Limeglass is a UK based fintech start-up that has developed proprietary software-as-a-service for ‘atomising’ or breaking down financial research into detailed, identifiable constituent parts, enabling users to locate accurate and relevant information quickly. The company was founded by Rowland Park and Simon Gregory in 2015. The co-founders have several decades’ of experience in the financial markets, working in research, business intelligence, and technology.

Limeglass has a team of employees and advisors with expertise in a range of financial and legal skills. The company received early-stage funding from many angel investors to begin developing its proprietary financial research platform. Limeglass is currently used by its largest investor, JP Morgan, as their preferred method for the analysis of financial research. Alongside this partnership, it is being trialed by some of the world’s largest financial service providers.

Limeglass was part of J.P. Morgan’s In-Residence Program, which incubates emerging technology companies to develop production-ready solutions solving for critical wholesale banking problems.

Project Scope:

One of the main challenges in financial research is that banks produce and receive thousands of pages of research every day on everything from geopolitics to company share prices. While this information is vital to effective decision making, the sheer volume of reports leads to information overload. This has led to various established practices coming under severe scrutiny, including the way research is consumed and distributed. In its current form, it is not sustainable or efficient to rely on analysts to comb through reports manually. This is exacerbated by inefficient methods for extracting insights from research reports, such as searching through an email inbox or via a basic ‘control-F’ function on text documents. These methods, reliant on finding specific keywords, are not only slow but also liable for missing vital key details.

Solution:

Limeglass uses Artificial Intelligence (AI) and rich Natural Language Processing (NLP) to enhance how research is accessed and distributed. Limeglass has a unique approach to research processes in that instead of just using taxonomy or ontology as the basis of its platform, it weaves the two together and also includes context. Consequently, it’s able to analyse documents with much greater potency than other providers.

For example, if an article is titled “Microsoft”, and the analyst is researching the company’s prices, the AI will flag a paragraph discussing shares as relevant, even if the company’s name is not explicitly stated. Similarly, if an investor was looking for information on ‘the US-China trade war’ there would be a multitude of possible combinations and synonyms from ‘non-tariff barriers’ to ‘US-Sino trade tensions’. Limeglass has been able to identify more than 40 synonyms for this specific topic, offering greater access to insights within the research for investors.

Rich NLP complements this by analysing text to identify where synonyms are being used and can draw connections between associated topics or phrases. This can then be organised into an asset-specific taxonomy, enabling users to very quickly access the right level of detail on any given topic in their entire library of research. To ensure that the financial research team can unlock the full value of their research, Limeglass’ software is specifically tailored to integrate with a user’s research library and can scan common documents like PDFs, making the adoption of the software seamless.

Impact:

For financial research creators, Limeglass enables users to customise their research consumption, by highlighting key paragraphs and providing relevant information. In turn, this leads to a better service to clients. The ability to break down a report into its parts, via AI, enables the analyst to provide not only specific information that the reader is interested in but also other paragraphs with associated information.

Limeglass enables analysts to streamline their research consumption by providing both a macro view of the topic as well as a granular level of detail on specific issues. Moreover, such personalisation means that the recipient can streamline their research process rather than having to wade through a mass of long documents to find what they want or risk missing out on key pieces of information. While a lot of information is needed this information is vital to effective decision making, the sheer volume of reports presents a barrier to utilising this research effectively. Additionally, technology is complemented with human expertise to ensure the best results. When it comes to identifying what subjects and terms are linked together, an algorithm can produce a statistical answer, but human expertise adds a layer of subtlety, making the technology as incisive as possible. Human experts are also involved in initially defining the taxonomy of tagged terms and updating them as additional market issues emerge.

Limeglass has made a positive impact on its biggest client JP Morgan Financial Research. JP Morgan is happy with the insights their research teams produce daily, leveraging Limeglass technology, which are a huge source of value to their clients.

IBSI View:

Limeglass tackles a very important problem of information overload, especially in areas of financial research. The technology allows analysts to streamline their research consumption rather than having to wade through a mass of long documents to find what they want or risk missing out on key pieces of information. The unique approach to processing research by combining context, ontology, and taxonomy ensures get relevant details within a report.

KEY FACTS

Bank: JP Morgan
Division: JP Morgan financial research
Bank Headquarters: USA
Solution partner: Limeglass