Blog

Innovation overkill: Why product innovation in financial inclusion isn’t always the right move

Posted on by MSC

Amid all the renewed talk about product innovation and a client-centric (or what we at MicroSave have, for over a decade now, called a “market-led”) approach to financial inclusion, I find myself wondering if we have lost sight of two key aspects:

  1. The differences between market leaders and market followers
  2. The challenges of advancing through the product development continuum

Market leaders vs. market followers

Corporations (including financial institutions) are, by strategy, either market leaders or market followers – first or later movers. First movers do the product innovation to lead the market. They invest heavily in product development and often have an outstanding in-house capability to conduct market research and create/test innovative products or (as we will discuss below) ways of marketing and communicating existing products.

Later movers watch and copy. They sometimes struggle with the back-office systems (IT, processes etc.), but the “externals” of new products are clear as soon as they are offered in the market (even, sometimes, at the pilot-test stage). Late movers assess if the product innovation is likely to be a success in the market and respond accordingly. Of course, if they do not see the potential of a product early enough they may struggle to counter the entrenched position of first mover market leaders – as we have seen, for example in Kenya, with M-PESA.

Corporations’ product innovation strategies may vary according to geography, depending on their market position and the resources they are willing to invest in any given market. So we may see one bank leading in one market and following in another where they are relatively new entrants – or indeed copying their own across products across borders. Thus Equity Bank has largely transposed its market-leading Kenyan products into the Tanzanian, Ugandan and Rwandan markets. Once it is well established in these markets, one can reasonably expect the bank to use its capabilities to refine its existing products or develop new ones specifically for these newer markets. Market leadership and customer focus is, after all, deeply engrained in Equity Bank’s DNA.

Key takeaway: it does not make any sense at all to expect all organizations to be market leaders and undertake product innovation … indeed many are simply not geared up to be anything other than market followers.

The product development continuum

MicroSave has already posted Key Questions that Should Precede New Product Development in the context of microfinance institutions (broadly defined to encompass banks, MFIs, credit unions, etc.). The same questions apply to mobile network operators (MNOs). At the core of some of these questions is the product development continuum.

Many products suffer from poor take-up because of inadequate or inappropriate marketing and communication. Across the globe, millions of potential clients are not using financial products because they do not know of their existence, or understand adequately how they might be useful. Quality market research can shed light on this, and on many other important issues that drive product uptake – see Market Research: Beyond Product Development. These insights can enable financial institutions to deepen their reach into existing market segments or reach new ones with the same product. And of course, repositioning a product with a new marketing campaign is much easier, quicker and lower cost than developing a new one.

Quality market research can also often provide important insights into minor tweaks to the delivery process or product terms and conditions. When MicroSave started working with Equity Bank (then called Equity Building Society) on product innovation in 2001, it had 109,000 customers. The first rounds of focus group discussions yielded the important insight that customers were confused by the plethora of charges that Equity levied. Within a few days, Equity had consolidated these charges into one single price and rolled out a careful communication strategy to ensure that the entire market was aware of the changes. And thus began a journey that would see Equity Building Society transform into a bank, list on the Nairobi stock exchange and become the dominant bank in Kenya, growing to serve 8.7 million customers by 2014 (see The Market Led Revolution of Equity Bank). The initial step was product refinement – accomplished with minimal fuss, and without even a pilot-test.

The final, most complex and most time and resource consuming step on the product development continuum is new product development. And it is not for the faint-hearted. Nor, as we argued in Who Is The User In “User-Centred Design”? is it something that can be easily outsourced to outsiders with a poor understanding of the marketplace, the regulatory environment, or the internal culture and systems of the financial institution. New product development requires a structured process to manage its inherent risk throughout – as we explored in the report Systematic Product Development. Due to the complexity of the process, it is often easier to follow rather than to lead.

NBFC-MFIs as business correspondents – Who benefits? (Part-II)

Posted on by MSC

As we highlighted in NBFC-MFIs As Business Correspondents – Who Benefits? (Part-I) the benefits for NBFC-MFIs and banks are pretty clear – we summarise them below.

Benefits of the BC Model for Key Stakeholders

(from Microfinance in India – Is Business Correspondent (BC) the Way Forward?)
Benefits for NBFC-MFIs Benefits for Banks
  • The model significantly de-risks MFI operations. Operating as agents of banks almost entirely removes political risk from the equation. Banks and their agents come under the sole purview of the Reserve Bank of India, and the possibility of state government interference will be significantly minimized if not altogether eliminated – particularly when poor people’s savings are involved.
  • Banks are struggling to establish agent networks that can profitably and reliably service the low income, unbanked segments of the economy. While they have been successful in opening accounts, they have not been as successful in promoting transactions. As they expand the range of services and seek to drive credit through agent channels, banks will have to depend on MFIs that understand this segment and have built systems and processes to serve it.
  • MFIs will be able to offer a wider range of products and thus to meet the real financial needs of the clients rather than pushing credit alone and optimistically claiming that it is for entrepreneurial activities.
  • Banks can reflect the assets and liabilities in their books thus enhancing their balance sheets. Banks can increase the spread and share the risk-return of lending to the low-income sector with the MFIs in a more realistic manner.
  • Savings is a service that is universally needed by people in the low-income segment. Offering the poor a range of financial products such as savings, pension, and remittance services will create a higher degree of client satisfaction, and thus customer loyalty and reduced default. The savings history of clients will also enable better credit appraisal and will help (to some extent) address the problem of multiple borrowing.
  • Banks will partner with MFIs that have built much more cost-effective outreach channels. The operating expense for an MFI branch that can service 2,000 – 3,000 clients is in the range of Rs.5-600,000 per annum. One entry-level officer in a bank will cost as much. Banks, on account of cost considerations alone, will struggle to directly service the low-end market; tie-ups with MFIs provide tremendous opportunities.
  • The BC relationship will also open up the possibility of appointing agents in villages to offer savings and pension services. While client origination can remain with the MFIs, agents can deal with day-to-day operations and settlement can take place with the MFI on a daily basis. This should reduce the cost of operations, and the economies achieved can be passed on to clients.
  • The banking regulator will be more satisfied when it knows that banks have their skin in the game. If banks use and monitor MFIs as banking agents this will inspire confidence in the regulator that they are maintaining the requisite oversight and due diligence.
  • While some of the existing MFI-bank credit relationships are manual in nature, most BC relationships typically ride on either card- or mobile phone-based technology as the front end. This will enable better and more efficient cash management at the MFI end. Currently, 1-3 % of the total cash at the MFI is typically either in transit or stacked in vaults in the numerous branches. This can be an instantaneous process riding on technology.
  • Banks can leverage MFIs’ local knowledge and cash and risk management skills to build a manage a network of agents that will allow them to reduce the cost of many of the core operations, reduce over-crowding in their banking halls, increase sales and service to high-value customers and increase the quantity and quality of their rural loan portfolio.
  • MFIs and their agent networks can also be conduits for direct benefit transfer payments, thus earning additional commission revenue to the model. This will diversify income sources, enhance revenue for the front-line agents and the MFI BCs and introduce the MFI to new clients and opportunities to cross-sell to them.
  • MFIs can build efficient channels to offer financial services and work with multiple banks to reach the under-banked/unbanked segments. Banks will be interested in such tie-ups not only because of the regulatory pressures but also because once they begin servicing this segment, they will realize the potential that it holds.

Of course, there are disadvantages too, and (drawing heavily on MicroSave’s study of the potential for MFIs to acts as business correspondents (BCs)) we highlight these below.

Beyond the opportunities presented by the BC model, MFIs need to be cognizant of the adverse impacts that embracing the BC model may have on their current business. We consider below three main types of impacts, though the severity of each may vary depending on the type of MFI. In many cases, these threats are flip-sides of some of the benefits discussed above.

Impact on group meetings. The group articulates the methodology of most MFIs, so they need to be careful in assessing how the introduction of front-end technology and BC operations can support or disrupt the conduct of group meetings. This depends on how the BC channel is structured:

  • Field officer or group leader acts as the front-line agent. In this case, savings operations would be conducted during the group meeting. Offering saving services along with credit will result in longer group meetings. In fact, MFIs might opt to shift savings operations to a separate meeting in order to preserve the focus of the credit meeting on repayment. Increased duration of group meeting or more frequent meetings will lead to a reduction in loan officer caseload, which will affect the business turnover and profitability for NBFCs.
  • Third-party outlets acting as front-line agents. BC operations with technology enablement can help in making meetings cashless, and that can reduce meeting duration or frequency. Cashless meetings may be more attractive to business-oriented MFIs and especially NBFCs, but they may challenge the models of those with a more didactic approach to development.

Erosion in repayment discipline. MFIs also need to make sure that BC operations do not lead to loss of group liability, which is one of the core principles of group-based microcredit. This may happen if meetings are less frequent, or if the availability of individual products from banks leads people to question the usefulness of group-based products. Loan repayments using technology-enabled BC channels may also lead to a situation where clients blame non-repayment of loans on the front-line agents – or technology-related problems (“The system was down”, “I’d lost my mobile phone”, “The agent didn’t have liquidity”, “I sent money to the wrong account”, “I forgot my PIN”, etc.).

Cannibalisation of existing business. Most MFIs suspect that in the long run, banks might be potential competitors for their lending business. NBFCs and not-for-profit MFIs, which offer microcredit as their core business, suspect that banks may gain access to their clients through the BC channel and start extending credit directly to them. Or, perhaps more immediately, NBFC-MFIs may be uncomfortable with the idea of lending off-balance sheet on behalf of banks; or struggle to negotiate the right commercial arrangements with banks to do so.

The burden on institutional capacity. MFIs need to evaluate the existing management capacity–skills and bandwidth—to negotiate with the multiple stakeholders involved in BC operations. NBFCs with larger operations and used to managing relationships with funders and investors may not find it very difficult to manage the new relationships, although for regulatory reasons they will need to place a separate corporate identity and team to manage this. MFIs and NGOs with smaller operations and with little experience of managing strategic relationships may find it trying and may need to bring in dedicated people with the right caliber. Institutional capacity will also need to effect the cultural change that MFIs need to undergo in order to offer savings along with the credit.

These advantages and disadvantages will require careful research, analysis, strategic planning, and negotiation – the steps to making the transition (and indeed the initial decision on whether to do so or not) are examined in the next blog “NBFC-MFIs As Business Correspondents – What Will It Take?”

More details available in our Agent Network Management for MFIs brochure.

Financial Inclusion Initiatives in Papua New Guinea

Posted on by MSC

The Microfinance Expansion Project (MEP) in Papua New Guinea has multipronged approach towards enhancing financial inclusion in the country. In this video, Jagdeep Dahiya, MSC’s Training Specialist in PNG, outlines how MEP is trying to achieve financial inclusion by stimulating both the demand and supply for financial services.

Challenges to agency business – Evidence from Tanzania and Uganda part- II

Posted on by MSC

In the previous blog, we looked at the challenges presented by fraud, armed robbery and customer service. In this blog, we look at perhaps even more fundamental issues of agent profitability, and the related issues of customer education (or marketing!) and the product range available to drive transactions.

Making enough money to cover costs

Just under half of the Tanzanian agents are making more than $100 profit per month (see graph) … but many are making losses or very small profits. Those agents making less than $50 profit per month are inevitably going to start asking themselves whether offering digital financial services is worth the time, energy, money (for float) and effort invested.

In addition to the limited product range (discussed below), agent profitability is also compromised by two other important factors. Firstly, system downtime is still prevalent- in Tanzania and Uganda in particular. While it varies significantly across the different providers, almost all agents report experiencing system downtime, and estimate that on average each episode of downtime costs them about 10 transactions.

Secondly, but more importantly, in both Tanzania and Uganda agents are turning down transactions for want of e-float (or in some cases cash). In Uganda, agents turn away an average of 3 transactions a day for want of float, in Tanzania, they turn away 5 transactions or 14% of average volumes each and every day. Small wonder that many are struggling to achieve profitability!

Sadly, in the absence of interoperability amongst providers’ systems, non-exclusive agents have to hold separate e-float for each provider they service – thus increasing both the cost of their digital financial services and potential of them running out of e-float for any particular provider.

Time taken to teach customers about the service

Even though the markets for digital financial services is relatively mature in Uganda and Tanzania (both countries contain several of MMU’s “Sprinters”), agents are still struggling to educate customers about the service. This is probably one of the reasons why over the counter (OTC) transactions are so popular in many markets (particularly Bangladesh, Ghana and Pakistan). If providers want to avoid the OTC trap, there is clearly a little more thinking to be done on how best to support agents (and ideally the sophisticated customers using the service extensively) to explain the nature and potential of the service to existing and potential customers.

Broader range of products

But there is a more fundamental concern here, services at agents in both Tanzania and Uganda are almost universally limited to cash in/out, account opening, limited bill pay, airtime top-up and some OTC transactions. Despite the apparent lack of competition, Kenya is seeing a rapid growth in savings/ credit products (primarily through M-Shwari), and in merchant and other payments (primarily through Lipa Na Pesa).

However, one of the most striking results from the ANA surveys has been how narrow a range of products is offered through different provider channels in both Tanzania and Uganda. In both countries, very few agents are offering much beyond cash in/out, account opening and over the counter (money transfer) transactions – which of course undermines the potential of a wallet for self-initiated transactions and ultimately a cash-lite ecosystem. Fewer Tanzanian agents (see graph) offer OTC money transfers (23%) than Ugandan ones (30%), and fewer Tanzanian agents offer bill payments (5%) than Ugandan ones (17%). And while no Tanzanian agents offer airtime top-up, 17% of Ugandan agents offer this service. So even amongst the limited range of products offered by agents in the two countries, Tanzania seems to be lagging. Perhaps this will change with the recent announcement of M-Shwari for Tanzania.

But, ultimately, to reduce the agent churn that appears to be so prevalent even in the mature East Africa markets, providers will have to increase agent profitability. This is unlikely to come through enhanced commissions, which are under pressure from competitive forces already, so increasing the number of transactions that each agent processes are likely to be key. The current rates of 30-35 transactions per day in Tanzania and Uganda, and of around 45 per day in Kenya, leave too many agents struggling to make the money they need to stay in business. This is evidenced by the fact that in the Kenya ANA study only 58% of Kenyan agents said they thought they would be an agent in one year’s time.

Clearly mature markets are by no means stable markets!

Competition in Tanzania – Fact or Fable?

Posted on by MSC

Tanzania is often cited as a model competitive market in digital financial services (DFS) with the three major providers healthily competing against one another. However, the story is a bit more complicated than that, and to really understand market dynamics in Tanzania it helps to examine competition from a few angles, and on a more granular level.

In markets dominated by Mobile Network Operators (MNOs), there is a basket of methods to measure competition on the provider level, however the most common  are voice market share and DFS (mobile money/agent banking) customer market share.  The first is important, because it indicates the ability of the provider (relative to its competitors) to transition people to their DFS platform (except in countries based on Over-the-Counter Transactions), and the second is important because DFS is an industry where scale is the only path to profitability, and at scale, a network effect can potentially create autocatalytic growth.  However, here, we present a third method for measurement, which is the proportion of agents serving each provider in the field.  This is an important contribution because it provides a metric for supply-side infrastructure that indicates how accessible the service is on the ground.

Figure 1: Three Metrics of Competitive Dynamics in Tanzania*

*Sources include: Communications Commission of Kenya, Uganda Communications Commission, Tanzania Communications Regulatory Authority, Provider’s websites and press releases and GSMA MMU research. Please note that these figures are best estimates. Also please note only figures from dominate providers are noted here and therefore numbers might not add up to 100%.

**This figure is reported by GSMA MMU, which represents customers who use multiple services.  It is very likely that the majority of people in this category use Vodacom and one of the other services, meaning Vodacom’s market share might be underrepresented here.

Regarding voice market share, Tanzania is almost ideally competitive, and clearly more competitive than other East African markets, however, it is interesting that the relatively equal market shares in Tanzania do not translate to equivalent proportions of the market in the DFS space.  Instead, market leaders have clearly accentuated market shares in Tanzania, Kenya and Uganda in the DFS space relative to the voice space.  This is an indication that differences in resource allocations and levels of operational excellence make a big difference in the ability of a provider to translate success in the voice market to the DFS market.

Competition at the Agent Level:

Looking more granularly at DFS market share at the agent level, Tanzania has fiercer competition than Kenya and Uganda, meaning the market is more evenly distributed.  However, with the Agent Network Accelerator research we filtered these results by location, and the analysis is intriguing, showing that competition in Dar es Salaam is especially intense, however, outside of Dar Vodacom Tanzania is just as dominant as MTN is in Uganda (see chart below).  Further, we note that while only about 8% of Tanzanians live Dar es Salaam, about 30% of DFS agents are located there.  Therefore, not only do providers have more equal footholds there, but agent density per capita is also highest, meaning the market is more saturated than in other regions of the country.

Figure 2: Agent Network Market Share Across The Three Leading DFS Providers in Tanzania, Uganda and Kenya

*In Tanzania Dar es Salaam was surveyed (largest city). In Uganda and Kenya we used the capital cities

Catalysing Shared Agents:

The highly competitive nature of Dar es Salaam resonates, as all providers have focused aggressively on that market, and to lesser degrees in the rural/non-capital (largest) city urban areas. However, the question remains as to what impact it has had on the development of the market itself.  The most noticeable difference on the ground is the level of exclusivity of the agents, with 84% of agents in Dar being shared by providers, while the majority of agents outside of Dar remain exclusive, almost entirely to Vodacom.

This high level of shared agents was certainly facilitated by the initial competitive market shares of the providers in the voice markets, which meant that no one player had enough market power to force agents into exclusivity agreements.  However, the small geographical concentration of non-exclusive agents in Tanzania makes it clear that even if there is the potential for shared agents across the country, aggressive investment in growing an agent network by multiple providers is still a necessary condition for non-exclusivity to thrive.

Figure 3: Exclusivity of Agents By Location – Tanzania*

*Taken from Slide 19 of ‘Agent Network Accelerator Survey: Tanzania Country Report 2013’

The Profitable Result for Agents:

The ironic outcome for agents operating in the competitive, saturated market of Dar es Salaam, is higher profits per agent.  However, this needs to be unpacked a bit, because Tanzania has higher profits overall versus Kenya and Uganda partly due to very low operational expenses.  Again, when we delve into a locational analysis we see that Dar es Salaam is considerably more profitable for agents than other areas of the country. When we separate the locational variable of being in Dar es Salaam with the exclusivity attribute, it is clear that serving multiple providers is the factor that has a strong positive correlation with profit (as opposed to just being located in Dar). Therefore, we conclude it is likely the ability of agents to serve multiple providers, and therefore earn multiple revenue streams which drives higher profitability in Dar es Salaam.

Figure 4: Median Agent Profits Per Month – Tanzania, Uganda, Kenya

*In Tanzania Dar es Salaam was surveyed (largest city). In Uganda and Kenya we used the capital cities

Figure 5: Media Agent  Profits Per Month – Tanzania

Conclusion

Although Tanzania has a competitive voice market, it has only translated to heighten competition in the DFS market in Dar es Salaam. While the data provided here can only show correlations, and not causation, combining it with qualitative interviews with agents and providers, a story does emerge.  It seems that initial tight competition in the voice market enabled the development of a shared agent network in Dar es Salaam where the three market leaders all focused their efforts.  The ability of agents to serve multiple providers increased their profits, as they were then able to collect revenue from multiple providers with minimal additions to their costs.

However, this dynamic has yet to spread beyond Dar es Salaam, where Vodacom remains just as dominant as MTN in Uganda.  Currently, challenging providers have intensified their focus on other parts of the country, and given it continues, we expect the whole country to look more like Dar es Salaam with profitable shared agent networks in the near future.

To read either our reports in full, please download them using the links below:

Agent Network Accelerator Survey – Tanzania Country Report

Agent Network Accelerator Survey – Uganda Country Report

Agent Network Accelerator Survey – Kenya Country Report – COMING SOON!

Building and sustaining agent networks – Evidence from Indonesia

Posted on by MSC

Indonesia is the world’s fourth most populous democracy with a population of 238 million people. An archipelago with more than 17,500 islands, the country is rich in cultural, ethnic, religious and linguistic diversity. However, access to formal financial products/services still remains elusive for most people. According to World Bank estimates only 20% of Indonesians (above 15 years) have an account with a formal financial institution. The country has relatively lower bank branch and ATM penetration with 10 branches and 36 ATM respectively for every 100,000 adults.

The Government of Indonesia and the regulators (both Bank Indonesia – the central bank, and Otoritas Jasa Keuangan (OJK) – the financial services authority that regulates and supervises financial services activities in banking, capital markets, and non-bank financial industries) recognize this fact. Both have been proactive in efforts to extend financial access to the poor and unbanked sections of the society. Bank Indonesia, the central bank of Indonesia, has recently issued electronic money regulations and OJK is expected to release the branchless banking regulations by the end of the year.

In the light of this, MicroSave conducted a nationwide study on “Building & Sustaining Agent Networks for Branchless Banking/Digital Financial Services”.The research was commissioned by e-MITRA (USAID’s Mobile Money Implementation Unit in Indonesia) and Visa to support the rollout of Digital Financial Services (DFS) in Indonesia. The main objective of the research was to identify support requirements and expectations that agents have for agent management (such as training, support services, marketing, communications, etc.)

In order to ensure diversity among the respondent profiles/businesses, we interviewed various institutions and individuals such as: organized retail (multi-brand) outlets, government entities, wholesalers/distributors, rural banks, cooperatives, SMEs, corporates, individual mom & pop stores, airtime sellers and community leaders.

The highlight results show that the challenges and concerns of Indonesia’s agents are similar to those across the globe, suggesting that the core issues facing agent network managers are similar across many, if not most, markets. We go into further detail on these when The Helix Institute of Digital Finance undertakes the Agent Network Accelerator research in Indonesia at the end of 2014.

Likes and dislikes:

Respondents liked the facility of banking within the neighborhood offered by digital financial services (DFS), which enables the customers to conduct financial transactions within their community. Also, they appreciate the fact that they can transact at a time convenient to them. We have observed that most of the branchless banking/mobile money transactions are conducted outside the banking hours i.e. before 10 AM and after 5 PM. This is because the customers find it convenient to transact outside their work/office hours.

On the other side, the risk of fraud and cash handling figured prominently in dislikes of DFS. This is understandable because – 1. The concept is new and relatively unknown to many Indonesians; and 2. Fraud is indeed a big risk as we have seen in other deployments across the world. Research conducted by The Helix’s Agent Network Accelerator (ANA) survey shows that the risk of fraud remains the biggest agent management issue. We have discussed this extensively in our blogs – “Challenges to Agency Business – Evidence from Tanzania and Uganda (Part- I)” and “Why Rob Agents? Because That’s Where the Money Is”. This highlights the need for effective risk management practices, and more importantly of educating agents and end users about the risk/fraud management. See MicroSave’s “Fraud in Mobile Financial Services” for a comprehensive classification of risks and effective mitigation mechanisms.

Motivation to become an agent: Commissions remain the key motivator to become an agent.

Increased customer footfalls and customer loyalty also act as important motivators. Respondents feel that they could get more customer walk-ins if they become DFS agents, which could in turn lead to higher cross-selling. They also feel that association with a reputed brand (of bank/MNO) could lead to higher customer loyalty.

Interestingly, ease of payments also figures prominently as the second most important motivator. Our research shows that 99% of the transactions still take place in cash. With huge volume and value, digital payment services provide good potential for deployments in Indonesia.

“My customers can purchase goods and use the system for payment rather than bring loads of cash”- Hamid, a retailer from Kutai Timur, East Kalimantan, Indonesia.

Requirements for Support:

Agents are the first point of contact for the customers. So it is important to provide adequate support to the agent to deal with customers/customer queries. Respondents suggest that the deployment should provide: call center support for agents and customers; supervisory visits; branch office support; and on-site support for technical/operational issues. These services are essential to increase the adoption by agents and customers, especially during the initial stages of deployment.

Our experience of working with other deployments across the world reiterates the importance of having separate helplines for agents and customers. Deployments often provide remote support to frontline agents through call centers. Call centers can also be helpful in addressing queries or grievances that customers may have. They can also be effective means to market the products, introduce special offers and take feedback on services offered. Equity Bank in Kenya recently conducted a huge marketing campaign to launch its m-banking product. Their call center played an important role to measure the effectiveness of various media channels used and provide inputs for future marketing strategy. See MicroSaveBriefing Note #129 on Customer Support for E/M-Banking Users. Other successful deployments such as WING Cambodia, M-Pesa and Eko have a toll free number that customers can call to seek answers for their queries and register complaints if any.

Training Requirement:

Agent training is a powerful tool to drive agent performance. Many agents have no/little exposure to selling or offering financial services, and of course, training is an important tool for the management of risk and fraud.

Respondents in the research felt that agents need training on product, technology, cash management, customer education, and regulatory related aspects.

As digital financial services are a relatively new concept in Indonesia, deployments need to focus a lot on training and education initiatives– both for providers’ internal staff and the agents. In addition, specific emphasis needs to be placed on customer education (marketing) in order to educate both agents and customers about the benefits of the service. Regulatory aspects – specifically related to authentication, KYC, fraud detection and money laundering should also be incorporated into agent training. See MicroSave Briefing Note #135:  Training E/M-Banking Agents: What is Missing?and#138 Implementing Training for E/M-Banking Agents.

Our research suggests that mobile and internet usage among the respondents is particularly high with 61% of them using mobile for internet browsing. Evidence suggests that 48% of regular internet users in Indonesia use a mobile phone to go online – which, perhaps surprisingly, is the highest among its Southeast Asian neighbors. With 66% of the respondents using smartphone, increasing usage of smart-phones was also prominent during our research. A recent GSMA study shows that global smartphone penetration as a percentage of the population is expected to rise from 19% in 2012 to 32% in 2017. This makes a case for the deployments to develop smart-phone based applications which offer enhanced user experience for the customers and better growth prospects for the deployments. Some of the telco-led deployments already using smart-phone application based interfaces include GCash (the Philippines), Zuum (Brazil) and Dompetku (Indonesia).

The findings of the research provide critical insights to deployments in Indonesia to design, build and implement well- functioning agent networks for delivery of formal financial products/services to unbanked and under-banked Indonesians. Further deployments in Indonesia are uniquely placed with latent demand/huge target market on one side and an opportunity to learn from successes/failures of deployments in other geographies on the other side.