We all have at one point in our lives groaned at the tedious nature of having to manually track our expenses. These days, Personal Finance Management (PFM) tools make this easy and intuitive. With these tools, which are sometimes embedded in our digital banking channels (web or mobile), users track and categorise their digital financial transactions as an input to taking the decisions that can help them achieve their financial goals.
Let’s start by defining what we mean by categorization. The term categorization is used to describe the classification of financial transactions, labelling each transaction with a predefined transaction category. The immediate benefit of categorization is in generating actionable insights from any given set of transactional data. For example, knowing that you spend 35% of your monthly income on PayTV (DSTV & Netflix for example) might prompt customers to reassess their expenses to possibly focus on one. Similarly, bank customers may choose to tag bigger, recurring expenditure such as utilities in order to track and control their fixed monthly expenses.
Categorization of data can also be applied to classify and get an accurate glance of income. For example, breaking down salary, benefits, and other income such as part-time or freelance work. Transaction categorization can be a route to gaining deeper insights about customers’ biodata, their financial behaviors, spending patterns and preferences. Other important data points like customer attitudes to credit, how often they save & invest as well as their lifestyle choices across socio-economic classes, can also be gleaned from this rich source of alternative data.
From a financial services provider’s perspective, insights from being able to categorize bank transactions can be used to build customer personas, which is a rich raw-material needed for product personalisation, as well as product up-selling and cross-selling. For user acquisition, categorisation features ensure that products offered to customers match their learned preferences and are offered to them digitally and efficiently. For customers, this means that FSPs would increasingly, and with greater accuracy, offer financial products & services that are more tailored to their preferences, needs and wants.
Operationally, analytics on categorised data allows for data-based decisioning on an automated scale. This can very easily, for example, be used to originate credit, assess credit default risk and take an efficient decision. This is a drastic shift from the old way of doing banking with blanket offers and products that hardly considered specific customer needs.
The Nigerian economy is expected to contract from the third quarter. Analysts believe that credit will play a critical role in reinvigorating the Nigerian economy in its recovery from the twin effects of the oil price slump and COVID-19. Transaction categorization will prove key in this regard as it can empower and supplement the existing credit framework. It will enhance the building of fairer, contextual credit scoring models, which will improve financial inclusion in these times. It could also very well be the solution to bridging the information gap currently preventing FSPs from servicing the unbanked and underserved market segments. All these factors when combined will have the effect of driving economic growth.
Transaction categorisation is a technology that is catalytic for commercial objectives like increasing average products and average revenue per customer. It also serves to achieve social objectives like improving financial inclusion, which is critical in Nigeria as the Central Bank of Nigeria (CBN)is still some way off its inclusion objectives. The CBN sponsored the National Financial Inclusion Strategy was projected to reduce the percentage of adult Nigerians that are excluded from financial services from 46.3% in 2010 to 20% by 2020. The current figure is at 37%, therefore there is still some way to go.
The harsh reality is that comparing a traditional lender’s ability to build a categorisation engine to a specialized technology partner may be unfair. Engineering a machine-learning system requires experience in text processing, automation capabilities and a huge amount of data. These are just three major hurdles – and there are more. However letting a tech partner do this heavy lifting for you will mean you can bring services to market in a fraction of the time, gain access to an ever-growing set of machine-learning technologies – and offer a better PFM experience for your customers. We are happy to speak to you on how we can use our proprietary technology engine to achieve these objectives and deliver enterprise value.
KliQr has built big data computing capabilities that can process large data sets to generate actionable insights that support decision making. You can speak to us for more information on how we can support your growth ambitions.