There is a worldwide movement towards the use of Big Data[1] across many industries, including financial services. Many financial institutions in the South now have economists reporting to the Board, and a select few, including Equity Bank and Kenya Commercial Bank, have Chief Information Officers. Data warehousing is a common terminology, even if evident in few institutions; Fintech and technology-enabled services are current buzz phrases.

But, and this is a big BUT …

Despite this, it is remarkable how little use is being made of ‘small data’ ― the high level information that is directly available on a financial institution’s customer base, from the banking system itself.  Let me give some examples.

A few years ago, a financial institution in East Africa had installed ATMs and was puzzled at why the machines were not being widely used. MicroSave asked the institution to compile basic data on usage and customer base. The data showed that of the institution’s customers, the vast majority were using the institution to repay loans, and, in fact, less than 10 per cent of the customer base was transacting regularly on their current or savings accounts or, indeed, maintained a balance to enable them to withdraw regularly. At the time, the institution wanted to move towards a large-scale advertising campaign, and our advice was simply to give cards to customers who maintained funds and/or transacted on their accounts, initially; and to follow up with targeted marketing campaigns at a later stage.

Whilst, in this case, it took some time to gather the data, it took just a few minutes to detect and then validate the patterns being shown in the data.

Shortly thereafter, another financial institution was preparing a strategy to significantly expand its services and was considering its delivery channel strategy, using a combination of branch and alternative channels. However, MicroSave’s data analysis showed that the institution had significant levels of dormancy across its network, but that it had a small, but very important, customer segment, which transacted regularly and maintained value in the bank. The initial strategy was focused on a mass market approach; however, data quickly showed that refining the value proposition for the small, minority of customers would probably provide quicker returns.

In a third example, a commercial bank was considering the case for major changes in its product strategy. It called in MicroSave for advice. Again we asked for data from the banking system. Data was prepared on product usage, dormancy levels, channel usage, and customer balances. The data showed that, whilst the institution was large and valued, it had growing dormancy, and some of its products and services were struggling to gain acceptance. The data revealed a need for more in-depth understanding: this led directly to targeted customer based research, which further demonstrated the need for product refinement. Core deposit liability products were refined and re-launched. Partly as a result of these strategic changes, within two years, the commercial bank was reporting record profitability and growth.

There are a number of key lessons in the examples above:

1. Simple data and trend analysis across a small range of indicators is sufficient to identify where to look, but analysing ‘small data’ is of itself, often, not enough. In the last case decisions were taken as a result of ‘small data’ to launch a more in-depth study, and it was this more in-depth study which led to the subsequent decisions on products and product delivery, which were to double profitability. However, it was the small data that helped to focus the research and take key decisions.

2. Many financial institutions are not using their own data to inform their strategic decisions. This remarkable observation means that even some large commercial banks are planning in a relative vacuum. With insufficient data, senior management teams in these institutions typically plan on the basis of their own perceptions about the institution and its client base, rather than an objective reality.  This approach can result in confusion, misdirection and frequent changes of course – as perceptions change. 

3. Significant expenditure can often be avoided; or better targeted, simply by trying to understand an institution and its market before taking key strategic decisions. Periodic system-based segmentation analysis should be conducted by every financial institution.

[1] Big Data is a term for data sets that are so large or complex that traditional data processing applications are inadequate. Challenges include analysis, capture, data curation, search, sharingstorage, transfer,  visualizationquerying and information privacy. The term often refers simply to the use of predictive analytics or certain other advanced methods to extract value from data, and seldom to a particular size of data set. Source: Wikipedia.

Leave Comments

*   can't be blank

*

can't be blank

<a href="" title=""> <abbr title=""> <acronym title=""> <b> <blockquote cite=""> <cite> <code> <del datetime=""> <em> <i> <q cite=""> <strike> <strong>