The PDF problem
It has been a long-lasting challenge for financial services professionals to efficiently analyse long PDF documents.
Whether the document in question is a company filing, an annual report, sustainability report, analyst research or even a company’s product marketing collateral, professionals from auditors and accountants, to quantitative analysts, data science teams and business teams all need to absorb more information, more quickly and more accurately than ever before. The sources they must analyse span an array of different file formats that constitute ‘unstructured data’, of which PDFs are just one example.
PDFs are currently analysed in an analogue way: by having financial services professionals read them in full, making notes not just about what they found, but also where they found it, providing citations to prove their assertions, as well as to help satisfy regulatory compliance.
If we consider the number of different PDFs which must be analysed, this process has to be exhausting in order to be exhaustive. This eats into the working days of many roles, meaning people handling data spend more time gathering, referencing and structuring information, leaving less time for interpretation, extrapolation and creation – i.e. the more valuable and often ‘fun’ parts of their jobs.
But the days of inefficiently ‘wrangling’ the information held in PDFs may now be over for financial services professionals, for good.
We invite you to test-drive Orbit’s free PDF query tool – chat.orbitfin.ai
Orbit Financial Technology helps financial services organisations master unstructured data, allowing them to unlock the information held in hundreds of millions of disparate, untapped sources, from news articles and social media posts to handwritten PDFs and company exchange filings, earnings call transcripts and even presentations.
Our core platform, Orbit DataStudio, uses AI to turn any conceivable format or volume of unstructured data into a wealth of accurate market intelligence, providing more information, more quickly, to guide better financial decisions. Orbit DataStudio currently powers three SaaS products, developed for specific client use cases in the Asset Management, Quant Hedge Fund and Banking spaces: Orbit ESG, Orbit China Data and Orbit Insight.
We are currently in the process of integrating Chat GPT functionality into our platform to create a powerful new user interface, which will allow our clients to ‘chat’ with immense, verified databases of unstructured data, helping them get to the answers they need more quickly and with more certainty, as the sources of data will always be offered along with the insights we provide.
To trial this approach, we decided to help solve the PDF problem described above.
We have created a free-to-use online PDF query tool, allowing people to take our new capability for a test-drive and save time interrogating PDFs while they are doing it.
Our free ‘Drag & Drop’ feature is simple to use and can start generating insights from any PDF document in record time, right now.
Using Orbit’s online tool integrated with GPT-3.5-Turbo API, you’re now able to run multiple queries across any PDF document, providing users the ability to trial the functionality instantly and convert automatic searches.
By early May 2023, this technology will be fully integrated into Orbit’s platform, enabling users to seamlessly generate research and niche data sets across millions of company documents, including annual reports, exchange filings, ESG disclosures, news, press releases and investor documentation, across any language into English.
Here are a few use cases:
Whether you have questions from mainstream ESG frameworks or your bespoke ones, drag and drop the PDF into the tool, then directly ‘ask the document’ in plain English. Our PDF query tool will give you the answers you seek along with page numbers for you to check. You can query subjective questions or even numbers from tables or graphs.
In the example below we looked at American Airlines’ ESG Report and asked a direct question about the company’s Scope 1, 2 and 3 emissions figures, followed by a question about reported accidents.
For Investment Research
Efficiently analyse exchange filings, earning call transcripts, investment research and any other documents. Upload your document and then ask the tool anything in English, e.g. management discussions, latest financial performances, or major corporate actions.
Below, we looked at BP’s 4Q 2022 results call transcripts and asked a direct question about KPIs, but then something trickier, asking the tool what the key themes discussed in the meeting were.
For Private Equity
Retrieve key financial indicators on investee companies from reports that are all in different layouts. The tool allows the automation of your data consolidation without limitations on formats and in addition this process requires zero manual labelling.
Below, we uploaded the Rockefeller Foundation’s financial statements from 2021, and asked the tool a very specific question about obligations under the company’s defined benefit plans.
Now, over to you…
Please take our PDF query tool for a test-drive today and see how quickly this technology can help you discover meaning in unstructured data – we hope it will give you back some valuable time in your busy day.
Finally, imagine how powerful this capability will be when we have integrated it into the rest of Orbit’s platform. This is coming soon, so please watch this space.
You can follow us on LinkedIn or on Twitter at @orbitfin.
To learn more about this functionality and how Orbit Financial Technology can help you master unstructured data, please get in touch here.