Shaping a Big Data Strategy for Finance

Rui Wang
4 min readAug 20, 2019

--

Big data analytics will be the key to the future in the world of finance, but in a world in which data science is evolving, what are the key pillars of a good data strategy?

It’s easy to underestimate the speed and impact of the big data revolution. 90% all the data ever created came into being within the past few years, but that’s just a hint at what’s to come. According to some estimates, the amount of data coming into businesses could increase 10-fold by 2025. By that time the world could be producing 175 zettabytes of data.

Data is likely to be crucial to the future. An estimated 89% of businesses believe that if they do not adopt big data strategies within the next few years they will lose their competitive edge. If they do not, their competitors will and will gain advantages across the spectrum. It will, then, truly redefine the competitive landscape of all businesses, few more so than the financial sector. Because of this the number of companies harnessing big data has soared over the past few years from 17% in 2015 to 59% in 2018.

Financial institutions are among the most likely to benefit. Back in 2016, investments in big data within the financial sector topped $20bn making it one of the biggest customers for big data solutions.

* Personalisation: Companies can collect a huge amount of data about their customers and use it to shed light on buying habits and trends. This allows them to offer a more tailored, personalised service. It can improve customer satisfaction and increase the amount they spend.
* Insights: Real time data can provide more granular detail about business operations. It provides real time financial information, shows where the company is making money and where it is losing money. It helps executives focus their strategies and make better business decisions.
* Trading: Big data has led to a rise in algorithmic trading which harnesses vast quantities of data and identifies trends human traders might miss. Already it is being used for over 70% of total orders as technological developments and computer power reaches the scale at which algo trading becomes viable.
* Security: Data such as location intelligence can help banks to identify customer spending habits and highlight any abnormal behaviour. For example, if a customer were to suddenly start making large purchases from the other side of the world, the system would flag this as suspicious behaviour. Big data, machine learning and automation are also being used by cyber security professionals to identify security threats faster.

We’ve already seen high profile cases such as the breach of Equifax and the hack of Tesco Bank in which criminals stole £2 million from 34 bank accounts.

When a company is deemed to have fallen short of what is required, the regulators will act. Tesco Bank was fined over £16 million for what the regulator deemed was an avoidable data breach. The only saving grace was that this happened in 2016 before GDPR came into force. Had it happened in the last year, the fine would have been much higher. Serious breaches could attract fines of €20 million or 4% of annual global turnover which means penalties are now reaching unprecedented levels.

This year, British Airways was fined £183million, Marriot was fined just short of £100million while Facebook broke a record with a $5bn fine for its handling of data from Cambridge Analytica. Each of these fines could have been even higher if the full powers of the regulator had been used.

* Volume: The amount of data collected. Technological developments such as the internet of things make it possible to collect vastly more data about customers and operations than ever before. This volume is an opportunity in that it can help companies to make investment decisions or gain insights into the needs of their customers, but it can also be difficult to process.
* Variety: Data comes in two forms: structured and unstructured. Structured is clearly defined and recognisable which falls into simple categories. Unstructured data is hard to recognise and does not fall into a standard model. This includes social media posts or video content which provides a lot of insights but can be difficult to sort and manage.
* Velocity: The speed with which data is stored and analysed. A business can differentiate itself from the competition by the speed of data collation and analytics.

Using Open Source in Finance, a presentation delivered by Benjamin Tang, Quantitative Analyst, BNP Paribas at the Big Data & AI Leaders Summit.

Originally published at https://forwardleading.co.uk.

--

--