their financial organization doesn't run afoul of regulations. One of the challenges these teams face is ensuring that their internal compliance systems stay updated with the latest rules and guidelines from regulators. These rules and guidelines, in most cases, are reported in, once again, PDF documents. Clients use Parsel to automate the ingestion of these document types so that their internal reporting systems have all the latest updates from regulators.

Due to Parsel's serverless and massively parallel architecture, Parsel can process large documents very quickly, reducing weeks' or even months' worth of analyst time to minutes.

Use Parsel to build a financial database

Each quarter, publicly traded companies report their quarterly earnings in consolidated statements that they must share with the public. These documents, which show the health and growth of the company, are published as, you guessed it, PDF documents. Parsel was originally built to extract and store exactly this type of valuable data in a database using the JSON output format that comes with all of our paid plans.

What did we miss?

As you can see, the options are truly endless. Did we miss any obvious financial market use cases here? Get in touch with us at hello@parsel.ai, and let's discuss how you can best use Parsel for your own purposes.

Sign up and get started today!