This is a guest post written by Maxime de Lisle and Emmanuel de Dinechin from Altai Consulting. Maxime and Emmanuel are co-authors of our recent publication, “Spotlight on Rural Supply: Critical factors to create successful mobile money agents”. This blog is also a brief excerpt from a workshop on the rural research, held at the Mobile Money Global Event in Cape Town earlier this month. View presentation slides from the workshop.
Despite the massive buzz around the potential of big data to radically change businesses, the reality is that very few actors within the digital financial services space have fully harnessed data-analytics technology to drive their business forward. The reason for this stems from a lack of urgent need (i.e. often their business is growing fast enough without the need to dive deeper on past data), a lack of resources (no in-house data analytics capabilities or computing power), or finally, the mistaken perception that data analytics equates to “big data” which itself equates to something necessarily arduous and complex.
However, this perception could not be further from the truth.
Uncovering data, analysing data, and transforming it in action
Mobile money providers, particularly mobile operators, are sitting on a gold mine of data. Or rather two gold mines: one from their core GSM operations (Telco CDR data , detailed coordinates of their Cell IDs, etc.) and one from their mobile money operation (KYC data for customers, agent registration forms, transactional databases, etc.). While CDR data is already being used regularly by operators to adjust their pricing model, reduce customer churn, or do well-targeted promotions, few have linked this data with their mobile money transactional databases. This is regrettable, as it is easy, fast and cheap to do. It can also yield very powerful insights to track mobile money fraud, better manage an agent network, manage float and cash, drive the marketing expenditures, etc. Data analytics can feed into most of the key business decisions a mobile money manager can make. GSMA has previously explored this space in the context of how data can create visibility around the customer journey.
What do you need big data for?
Over the past two years Altai Consulting has been working with more than 20 mobile operators in Africa. While big data is at the forefront of many CEO agendas, too often the responsibility is passed down to an IT team, often where the big data journey starts by installing new hardware and software. However, because the solution was chosen before the question was crystallized, often this hardware and software ends up not being flexible enough to the business’ requirements.
Instead, operators—especially those in the financial inclusion space—should ask themselves: “What are the key questions I need to answer? What data is required to track fraud (or recruit agents/decide which product I should focus on/spend my marketing budget wisely/etc.)?” And only once this list of questions is clearly established is it worth asking “Can my own data answer that?” The answer tends to be sometimes yes, sometimes no. But when the answer is “Yes”, the dive into the data ocean is likely to generate a find because you will know what you are looking for.
Recently, the GSMA released a case study on identifying critical factors for successful rural agents in both Mali in Chad. In addition to these findings, the case study offers more information about developing a go-to-market and geo-mapping tool to identify what geographical areas are well penetrated by mobile money services and agents, and what other areas should be targeted to extend customer reach or increase or decrease the number of active agents.
This tool is available to the industry and is aimed to be adaptable to any market seeking to develop its rural footprint more strategically. However, the approach to this tool is equally relevant to providers seeking to better leverage the potential of the data they have. The four-part approach is straightforward, and can help mobile money providers guide their rural expansion, but also provides a framework to ensure data analytics is used more effectively.
The Special Forces approach to data analytics: Focus, Speed, Flexibility & Agility
Finally, from our experience, a data analytics engagement needs to have the following characteristics to be successful:
- Focus: What is the business objective and what role does data play to achieve that objective? This needs to be clearly defined and as narrow as possible. The more narrow the scope, the lower the risk of excessive and unnecessary data analysis that is irrelevant for the business.
- Speed: Given how fast the financial inclusion industry is changing (in terms of products, customer base sizes, etc.), data analytics should be built around very short time-frames. Delivering results on data-sets several months old may prove of little value.
- Flexibility: To maintain flexibility for changing business needs, data analysis & processes, to every extent possible, should be software and hardware agnostic. With the pace of change in data-analytics technology being so swift, there are literally new solutions being launched every week which are more powerful and/or cheaper than existing ones.
- Agility: Roadblocks are likely to emerge which cannot be identified in early stages. As such, the team should be ready to re-deploy resources, to change strategy, and to display creativity
To request a copy of the Excel go-to-market tool, please email mobilemoney@gsma.com.
Image: Courtesy of Altai Consulting. Note that dummy data has been used for illustrative purpose.