Natural Language Processing boosting investor engagement

-9 September, 2021
natural language processing in investor relations

Estimated reading time - 4 minutes

In the context of rapid process digitalisation, IROs now have access to simple, understandable and synthetic data which is geared towards informed decision making. 

But how can technology help you leverage your data to support your investor relations strategy?

At Praexo, we apply tools such as Natural Language Processing (NLP) and data analytics on investor feedback, comments and meeting notes collected. Our purpose is to help IRO’s and C-suite management become more efficient and insightful during their investor engagement. This article intends to demystify the complexity of such data processing techniques and show how these can be applied to the financial industry and more specifically IR.

What is NLP and how can it be implemented?

NLP is a key component of Artificial Intelligence (AI) that automatically exploits text and speech. Relying on machine learning, it also analyses patterns in data to help improve the understanding of text or speech. This boils down to text classification, specific information extraction and text generation.

As part of text classification, sentiment analysis examines the positive, neutral, or negative polarity of speech or written text. Many powerful models and services already exist on the market making it easy to implement sentiment analysis.

On the other hand, topic identification is a more challenging technique which requires a deeper and more granular level of analysis to determine precise segmentation. Finally, one of the most recent breakthroughs that has been happening in NLP is text generation, which for example is used for automatic translations, chatbots and summarisation.

Companies encounter many challenges when implementing Natural Language Processing techniques especially due to:

  • Unstructured nature of data
  • Difficulty of information extraction that must be adapted to the industry
  • Lack of available datasets to train domain specific models

Three main implementation strategies already exist to help companies integrate NLP:

  • Cloud technology services providing firms with a quick and initially low-cost solution, however this is usually not trained on specific models for distinctive sectors
  • In-house development requiring more effort and resources in the acquisition of new talents and data collection but brings models trained on relevant industry notes
  • Providers delivering powerful NLP tools designed for defined industries, that do not require preliminary data collection or many resources

How can Natural Language Processing be applied to IR?

As financial markets transform due to technological innovations and new regulations, investors are more than ever becoming involved in companies’ strategic decisions. MiFID II has particularly impacted change in the industry by disrupting the role and involvement of brokers and intermediaries. The IRO’s role is thus facing a growing need for more transparent and qualitative information. 

Adding on, IR must manage an expanding volume of information and topics while coping with lesser involvement and decreasing quality from traditional intermediaries, such as brokers. Companies now need to be more independent in dealing with these matters for the sake of governance purposes and in order to contain potential activism threats. For more information on how technology and our solution can help improve governance, see our article digitalisation and ESG standards.

“86% of financial services adopters predicted that AI will be very or critically important to their business success in the next two years” – “Making the investment decision process more naturally intelligent”, March 2021, Deloitte1.

In this regard, we foresee that NLP plays and will have a greater role in facilitating real-time investor relations analysis and detecting evolutions in investor’s opinions. More specifically, it helps capture and synthesise information, and anticipate issues quickly in an unbiased manner where every bit of information is examined precisely no matter the quantity.

As companies seek to digitally transform their operations, NLP will undoubtedly play a larger role by bringing:

  • Autonomy: synthesise data to provide a clear overview of opinions providing actionable insights
  • Efficiency: quickly centralise and analyse investor feedback 
  • Consistency: structure lengthy amounts of data in themes and identify trends
  • Accuracy: process feedback in an unbiased and accurate manner

Today, companies do not have to develop their own in-house tool. Different options are available with the only real question remaining is whether or not these models are optimised for their needs. 

At Praexo, we believe the solution is to focus on digitisation and management of investor data using AI, and more specifically NLP, to boost engagement with investors. Indeed, our advanced algorithms trained specifically for the financial sector help to automate collection and analysis of your investor feedback process. Using topic classification and sentiment analysis, they can track their own evolution of key information, anticipate anomalies and spot trends & expectations among investors. 

Ultimately, NLP combined with data analytics allows IROs to target and better qualify their investors by identifying opinion outliers.

To discover more about the potential of NLP in your investor relations, you can access the replay of our webinar: leverage data to boost investor engagement.

If you wish to discover more on this topic:

1Making the investment decision process more naturally intelligent, 2nd of march 2021, Partick Henry & Dilip Krishna

2Natural Language process investors future is here, 1st Feb 2019, Oliver Schutzmann

3 The changing role of the Investor relations Officer, 20th May, 2021, Dennis Carey, Ram Charan & Bill McNabb