Blog

Accelerating agent banking in Kenya – Part two

Posted on by MSC

This blog was originally posted on the CFI Accion Blog

The Helix Institute of Digital Finance recently launched the Kenya Country Report 2014 as part of their Agent Network Accelerator (ANA) project. The ANA project is aimed at increasing global understanding of how to build and manage sustainable digital financial services (DFS) networks by conducting large-scale research among DFS agents and issuing training to providers and other stakeholders. In this two-part interview, Dorieke Kuijpers, Research Project Manager at the Helix Institute and co-author of the report, provides insight into the ANA project and the Kenya Country Report. The following is part two. Part one can be found here.

One of the big findings of the survey is that banks’ agents now account for 15 per cent of the agent banking market in Kenya – a threefold increase over last year. What are some of the other key developments in the market?

We have identified a number of market developments by comparing the Kenya 2014 survey findings with those of the Kenya 2013 survey. Mobile network operators (MNOs) have led the success in the digital financial services industry in Kenya and historically have been considered better in marketing and distribution than banks, which is not surprising given that many MNOs in East Africa have more clients than banks do. Nearly a decade of development later, we see this changing: banks are now making large investments in the DFS business and they are approaching it in a very different way.

An interesting finding is that although we observe a significant increase in the market presence of bank agents, the products and services they offer are in many ways additive as opposed to competing with those of MNO agents. While MNO agents are still conducting a higher number of transactions (almost twice as many as bank agents), bank agents are offering a greater and more sophisticated array of services, including bill payments, savings, and credits. Also, the median amount transacted among bank agents is roughly 50 per cent higher, which means their revenue is now similar to that of MNO agents. This is reflected in the fact that out of the 32 per cent of agents that report wanting to open a new till for another provider, the overwhelming majority of agents would like to join a bank’s network, with Equity Bank being the most popular option.

Among bank agents, in terms of exclusivity (agents working exclusively for one provider), nearly three-fourths (73 per cent) were exclusively working for one provider in 2013, where only one-fourth (26 per cent) were exclusive in 2014. We believe this indicates that banks are using a different strategy of agent selection, favouring existing agents for recruitment. More on the key developments in the Kenyan market can be found in the here.

By and large, Kenya’s agent banking is still dominated by agents serving mobile network operators. Can you give us a breakdown of the big providers by market presence? Does this change in a meaningful way across different cities and rural areas?

Safaricom, the leading provider in Kenya by far, has a market presence of 79 per cent, followed by Equity Bank with 8 per cent, Airtel with 5 per cent, and Co-op Bank with 4 per cent. In Nairobi, we find that Safaricom’s share of market presence is less, namely, 70 per cent, while Equity and Co-op Bank are doing better for themselves in the capital with shares of market presence at 11 per cent and 8 per cent, respectively.

In summary, the banks’ share of market presence has doubled outside of Nairobi as compared to 2013, while they have expanded their networks most aggressively and rapidly in Kenya’s capital. This makes sense because banks typically use the hub and spoke model, meaning they build their DFS agent network around their existing bank branches. In the long-term, banks will want to do more to extend their agent networks to non-urban areas. Banks are generally better placed to understand lower-income customers’ needs than MNOs but will need to think about how to manage liquidity and offer operational support to agents located further from their branches.

Agents seem to be making more money by investing less time and energy. What’s responsible for these changes?

Yes, we found there to be a 15 per cent increase in non-dedicated agents (agents who also run other businesses) as compared to 2013, and that the non-dedicated agents spend a lower percentage of their time on the agency (50 per cent) as compared to 2013 (60 per cent). This points to agents diversifying their business activities and spending less time on DFS. We also see that agents are allocating less electronic float and cash, yet the number of transactions they carry out remains the same, and the profitability has increased by 10 per cent.

One of the explanations for these findings is that operating costs have decreased by 37 per cent in the last year. Because of this significant drop in expenses, agents have been able to increase their profitability without actually conducting more transactions. This may help explain why agents are more optimistic – the percentage of agents predicting they would continue as an agent increased from 58 per cent in 2013 to 80 per cent in 2014.

What changes to the agent banking industry do you envision in the years going forward?

We have observed many differences in how banks and MNOs approach the DFS industry as well as the type of products and services they offer. More recently, however, we also see convergence, and this is a trend we expect to continue.

Earlier this year, KCB and Safaricom launched the KCB M-PESA account, while Equity is partnering with Airtel to launch Equitel, a thin SIM that offers voice, data and money transfer services. We believe that these kinds of collaborations will become more common and that they will lead to greater operability through the sharing of agents and switches. With more players in the market that have a significant market presence in Kenya, we also expect to see increased competition and new market dynamics.

We, of course, hope that the recent collaborations between banks and MNOs will result in exciting innovations on the DFS front, including a range of new products and services. The Helix Institute looks forward to continuing following the developments in the Kenya market closely and training DFS providers in the country in accelerating their agent networks and designing new, customer-centric products.

 

PMJDY Assessment Round 2: Well Begun is Job Half Done

Posted on by MSC

The Hon’ble Prime Minister of India launched ambitious Pradhan Mantri Jan Dhan Yojana (PMJDY) on August 28, 2014 with an objective to ensure at least one active bank account per family. MicroSave and the Bill & Melinda Gates Foundation designed a survey to track the progress of Pradhan Mantri Jan Dhan Yojana (PMJDY) with focus on the presence and performance on Bank Mitrs (BMs). Till now two rounds have been conducted. Situation with regards to availability of BMs on ground has definitely improved, however for how long they will remain active in absence of attractive remuneration and support from banks/BCNMs, is doubtful. We recommend steps such as development of standard training module for BMs, ensuring monitoring and capacity building support from banks/BCNMs, ensuring minimum commission of Rs. 5,000/month and roll out of additional products through BM network.

The “I don’t have enough float” quandary!

Posted on by MSC

Ugali is cornmeal porridge and a staple of the Kenyan diet; it is as Kenyan as M-PESA. Last time my mother was making it, she ran out of cooking gas and texted me frantically to send her money so she could buy gas and finish cooking dinner before it got soggy. I ran to the M-PESA agent near my office, but they did not have a float, and then ran frantically from agent to agent trying to buy float, until the ugali had long become non-edible. If you ask any Kenyan they can tell you their own harrowing tale (some much less funny) about trying to conduct a transaction with an agent that was unwilling to and recent research like the Intermedia Kenya Wave 1 FII Tracker Survey on the customer experience with digital finance agents and The Helix Institute’s Agent Network Accelerator Survey – Kenya Country Report (2013) highlight this as among the top three issues hindering transactions.

Agents without float are perennially frustrating for providers who are constantly making new partnerships and products to bring the amount of time needed and cost incurred in rebalancing down. The Helix Institute research from across East Africa shows that agents report that rebalancing is fairly quick and easy, so the question remains, what is prohibiting agents from carrying more optimal levels of float? Given my personal frustration and the mounting evidence of this as a systematic issue, I set out to interview 50 agents in Kenya as to what was causing this problem.

Some of the answers are quite surprising and necessitate using a behavioral science lens to better understand how agents are thinking about profit maximization with agents around them, over time, and by the customer. In fact, in some instances, the agent actually has a float and is refusing to do the transaction for other reasons. Understanding these mental frameworks will help providers to better address the quality of customer service their agent networks are providing.

Agent transactional revenue honing

Some M-PESA agents choose to target the maximum revenues per transaction instead of the more prudent strategy of maximizing profits over time. The amount of revenue earned is based on the value of the transaction conducted. Values are arranged in tiers so that all values in that tier yield the same amount of revenue. Hence, agents only agree to conduct deposit transactions closest to the tier’s lower limits, informing a customer that that is the maximum amount of e-float they have to transact. For instance, an agent speaking about the value tier from Ksh.3,501 to an upper limit of Ksh.5,000 explained:

“….to me, it doesn’t matter, whether the customer wants to deposit Ksh.3,501, Ksh.4,000 or Ksh.5,000. I will only deposit Ksh.3,550 because it is the least amount and which will earn me the same Ksh. 14 commission, as that which I would have earned; if I were to do the Ksh.5,000 deposit … so then I have to save some float for more transactions later….”

Most customers are then left to abide by the amounts quoted by the agents and not fully accessing deposit services that address their needs. Depending on the customer needs, some opt to search for other agents while others end up splitting their deposit transactions through multiple agents. Providers need to better communicate to agents the effects of focusing on revenue maximization per transaction as customers are inconvenienced by the resulting limitations on access to e-float, and makes them prone to completely losing their customers in the long run.

Agent default stickiness

There are other agents who opt not to increase the float they hold despite the growth in the demand for digital finance transactions over time. While it may seem simple to decide to invest more in a business that is noticeably growing, the truth is it is not. The proprietor must weigh increasing float vs. all the other investment opportunities they have in the other products in their store. Even if they did decide float was a prudent investment, how much more should they buy and how might that affect the risk of getting defrauded? These fundamental decisions can be complex enough to drive agents to just stick with their default amount – the original amount required by the provider to start their agency business. In which case, agents report rationing the float they have for their most loyal customers. Meaning they will reserve float for regular customers, and deny transactions even when they have the float to make them.

To address this, providers or master agents may push such agents to grow by automatically increasing the default. There could be percentage growth mandates for agents with clearly defined timelines for instance 20% growth every year, which would aid in determining the progressive defaults over time. Agents also need to be more knowledgeable about risk mitigation and risk reporting to reduce their concerns about facing risk or fraud.

Agent transaction pooling

While some agents opt to maximize revenue by the transaction, others try to limit volatility in their earnings by forming groups with other agents and pooling their risk. Group members help each other by lending each other small amounts of money to make change for customers, referring customers to each other, and where all the group members are under one master agent, the agents take turns going to the bank branch for rebalancing.

The group support also extends to slow days, when specific agents are not doing many transactions. They will communicate it to the group, and other members, even when they have a float, will turn away customers, directing them to the agent needing more transactions that day. This helps ensure that revenues are more reliable on a daily basis for everyone in the group. It also makes sense, since it will mean agents will be much more likely to have in sync needs for rebalancing, which they benefit from doing together as well. Providers must recognize this need for predictable revenue streams, and the gains agents experience from pooling the work of rebalancing, and provide superior solutions for them, or risk having this system persist.

Addressing agents’ mental frameworks

The above are just initial insights, and so while these trends were clearly observed, we are still unclear of the prevalence of them on a national scale. These practices indicate a clear disconnect between the mental frameworks of providers who are trying to facilitate rebalancing and agents who have a float and are still denying transactions. Probably the scariest realization here is providers cannot see these problems on their virtual dashboards, because the agents actually have a float, they are just not always willing to use it, and therefore customer service will continue to suffer. These issues highlight the importance of regular agent monitoring and research techniques like mystery shopping that will give better data on the prevalence of these issues, and cost-effective solutions to address them.

Helix FI2020 Webinar – Digital Financial Services: Opportunities for MFIs

Posted on by MSC

Well, there are four basic paradigms MFIs can use to effectively go digital:

  • Buy a platform and become a provider.
  • Act as an agent / agent network manager for an e-money issuer (usually a bank or an MNO).
  • Leverage an established digital financial system to deliver its own products and services.
  • Limit its engagement to leveraging telecom networks for information sharing.

The Helix Institute hosted a webinar exploring the opportunities abound for microfinance institutions in digital finance during the FI2020 Week event. This event was part of a week of global conversation exploring the most important steps to achieving full financial inclusion which will include partners from around the world participating in order to advance financial inclusion.

Click the download button to read the report in full.

 

Embracing A Market-led Approach To Developing Product Concepts

Posted on by MSC

According to the World Bank’s Global Financial Inclusion Database, more than 2.5 billion adults do not have an account at a financial institution. Among the many factors that are responsible for this high level of financial exclusion, poor design of financial products is key.

Why, then, is it so difficult to design financial services for the poor?

Unlike populations who receive predictable incomes into their bank accounts, poor people’s financial flows are far more complicated. There are multiple contextual elements – scarcity of income, lack of buffers (emergency funds), inadequacy of information, inaccessibility to financial services, and uncertain cash flows. Equally difficult to understand is the poor’s choice of financial services such as apparently extortionate Bombay 5-6s or in-kind savings, or village level ROSCAs. Then how does one approach financial product development for the poor?

MicroSave’s guidestar – market-led solutions for financial services – has helped us develop product concepts that are now used by millions of poor people across the globe. In this blog post, we reveal the secrets of how this mantra of a “market-led approach” actually works.

Secret 1: Significance of understanding the “market” 

A key differentiator of our approach is the premise that the market involves both financial service providers and their customers. A product will succeed only when a provider’s strategic vision and institutional capacity are aligned with the product ideas emerging from market research. Although this may sound obvious and basic, this fact differentiates success from failure. A provider may conceive a fantastic, ‘out of the box’ product idea, but whether it will be developed and/or implemented successfully depends on the provider’s intent and capabilities as well as business environment.  CGAPs’ Insight into Action, which documents their experiences in using Human Centred Design to develop product concepts, reports that many product ideas could not actually be implemented either because of lack of buy-in from departments of providers – Bancomer in Mexico, or for other pressing needs of provider – MTN Uganda.  The No-Frills-Accounts offered by multiple banks in India are a similar story. The Government focussed on addressing only one challenge faced by the financially excluded population – an extremely complicated account opening process. They overlooked many aspirations of these people from a bank account, that is – quicker services at branches, access to information on banking services, respect associated with a bank account, etc. The product design also ignored the capacities of the banks and the aspect of motivation of the staff who were to offer this product.

Of course, the starting point for a product concept is the basic understanding of existing behaviour. While talking to customers of a leading bank in Kenya on a savings product concept, a key insight was that a ‘lock-in feature for a fixed period’ was actually desirable and could encourage their discipline to save.

Understanding market insights does not always lead to ‘successful’ product design unless service providers strategically align themselves to the product concept and are thus ready, willing and capable to deliver the product. In Nepal, many practitioners embraced digital financial products and created agent networks. Agents began to offer banking products but there were not many adopters. The providers had assumed that customers wanted to use formal financial services and that agents would solve the problem of inaccessibility to these services. This was a correct supposition, but there was more to it. People did indeed want access to formal financial services, but not at the cost of convenience. Opening an account required at least four visits to agent outlets – as a result, very few people transitioned from village level cooperatives and most agents remained unused. In this case, the service providers did not have the capabilities to develop and support a network of agents robust enough to service targeted customers. (See also Lessons from emerging markets, Why robust agent network is crucial, How agent networks fuel M-PESA’s success).

While developing product concepts with a dozen banks in east and southern Africa (2001-2006), MicroSave documented the importance of strategic planning and the need to ensure buy-in from  all departments.  Strategic planning helps to manage the growth process and build systems so that a product concept does not face any unnecessary challenges. For example when Teba Bank upgraded their banking system, they found that customisation took longer than anticipated. Worse still, as banking systems fill to capacity, they slow down – each transaction takes longer. Ensuring buy-in from all departments helps to avoid problems with legal, regulatory, audit and IT systems; as well as to reduce the risk of “product orphans” by ensuring that the departments that must roll the product out are involved from the beginning. (More lessons from MicroSave’s Action Research Programme – 2002, 2003, and 2006).

Secret 2: Participatory creation and refinement of concepts through testing  

Understanding the market gives researchers multiple ideas for product concepts. While some of these ideas evolve in the field, some are discovered in participative concept distillation workshops.

In the example from Kenya above, we knew that we needed to build in a contextual factor that would curb the temptation to spend. We took the market insight to the concept distillation workshop. We encouraged bank staff to think about how the bank could induce self-disciplinary saving behaviour amongst its clients. In the light of this understanding of the market, various ideas generated in the workshop suggested a savings product that committed the customer to a goal, a specific amount of regular savings to achieve it, and a lock-in period to avoid premature encashment. However, this was just the starting point. We needed to understand what clients actually thought about it.

In a market-led product development process, understanding customer need is not a one-off activity. It is a continuous process where product concepts are developed, tested and modified through a series of iterations. The objective and scale of each iteration may vary. While the first phase essentially focuses on understanding the behaviour of existing and prospective customers, in the later phases we evaluate whether the product concept will be able to bring about the behaviour change that the service provider expects from the new/modified product. This process helps optimise the product’s value proposition for users. The result of this exercise is a product concept that works –  M-PESA provides an excellent example of this. The initial idea (developed without market research) was wrong – for both the clients as well as the microfinance institution. However, the process of concept-testing made way for the re-invention of the product – right down to the processes and marketing/communication around it. Testing continues throughout the journey of product development. It starts at a conceptual level and continues until the development of systems and processes.

In Kenya, we tested the idea of commitment savings at various levels (customers and staff) and stages of development – from on-paper concept to soft launch. During all the testing, we focussed on assessing clients’ attitudes, perceptions and behaviour towards the adoption of these products, identifying the key risks involved, and testing the processes, technology and operational activities. The result of this exercise led to the development of a successful savings product that we now know as Jijenge product offered by Equity Bank.

Secret 3: Focus on 8Ps

Product design, pricing, people, process, place, physical evidence, promotion, positioning – collectively the “8Ps of marketing” is all about focusing on the details and can be used in various contexts – 8Ps for DFS, 8Ps for savings, 8Ps for marketing. At any stage of product concept development, attention to the 8Ps keeps product design experts focused on all of the elements they need to elucidate and develop. If these details are not well thought through, recorded, and shared with the team, there are high chances that many crucial aspects might be missed or overlooked while a product is tested. In our journey to develop the refined version of Jijenge in 2012, formulation of 8Ps indicated  that the product would also be launched on the digital Eazy 247 platform. The product development team knew this and subsequently modified the Eazy247 platform to incorporate Jijenge. The market research also provided insights into the optimal channel and messaging for the promotion of the product. Adoption of the 8Ps-based approach ensures that product development team members will also initiate activities to market/communicate the product.  The 8Ps framework implies that all aspects of the product are examined in an integrated manner and everybody in the team is on the same page.

MicroSave has developed more than 200 products for a wide range of banking, digital financial service provider, MSME, microfinance and third party aggregator clients for nearly 20 years. We can confidently say that there are no magic formulae for designing financial products for poor people. Success lies in following a careful and systematic product development process. This process starts with market insights through rigorous market research (remember market = customers and providers); a participative approach to concept development and iterative testing of the concepts to final rollout; and attention to all the 8Ps.

 

The Bricks and Mortar of Agent Networks: Training and Support in India

Posted on by MSC

While the first blog in this series explained how Indian Government mandates have determined the current character of the digital finance services (DFS) market in India, this blog is more forward looking,  focusing on addressing issues of agent training and support.  The 2015 ANA India data shows that these are areas where providers are struggling, and are therefore also opportunities for creating competitive advantages as new players enter the market.

The government has recently licensed payment banks, which will hopefully bring more aggressive competition to the DFS space in India, driving usage for customers and profitability for agents.  However, payment banks must ensure they build their offerings on a solid foundation and that means focusing on the bricks and mortar of the agent network from the beginning.  This blog provides advice for where these payment banks and existing providers should focus, so they can ensure a solid foundation for scaling their roll-out in the future.

Agent Training

Support metrics are among the lowest in India compared with other ANA research countries. For instance, only 59% of agents received training in India compared to 92% in Kenya and 68% in Bangladesh (Figure 1). Training is important so that agents can provide a consistent, high quality customer experience, and in certain cases also help with sales of accounts and new products, as explained in this video interview “Why Is Agent Training So Important?”

The Helix Institute in partnership with the Harvard Business School conducted econometric analysis, and found that agents who are more knowledgeable about mobile money policies experience a significantly higher transaction demand than their less knowledgeable peers. Knowledgeable agents are a product of careful selection, followed-up by high quality, targeted, and repetitive training.  The study also found that these well informed agents experience an even greater increase in demand when there are competing agents nearby. The findings suggest that expertise is a dimension upon which mobile money agents compete and indicate that service quality is critical to a healthy agency.

Monitoring and Support Visits

Regardless of how good the training is, issues will still arise in the field and providers need to support agents through regular visits and a call centre.  Support visits are important to ensure agents are well informed, can provide a high quality of service, and help build agent loyalty to the brand. Support visits can also be used to collect important business intelligence from the field.

However, only slightly more than half of Indian agents (58%) reported receiving regular support visits –again lower than other ANA countries like Kenya (86%) and Bangladesh (69%), as demonstrated in Figure 2. This means that providers are blind to many aspects of how their agents appear and operate, as well as what competitors are doing in the area.  Extending regular visits to a higher proportion of agents will give providers higher quality of service, happier agents, and the ability to adapt strategy faster than their competition.

Support can also come in the form of immediate assistance via a call centre. At present, the evidence indicates low awareness and usage of this facility—only 59% of agents are aware that there is a call center, and those that do know, only call it a median of two times a month.  Again, these metrics are significantly lower than other countries.  Agents further report that, “dealing with customer service when something goes wrong” is the biggest challenge they face as an agent. The first step will be to increase agent awareness of this support option, and the second will be to encourage them to call in, by giving them immediate answers to the problems they face.  This is also an opportunity for collecting business intelligence, which will help improve the service offering, given the system developed is sophisticated enough to detect trends in the issues agents are having.

Marketing and Communications

On the customer side, data from the Indian Government (PMJDY) and Intermedia (2014)  shows that up to half of the bank accounts opened by Indian agents are not active, meaning that the extension of access that the government drove has still not translated into efficient usage of these systems. Further, InterMedia (2014) also shows only 0.03% of adults are using a mobile wallet, and only 13% are aware of mobile money.  This is consistent with the agent perspective as they rank the lack of customer awareness around DFS products as the numberone barrier to conducting more transactions.  A little over half of agents (52%) also don’t think their providers’ marketing has helped generate awareness of DFS products among consumers.

Research conducted by MicroSave indicated that marketing support from providers was often limited and only amounted to giving agents some marketing collateral. Above the line (ATL) and below the line (BTL) marketing is expensive, but unless providers invest in it, customers will not use the systems and agents will be left without the business they need to drive profits for their business.  Scaling agent networks must be sequenced to carefully balance new customers with new agents, so a sustainable ratio between the two is kept, as explained in this HelixInstitute video.

Agents are the bricks and mortar of a DFS business. With payment banks designing their strategies for building their agent networks, they must also ensure they get the operations right of putting high quality agents in the market, supporting them effectively, and playing their part in seeding the demand they need to earn a decent revenue. Providers who focus on these factors will be able to create a competitive advantage with those currently in operation, and help create a better paradigm for moving the country forward into digital finance.