Reading The Signs: Using Mobile Data To Empower Customers And Address The Literacy Challenge In Financial Inclusion

About two-thirds of illiterate people in the world are women. UNESCO estimates adult literacy in Sub-Saharan Africa to be at 64%. That means more than 1 in 3 adults cannot read and write; a problem that will persist given that 48 million youth between the ages of 15 – 24 are illiterate and 22% of primary school-aged children are currently not in school. The implications of limited access to education is a burden balanced unfavourably against women and youth, not surprisingly then, CGAP has described literacy as a “hidden hurdle” in the pursuit of financial inclusion. The good news is that hurdles can be overcome, however, the solutions come at a cost and therefore need to be intelligently targeted. To be targeted, real-time and relevant data is critical … that’s where it gets complicated.

Technology and specifically mobile financial services have accelerated financial inclusion by removing a number of barriers to entry such as costs, complex processes and onerous application requirements. At JUMO simplicity and transparency are a dominant theme in our product design process; we even designed our loan access product and customer journey to provide credit education scaffolding to customers, many of whom are new to formal borrowing. Clearly though, customers who have limited or no ability to read and understand communications are at a disadvantage when it comes to accessing the full potential of even the simplest mobile financial services. Like many organisations, at JUMO we have struggled with using conventional literacy data to address this issue in the markets in which we operate.

The trouble with applying publicly available literacy data to our business for customer and commercial decision making is that it’s not purpose built … it’s “their” data, and it is not updated regularly due to the costs and effort of collecting it. So we asked the question, can we leverage the mobile data we have to predict literacy levels and optimise customise journeys and communications? It sounded like a great challenge so we started developing the capability to predict literacy levels from mobile phone and wallet usage; an exciting application of big data to real world problems!

Mobile applications, outbound IVR and “missed call marketing”, and other image and audio based tools and interventions mean that quality communication and education is actually feasible for customers that might otherwise continue to be financially excluded or disadvantaged due to low literacy levels. The ability to predict customers’ literacy levels clearly increases the effectiveness and efficiency of key decisions such as where to invest and deploy resources to better support and empower those who need it most. This is critical when trying to build healthy savings and credit cultures in previously excluded segments.

In our research we have found that customers with limited reading and writing skills often have access to technology and are highly entrepreneurial. It is important that they do not get left behind. Knowing and acknowledging these customer qualities is important because, as Professor Roger Martin put it, there is a real risk in the “obsession with the manipulation of quantities without the appreciation of qualities.” It is perhaps because of the prevailing obsession with quantities that many people fear advancements in the use of big data, believing it will erode customer privacy and increase discrimination, especially for vulnerable groups. To my mind, this is only a risk if the data is considered in isolation. It is essential to blend it with deep qualitative insights that surface the most important qualities of the problem context and customer needs. This is how we can ensure that customers’ data empowers them to own their economic future regardless of where they live and their level of literacy.

My interest in predicting literacy levels is geared at financially including and empowering more customers on the JUMO platform, however, the other practical applications of work in this area are equally exciting. For instance, Telenor Group Research have done work along the same lines and have pointed to the application of literacy prediction for Aid Agencies to better allocate resources to improve outcomes in breaking the poverty cycle. The potential applications of this data for governments has also been explored.

In the end, we need to read the signs and see the size of the opportunity to empower customers when we focus big-data efforts on a fair exchange of value and maximising customer utility. Because when we do that, the outcomes for customers can be positively life-changing.