By Anca Bogdana Rusu
For the past decade agent banking has been touted as one of the solutions to financial inclusion. The concept is the embodiment of responsible finance: with a network of agents a limited but essential range of financial services can be offered to previously unbanked populations in an economically viable way. However, questions remain about what makes a successful agent and how to expand a network in sustainable manner.
Answering these questions are fundamental for expanding digital delivery channels to benefit both the institution and the consumer; to realize the promise of responsible finance. For the institution, the agent network supports profitability and is also an essential tool for customer acquisition and relationship management. The agents become proxies of the institution: successful agents engender positive customer relationships; while poor performing agents and agent churn has a negative impact on the trust customers have, both for the institution and for the broader financial system more generally.
Through a long-term research engagement with FINCA DRC, IFC’s Applied Research and Learning team analyzed agent information, including agent characteristics and agent transactional data and built a model to identify the characteristics of successful agents.
In DRC, FINCA launched its agent network in 2011 by employing small business owners to offer FINCA DRC banking services. The agent network grew rapidly and by 2014 over 60% of FINCA’s transactions were done via the agent network. Much of the agent selection up to that point was done in a rather opportunistic manner, by reaching out to existing retail outlet – clients and signing them up as agents, which was an efficient strategy at inception stage. However, this meant that despite having rolled out over 300 agents at the time, there was little understanding of what determined agent efficiency – as defined by high number of transactions and volumes transacted. Today, 80% of FINCA transactions are through their agent channel.
To better understand what makes a successful agent we compiled existing data on FINCA’s active agent network. Data for the model included location/market characteristics, business and socio-demographic data on the agent, transactional data as well as monitoring data on the agent’s cash and e-float, the shop condition, quality of the agent’s customer interaction as well as FINCA DRC product branding displayed among others. Data availability and data quality were the main challenges in developing the agent performance model. Digitized data were required for sources usually only collected on paper, like agent application and monitoring forms.
- Results: Financial inclusion and business development goals align
Successful banking agents in DRC were identified by the following statistically significant criteria:
- geographic location
- sector of an agent’s main business
- gender of the agent
- whether profits are re-invested in the agent’s business
Women-owned agents are found, for example, to make 16 percent more profit with their agent businesses than their male counterparts; the value of their business inventory is 42 percent higher.
Importantly for financial inclusion and responsible finance, we find that transactions are higher in low income, densely populated areas with high levels of commercial development. This suggests that the agent network can be best used for supporting financial transactions among the urban poor. In addition, visible FINCA branding and effective liquidity management are strongly linked to agent activity. The results suggest that agents can be effective providers of basic financial services among the urban poor who lack suitable alternatives.
These results were implemented to improve and streamline the agent selection process, which ultimately helped to expand the network into rural areas by incorporating factors into agent surveys and roll-out strategy and driving FINCA’s ability to achieve 80% of total transactions through the agent network today.
The value of this analysis comes from the fact that it starts answering the question of what makes a successful agent and helps mitigate some of the risk for all the parties involved. Rolling out an agent is a costly endeavor for any institution and being able to able to identify which potential agents are worth the investment is highly valuable. The value comes both from a more strategic investment and a more effective customer acquisition and relationship management via the agent network.
For the prospective agent who is a small business owner, building a trust relationship with the financial provider and the additional income from the agent activity are welcome. However, becoming an agent involves an investment of time and money and is only worthwhile if the relationship is lasting.
From the consumer’s perspective, an efficiently selected and managed agent network means stability, lower churn of agents and thus a better trust relationship with formal financial alternatives.
Equally important, this endeavor highlights the significant benefits that any financial institution – but especially digital financial services providers – stands to gain from employing data analytics to make better informed business decisions.
Adapted from a case study presented in the Data Analytics and Digital Financial Services Handbook (June, 2017), this post was authored by Anca Bogdana Rusu, IFC-Mastercard Foundation Partnership for Financial Inclusion, for the Responsible Finance Forum Blog.