The analysis of data and the application of statistical techniques to solve problems has existed in the world for decades. However, only recently has the impact of data science on businesses become palpable. The rise of Big Data and other technological advancements has resulted in this explosion of interest, deep exploration of new territories and decision making with precision. As of 2019, The market size of Data for businesses stood at a whopping sum of $3.93 billion US dollars, and scientists predict that this will grow at a compound annual growth rate of 26.9% between 2020 and 2027.
Deep diving into these techniques and decision making with precision, reference is often made to “data science “and “business analytics”. These terms are often used interchangeably. Businesses today are transforming at a much faster pace, searching for easier and yet more scientific ways of retaining their market share in a mutually evolving business environment. In the words of Franklin D. Roosevelt “There are many ways of going forward, but only one way of standing still”. Therefore, we must continue to embrace new cornerstones of building the knowledge required to navigate our new reality.
Business Analytics vs Data Science - The Yin and Yang
More often than not, people believe data science and business analytics mean the same thing. A further review reveals that Business Analytics is restricted to commercial uses and problems (profit, number of employees, etc.). Leveraging on technology, it relies on analytics tools to discern insights from data that a company can leverage for its strategy. Therefore, business analytics involves making several business assumptions and incorporating macro changes into the strategy. This will require more business expertise in order to drive decision making.
Conversely, Data Science focuses on the application of statistical analysis, machine learning, data visualization, and programming to help make better decisions. It has a much broader application area and a set of problems, requiring processing of vast behavioral data from customers and understanding hidden patterns. For this, the data analyst should have a very good understanding of problem formulation and algorithms.
Data Science is a superset of Business Analytics. Due to the pandemic-caused strain, businesses have a special need to find new, data-driven ways to ensure resilience and continuity.
In the article, “Harnessing the Power of External Data'' written by By Mohammed Aaser and Doug McElhaney, McKinsey's Chief Data Officer stated that "many companies have made great strides in collecting and utilizing data from their own activities. So far, though, comparatively few have realized the full potential of linking internal data with data provided by third parties, vendors, or public data sources. Overlooking such external data is a missed opportunity”. The article goes further to highlight that “The COVID-19 crisis provides an example of just how relevant external data can be. In a few short months, consumer purchasing habits, activities, and digital behavior changed dramatically, making pre-existing consumer research, forecasts, and predictive models obsolete”, Therefore, while companies tried to understand these changing customer patterns with their internal data, they realized that relying solely on their internal data did not allow them to make smart decisions. They needed additional wealth of external data that could assist them in planning and responding at a granular level. Data science, as a concept for businesses, has evolved into a powerful anti-pandemic tool.
In the world of today, companies can use data science to reveal trends, produce insights to make better decisions, and create more innovative products and services. With these emerging needs, data driven companies such as our new subsidiary, CRC Data and Analytics Limited leverages on this superpower, Data Science. We accelerate data-driven outcomes across lines of businesses, thereby developing faster results, and subsequently assisting with making better informed decisions.
What then are the benefits to lenders?
Simplifying the Loan Approval Process
Lenders are able to simplify the loan approval process using data driven insights, to understand the credit worthiness of the intended borrower, including consumers who have thin files.
Customer Segmentation
In the traditional collection process, lending businesses segregate customers into a few risk categories and set different contact strategies for each of these customers. By using advanced data analytics tools, businesses can move to a deeper, more nuanced understanding of their customers, properly segmenting their customers based on alternative data. Examples of these may include demographics, interests, spending patterns and financial behavior collections, and risk ratings.
Therefore, adopting this methodology, allows lenders to better understand their customers, the kind of services they may likely choose, and effectively market value-adding products and services to these customers.
Detect High Risk and Delinquent Customers
Leveraging the power of data, lenders can significantly lower their risk and take quicker corrective actions to forestall possible delinquencies on loans taken. Lenders are able to draw actionable insights on customer behavior, inclusive of spending patterns from a broad range of application domains, prior to approving loans and postpaid services. They are able to discern seemingly perfect borrowers, who appear as the perfect candidate based on their past behaviors but are likely to default, from borrowers who have perfect repayment histories and have been impacted by a temporary downward turn in economic and financial status.
By applying data analytics as a subset, lending institutions can use their customer data to better understand behavior and characteristics, and maximize collection yields from delinquency prediction models. These models often use various types of data including past loans, transaction records, the number of times a borrower has not paid in full, number of times they have gone past the due date of payment etc. At the end of the loan cycle, it helps to determine renewals of credit lines and ascertain if the customer can commit to their repayment schedule.
Probability of Fraud
Credit fraud is one of the biggest concerns of financial institutions and businesses granting lines of credit. Credit fraud can be described as the unapproved use of personal data and credit worthiness, with the aim of taking a loan or enjoying a postpaid service without the intention of paying back. The most common cases of credit fraud in the world are linked to credit card fraud. According to Nilson Report December Issue no. 1209 “Issuers, merchants and acquirers of merchant and ATM transactions collectively lost USD $28.58 billion to card fraud in 2020, equal to 6.8¢ per USD$100 in purchase volume”. This is higher than 4.5¢ in 2010 and slightly lower than the highest recorded value in 2016 of 7.2¢. The report further projected that the credit transactions loss would rise to USD$49.32 billion equaling 6.23¢ per USD$100 in purchase volume for the year 2030.
Words like “unapproved use of personal data” and “without the intention” does give cause for concern and alludes to the action being an illegal or criminal act. Lenders and credit grantors, with this understanding continuously require ways to identify existing customers and/or potential customers. Taking advantage of data science, lenders can validate a customer’s identity and authenticity. Continuously mining the data to classify, cluster, and segment customers, automatically finds associations and rules in the data. These associations and rules may signify interesting patterns, including those related to fraud. It therefore remains important to use the services of companies who have built expert systems to encode expertise driving decision making.
In conclusion, while we live in the big data era, giving rise to the importance of data science and analytics in the success of organizations of all kinds, large corporations to small businesses, we must recognize there may be shortcomings prevalent in this data. These may arise from the nature of the data itself and possible inconsistencies. Over the past decade, there has been a drive towards cleaning, verifying, standardizing and enriching the data arising in what is now called “Smart Data”. Smart data enables companies to extract even more value driven outcomes enabling effective marketing strategies and real-time smart decision-making. They also support a whole wealth of knowledge to other critical operations.
References
- Data Science vs. Data Analytics — What’s the Difference? (https://www.sisense.com/blog/data-science-vs-data-analytics/)
- Global Data Quality Tools Market Report 2020-2027 - Product Data Segment Corners a 16.1% Share in 2020 (https://www.globenewswire.com/news-release/2020/07/24/2067148/28124/en/Global-Data-Quality-Tools-Market-Report-2020-2027-Product-Data-Segment-Corners-a-16-1-Share-in-2020.html)
- Data Science vs Business Analytics (https://www.educba.com/data-science-vs-business-analytics/)
- https://nilsonreport.com/upload/content_promo/NilsonReport_Issue1209.pdf
- https://www.forbes.com/sites/forbestechcouncil/2021/03/23/big-data-is-dead-long-live-smart-data/?sh=4bafc5952bc3
- https://www.mckinsey.com/business-functions/mckinsey-digital/our-insights/harnessing-the-power-of-external-data?cid=other-eml-alt-mip-mck&hdpid=ab36470a-d237-4eeb-ad67-d33fc1ed4db5&hctky=12223383&hlkid=6495eb1bcd834d94bb7e76b23c93939b