Session 8: Combining Qualitative & Quantitative methods

October 31, 2012 9:00 AM EDT to November 1, 2012 1:00 PM EDT

The debate is often about whether quantitative or qualitative tools are better for sourcing client insights. This starting point for this session is that a basket of tools, including both quantitative and qualitative, will help to make better decisions and gain deeper insights about clients’ needs, behaviors and preferences. 

This conference has ended. Please feel free to peruse the commentary for each of the 10 sessions.

Presenters: 
Claudia McKay
Financial Sector Specialist, CGAP

Claudia McKay works with the Technolgoy and Business Model Innovation Team, focusing on research areas related to the large-scale adoption and usage of branchless banking by low-income, unbanked people as well as the business case for agents. Prior to working at CGAP, she spent seven years working for Opportunity International, a global network of microfinance organizations providing more than 1 million of the world’s poor with financial services. She worked as the director of product development for the global network, developing innovative financial products (such as agriculture microfinance, housing loans and index-based weather insurance) that meet the diverse needs of Opportunity’s clients. She also spent four years working as the head of microfinance banking at Opportunity Bank in Malawi, a commercial microfinance bank.

Akhand Tiwari
Analyst, MicroSave

Akhand Jyoti Tiwari is an Analyst in MicroSave’s Research Practice Group. He has over six years of quantitative and qualitative research experience in providing financial services to low income households. Akhand has worked on research concept, research planning and design, data collection and research analysis.

Mukul Kumar Singh
Senior Analyst, MicroSave

Mukul Kumar Singh is Senior Analyst at MicroSave’s research practice. For last two years in this role, Mukul has led and contributed to various research projects in Asia and Africa. Listening to the clients has been an important objective of all these researches that range in diversity from assessment of agriculture value chains in Africa to issues of financial access in India. Mukul has had the experience of using both qualitative and quantitative methods during these different research contexts.

Comments

Submitted by Sandhya Suresh on

No one method will work in totality when it comes to getting valuable information from the clients.The literacy levels of the clients,the time they could spare for sharing the information w.r.t their employment patterns,ability to comprehend and understand the context,urban and rural geographies etc plays an instrumental role while deciding the methods for research.Some of the PRA tools are the best when the clients are not quite vocal and hesitate to speak up.Much still depends on the facilitator who can make the process quite interesting.In the cities however where space and time is a constraint one to one quick surveys (which has to be conducted frequently though) could be effective.

I am glad these conference is bringing lot of valuable insights putting the research in the forefront for all of us in the research community.
Sandhya Suresh
ESAF,India.

ESAF,India

Submitted by Akhand on

Dear all welcome to session 8!
In the past one and a half days, we have discussed the use of qualitative and quantitative research tools at length. Often, quantitative and qualitative methods are seen to be completely different. Quantitative methods, as we have discussed, include tools such as surveys and deal objectively with numbers and statistics. Qualitative methods such as focus groups and individual interviews deal subjectively with intangibles like words and emotions.
There are situations where one can clearly identify whether quantitative or qualitative methods are needed. For example, quantitative tools would certainly be needed if you are trying to quantify a certain problem or understand how prevalent it is by projecting results to a larger population. On the other hand, qualitative tools are needed in an exploratory phase of research when you are not quite sure what is needed or how to approach the problem.
However, are quantitative and qualitative tools always distinct or can they be combined to yield even more powerful insights into customers?
As the best advertisements engage both the left and right hand side of the brain, so the most powerful research can combine quantitative and qualitative methods. Objective data can provide a structure to analyze subjective qualitative data and measure how prevalent a certain view is in the population. Qualitative methods can be used to dig deeper into certain issues that arose through qualitative methods like a focus group.

What do you think? Are quantitative and qualitative tools always distinct or can they be combined to yield even more powerful insights into customers?

Akhand Tiwari

Submitted by Mukul K Singh on

Dear Sandhya,

Welcome to the session 8 where we discuss about combining qualitative and qualitative methods. As you rightly said, each has its own positives and negatives. Does it also mean that combining the two will lead to more effective outcome?

MicroSave

Submitted by Diana Lewin on

In my experience, combining quantitative and qualitative research results yields to much more powerful insights. The quantitative research gives an idea on some of the phenomena going on, and then the qualitative tools allows to understand why it is happening.

MicroSave

Submitted by Aakash mehrotra on

The point you make looks interesting, but isn’t it very complicated to combine two different approaches. Why cannot the quantitative research be used to elicit whys? Quantitative methods also have tools that explain whys? Factor analysis is one such example! In fact tools like preference ranking can well be adopted into questionnaire schedules.

Sambodhi Research and Communications

Submitted by Mr.Kumar on

I think both methodologies complement each other, and a synergy is drawn by limiting objective of quantitative and qualitative methodology. Quantitative methodology can answer "WHY?" but qualitative method suits more to elicit a response from customers to understand their perspective and to gain in depth understanding.

Consultant

Submitted by Jagdeep Dahiya on

I second the views of Sandhya and Diana, that combining Qualitative and Quantitative tools indeed offers better results and more powerful insights for taking informed decisions.

If one have resources, one would go for both the techniques, but then we usually have to settle for either of the two in most cases. What we need to understand are the considerations for deciding on the right combination of the two techniques.

MicroSave

Submitted by PRAKASH JHA on

Is it question of which approach, qualitative alone or quantitative alone or a combination, is better. Or it is a question of what solves my research question.

IKYA HUMAN CAPITAL SOLUTIONS

Submitted by Tanaya Kilara on

Prakash, I think you have hit the nail on the head! Very often we talk about techniques but the driver really is the research question. Starting with the research question allows us to pick the best method to answer the question, regardless of whether it is qualitative or quantitative.

CGAP

Submitted by Achal on

Hi. This is very pertinent question. The question that “what solves my research question” is the basis of present thinking. In fact a lot of participants suspect it is the combination of qualitative and quantitative research that solves a research question better.

Sahayog micromanagement

Submitted by Mr.Kumar on

Thats a valid point Mr. Prakash. Many times we get entangled in the methodology and forgets the core issue. But it is also true that in most cases people have their in house capacity in one of the above methods, and so could not dig deeper into other. Also, I have witnessed that people are biased towards either methodology based on their personal experience.

Consultant

Submitted by Sharad on

I agree that both the methods complement each other and provide much better result when used in combination. The challenge I foresee in this approach is around which method to be deployed at the start. Qualitative methods are very useful to dig deep into selective issues and generate useful insights, thus researches might need to narrow down on those important issues. The most credible methods to short-list those important issues are quantitative (although qualitative means can also be tried). This means that a quantitative study should precede a more intensive qualitative one. On the other hand the general practice is to conduct a qualitative assessment first to fine-tune the quantitative methods and questionnaires?

Consultant

Submitted by Mukul K Singh on

It seems that the house is divided over what is the best approach. Some are in favour of combining quantitative and some other are in favour going either with quantitative or with qualitative methods. I think referring to practical experience on this would be useful.

To start with an experience of my own: we recently completed a unique series of studies called "Agent Journal Research". This was a study of m-banking agent networks. It used a combination of qualitative and quantitative methods. The quantitative data gave the network level broader picture and qualitative inputs helped us explain many whys thrown up by the quantitative data. For example, quantitative data indicated that many agents did not make enough profit. The qualitative inputs explained the reasons. You can refer a paper based on the study: http://microsave.org/briefing_notes/policy-brief-6-assessing-agent-profi...

MicroSave

Submitted by Sonal Agrawal on

In session 7, Olga, Mukul, and Rajarshi discussed about how and why qualitative researches inform decision makers better. It will be worthwhile to mention it here…
Findings are just one of the many factors that help the board to decide. The other factors may include the importance of the decision, attitude of the board about numbers and qualitative findings etc Quantitative research data gathered from a representative sample can help us generalise certain findings for a defined population. The questions that we can effectively derive answers for in a quantitative questionnaire would be limited in their probing capability compared to a qualitative tool's questions and techniques. For instance, we can very well use a quantitative research to gather client preferences (from a broader sample) in terms of say a suitable repayment frequency for a prospective loan product. We can use qualitative research on top of the quantitative research to ascertain the 'reasons' for why a certain attribute is preferred that way- this can go into further refinement of the product and also help profile the needs/ preferences for different segments of the client base (hence achieve more customisation and better fit to client needs)

MicroSave

Submitted by Nihal Kandanaar... on

We often see the usage of both the techniques. An example of using the qualitative and quantitative together could be product development. First, qualitative assessments like focus groups and individual interviews might tell providers what features are important to include in a product. Quantitative research will help identify the potential customer segments and estimate the potential market size for the product. Qualitative research can then be used to test customer reactions to the product prototype. In this way, both techniques together yield a client-centric product.

Sewa Micro Finance- Sri Lanka

Submitted by Tanuj Khandelwal on

I think combining qualitative and quantitative tools will always give a better idea and help to solve the problem more efficiently.
But its really difficult to combine both to get a good solution , as in some situations both will end up in opposite results.
Grameen Bank Intern

IIT Bombay

Submitted by rod dubitsky on

And this points out why it's critical to do both. Analyzing contradicting conclusions from qualitative and quantitative will help reveal analytic or data flaws and help improve overall design (was the qualitative data biased due to poorly executed survey or quantitative method was flawed). Or perhaps both conclusions are correct, but measure different things: qualitative results reveal strong improvement in empowerment and perceived ability to withstand shocks, while RCT reveals that average income effect is neutral. In the contradicting results, we have the ability to get deeper insights into the intervention as well as the research methods.

BRAC

Submitted by Diana Lewin on

Of course it would depend on the objective of the research. Here are some examples that I can think of where qualitative or quantitative research would be most useful:

When designing a product, using both approaches (quant and qual) can be useful and yield to stronger results (given that resources are available), but most of the information required could be gathered from qualitative research only (assuming resources are limited).

When trying to understand transaction patterns among clients, probably a first step would be to analyse existing data from the MIS (quantitative analysis - assuming that the data is available which is not always the case) and then combine it with qualitative research tools to understand why clients are transacting in certain way.

Does anyone have other examples to share?

MicroSave

Submitted by Tanaya Kilara on

This may be a good point to discuss specific research questions people have and the best method (more specifically than qualitative or quantitative) to answer those questions. I would love to hear the questions that practitioners are grappling with.

CGAP

Submitted by Mukul K Singh on

As the emphasis of the session is on exploring the the possibility of combining the two methods, we would like to go deeper into the instances where two have been combined. As most of the comments indicate that there is possibility of combining qualitative and quantitative methods. It would be interesting to ponder over how the two can be combined. The question assumes more importance as the two methods are essentially different and hence the attempt to combine them poses some challenges.

MicroSave

Submitted by Alok Kumar Mishra on

As to combining qualitative and quantitative, the two essentially needs to be done sequentially. That way it yields maximum effectiveness. As we also discussed on the previous day’s session on quantitative techniques, using qualitative tools at the beginning of the research helps refine the decision problem into testable hypothesis. This then drives the quantitative research design. However, qualitative research can also follow quantitative research to provide more insights into the essentially numerical outcome of the quantitative research.

ICICI Bank

Submitted by David Cracknell on

In Africa we are often asked to combine quantitative and qualitative approaches. In this respect I agree very much with Diana's earlier message which talked of using qualitative data to understand quantitative. There are challenges as always - where the quantitative data and the qualitative responses show different results, but often I have seen with some very large quantitative surveys a challenge in understanding the data that is presented. So the institution commissioning the survey is left with data but not enough knowledge as to what does this data mean. In the FinAccess Survey in Kenya this led to extensive qualitative studies to try to interpret the quantitative data.

In more recent studies we've worked with a leading market research company who handles quantitative surveys well, but we've worked with them to develop the survey instrument and to interpret results. It can sometimes be a problem for a company which specialises in managing data to understand the usually industry specific data they collect.

Yet it is the interpretation of the data which is fundamentally important. This is where detailed in depth industry knowledge is important, and where qualitative data is important too.

One of the challenges though is the organisation of qualitative and quantitative approaches. Ideally they would be in sequence, with one informing the design and development of the other. So qualitative results are used to develop (and focus) the quantitative questionnaire. But in practice due to time considerations I see that this approach is rarely taken and that the quantitative and qualitative surveys are conducted at the same time.

However, much of the value of a quantitative survey is lost if the survey is not well conducted, or the data is not carefully collected, or cleaned. If this is not done, then there is a real danger or misinterpretation. Its the same on the qualitative side, if you are not having representative focus groups, or a sufficient number of groups or triangulation of tools. So the key in combining qualitative and quantitative is in part down to design of the approaches and quality in implementation.

MicroSave

Submitted by Claudia McKay on

Nihal, I think that is a perfect example of how qualitative and quantitative methods can be used together. It sounds like the conversation is reaching a general consensus on this. To give one more example - when I worked at Opportunity Bank in Malawi, we used both methods to look at the reasons for high drop-out rates for a certain loan product. We first did data mining to really understand whether the drop-outs were stronger in certain geographic areas, amongst males or females, young vs. older clients etc. We also had some survey information and had a general idea of what people liked and did not like about our services. However, ultimately there was absolutely no substitute to going out and talking to people individually and in focus groups. How we targeted who we spoke with and the questions we asked were very much influenced by the quantitative information, but ultimately through in-depth conversations we found a level of granularity about perceptions of our brand, people's cash flow struggles etc. that we never could have gotten from just the quant. methods.

I think there is a danger when providers or researchers fixate too much on any one method. Never assume that doing more focus groups is a substitute for quantitative research or that a long survey will give you all the in-depth information you can get though qualitative research methods.

However, what if you can't afford both or you don't have time for both? How do you decide what is most useful? Also, do most researchers have the skills to do both types of research or does this mean contracting two different firms?

CGAP

Submitted by Mr.Kumar on

I also agree with you Claudia. But I do not find much material , when it comes to document such studies where both methodologies have been applied to conduct a research. Can you share some of the materials where I could find such studies?

Consultant

Submitted by Richard Meyer on

Claudia: Is your Opprtunity study available?

Retired OSU

Submitted by A. K.Sahai on

Combining both the technique s could be a good solution but it should be done on case to case basis. With the recent upsurge in agent based banking there has been a rising to need to gauge their effectiveness. While doing such research it is suggested that a combination of qualitative and quantitative techniques to be used. This will help to capture the financial performance as well as other qualitative factors such as customer relation, Stake holder support, etc.

Following is a report that uses a proper mix of both the techniques

http://www.ngobox.org/the-shadow-of-hunger/

NGOBOX

Submitted by Nara Hari Dhakal on

Dear All:

Experiences of the Nepalese researchers on access to finance indicate that quantitative and qualitative tools complementary to each other in order to source the client insights. There are hundreds of tools, both quantitative and qualitative, that can be used for undertaking researches on different dimensions of microfinance operation. Extending the microfinance services to the clients requires deeper insights about clients’ needs, behaviors and preferences on microfinance products and services.
Researchers often see quantitative and qualitative methods to be completely different, but it is not true. Often they are two sides of a coin. In general, quantitative methods include tools such as surveys and deal objectively with numbers and statistics and that of qualitative methods such as focus groups and individual interviews deal subjectively with intangibles like words and emotions. Most practitioners do not like quantitative findings and they comment such research pickles without salt, while quantitative research at times seems to be general and documentation of what we have observed in our day to day work.

There is no doubt that quantitative tools are needed if we are trying to quantify a certain problem or understand how prevalent it is by projecting results to a larger population. The essentially refers to use of sample statistics in the form of mean, standard deviation, normal growth, rate of growth, as well as relationship between different variable. This is basically use of numbers. This is not enough always. Under this circumstance, qualitative tools are needed in an exploratory phase of research when we are not quite sure on what is needed or how to approach the problem and develop a clear theory of change. Even to properly explain the rationale behind certain test statistics derived through quantitative methods, we need to rely on data collected through qualitative methods. In general, quantitative and more objective data can provide a structure to analyze subjective qualitative data and measure how prevalent a certain view is in the population. Qualitative methods can be used to dig deeper into certain issues that arose through qualitative methods like a focus group or observation or in-depth interview.

Against above notion, I think that quantitative and qualitative tools can be combined to yield even more powerful insights into customers.

Senior Advisor, Centre for Empowerment and Development, Nepal

Submitted by A.RAY on

Hi. Mr. Dhakal Thanks for this wonderful insight...

Social Entrepreneur

Submitted by Nara Hari Dhakal on

Hi A. Roy:
You are most welcome. This is basically how we blend qualitative insights with quantitative statistics and build the confidence of the microfinance practitioners.

Senior Advisor, Centre for Empowerment and Development, Nepal

Submitted by Abhay on

An example of using the two researches together could be where qualitative research can explain the findings obtained from the quantitative research. For example, quantitative research may indicate a significant causal relationship between field-staff behavior and client drop-out. Qualitative research in this context will explain how and why staff behavior leads to client drop-out. This finding is more valuable in addressing client the problem than a mere correlation between drop-out and staff behavior.

MicroSave

Submitted by Mukul K Singh on

Thank you Mr Dhakal for elaborate insights. What we are essentially talking about are three key dimensions: quality of the outcome, the cost effectiveness of the process and requirement of expertise.
1. Combining the two methods leads to better outcomes
2. However, it automatically entails additional costs
3. It also requires higher level of in house expertise or additional cost to buy it.
In practical world, one would need to strike a balance among the three dimensions. Is it possible to still combine the two methods and maintain the balance?

MicroSave

Submitted by Nara Hari Dhakal on

Thank you Mukul K. Singh for the nice synthesis and putting more precisely the key dimensions related to blending of qualitative and quantitative methods.

Senior Advisor, Centre for Empowerment and Development, Nepal

Submitted by Narasimhan Srin... on

Of course we need both to be used together - but do we know how to do it so as to get the best out of it?
A demand assessment for a niche product was being carried out which said that 90% of the sample was not really interested in what was offered. The qualitative study tried go in to why so many did not want the product. Being a niche product, even a 5% sample getting interested would have been a sufficient market and the qualitative study should have focused on what attracted the 10% to the product and tried to build on the same. If however the demand assessment was for a mainstream product the focus on those who did not like the product would been more appropriate. It is not just the use of the two techniques together, but also the the kind of issues that we pursue qualitatively arising from quantitative results that would determine both the cost efficiency and relevance of the results.
At times the enthusiasm for the research proper carries the researcher away from the most relevant questions and hypothesis. The 'compulsion within the researchers' to design a large and complex research project with quantitative and qualitative elements built in is an aspect which we should guard against.
Srinivasan

consultant

Submitted by Jagdeep Dahiya on

Using the two techniques is not easy however. The challenge can be as simple as in to decide which research to use first. Whether one should first go with quantitative research to understand general trends and then explore the ones which are critical qualitatively. Or Whether do a qualitative research to first understand critical issues and then understand if the findings could be generalised.
This is just an example of the complexity of combining the two research methods.

Jagdeep

Submitted by Chris Linder on

I think most of the examples and thinking here are built around special purpose events (once a year, etc), especially product development. I am wondering if someone can talk about what kind of a client monitoring system can be put in place for just everday operations - for monitoring client satisfaction, customer service, and possible unknown risks or operational problems. Clients and field staff can be excellent sources for all of this.

So I am wondering about thoughts and experiences on creating a simple day-to-day client monitoring system, combining qual and quant - using the MFI's own resources - such as field staff, specialist research staff, and even middle management - as well as the MIS. Thanks!

AZMJ.org

Submitted by Narasimhan Srin... on

I just reproduce an extract from a response given by me in an earlier session
"Many organisations do not realise that ongoing business operations actually constitute research-in-process and if the MIS and data capture is well organised, you avoid some expensive parts of quantitative work."

To your specific query on client monitoring, I know of a MFI which has real time SMS based quantitative data on what was the collection rate in a given cluster, what was the level of attendance of members and how much time the meeting lasted at the cluster on a continuous basis throughout the working day to a central monitoring desk. Where the levels of attendance and collection are below a threshold, the desk initiates an enquiry processes that reaches the centre manager within 30 to 60 minutes of the reprot coming - on the same working day. The reasons at each borrower's level does not matter to the central monitoring desk, but the pressure of the enquiry on the groups and cluster staff does rein in voluntary default.
The default and attendance data is also analysed to trace patterns, each day as also over a period to see whether customers are acting in concert, whether local economy has problems in some areas, etc. Data analysis reveals trends, and specific studies of the reasons for trends that are abnormal are carried out to garner more information to enable the MFI take new actions or formulate strategies depending on the nature of the insights.
But such use of operational data in realtime and using the same for business decisions is not commonplace; only a handful of institutions do it and in the banking system in India it is unheard of.
Srinivasan

consultant

Submitted by Tinu Jain on

Qualitative Research VS Quantitative Research has been interesting topic....
Experts of each level would criticize the other or say that the other method is secondary. I am into quantitative research and use the qualitative part as explanation of the outcomes from the quantitative analysis, but at the same time do believe that any disruptive research idea has to come through qualitative research and that would be more or less validated through quantitative research.
Basically two sides of the same coin, thus sometimes one would come up with heads while sometimes will have to take the tails.

IIMC

Submitted by Akhand on

Dear all, the discussion so far has brought up insights on why and how the two techniques of researches should be combined to maximise synergies. The discussion has brought up that using the two research techniques definitely increases the efficacy of results, there is an explanation of trends and vice versa how trends could be generalised.

However there could be challenges in combining the two methods. It will indeed be interesting to note the cost effectiveness of combining the two research methods. What has been your experience towards this? Further does it necessarily mean collaboration, as single institution may not have both capacities thus further entailing costs (thanks Mukul)?

Further, in terms of how to combine the two researches together, we can share more on how planning a combination of research methods shall happen, which part should precede the other?
What should be the ideal framework to analyse the findings? Does using the two techniques mean presenting trends and then explaining deeper insights or first explain the research questions and critical factors from qualitative exploration, and then present descriptive stats on critical factors. And whether it is also a challenge not to overwhelm the audience with two different type of research findings.

Akhand Tiwari

Submitted by Claudia McKay on

Hi everyone, in the last few posts we've had quite a few comments about the TIMING of quantitative vs. qualitative research and which precedes the other. My sense is that there is not a clear order that will always work. Ideally, this is an iterative process where one research informs the other and they each sharpen each other. For example, qualitative methods such as focus groups or individual interviews may be used to get a sense of key issues and the right way to ask questions. A large scale quantitative survey may then be done to elicit information across a larger population size. This survey may very well bring up issues or challenges which we then need to understand more deeply vis-a-vis more qualitative research. What do others think? Do you agree that this is an iterative process or do you think there can be a clearly defined order and that one should always precede the other?

CGAP

Submitted by Mr.Kumar on

Below is one such study where both quantitative and qualitative methods were used… “By collecting qualitative measures of instruction, we hoped to get an understanding of it what instruction that affects literacy development looked like. We used a detailed classroom observation guide, where an observer wrote a detailed narrative and rated teaching on five dimensions as the main source of data on instructional activities. Supplementing these data were bi-weekly logs of representative lessons provided by teachers, as well as periodic formal and ad hoc discussions with teachers about their instructional approaches and goals.” http://www.leslla.org/files/resources/RealWorldResearch.doc&ei=lQWRUIztO...

Consultant

Submitted by Claudia McKay on

Hi Chris, I think you pose a great question about how this can be done for the day-to-day running of the organization (and as much as possible by the organization itself on a limited budget) versus one-off research projects to tackle a specific issue, usually done by experts. I'll lay out some thoughts below but am certainly interested in hearing from others.
- MIS - We talked yesterday a lot about MIS systems and how this is the bedrock of information for MFIs and yet many MFIs still struggle to have a stable MIS. Even where they have a stable MIS, is there a user-friendly functionality for getting the reports that are actually useful? I do think this is the first step and can provide extremely valuable information on clients, performance of loan officers, popularity of products, vintage analysis to understand the quality of the loan book etc.

However, we all know that just because clients are buying a product, it does not mean that they are happy with the product or that it truly meets their needs. What are other ways to gauge customer satisfaction and performance of the institution?
- Customer comment boxes in branches are common in developed countries but will low-income customer segments use these?
- Has anyone had experience with mystery shoppers to objectively gauge the quality of customer service?
- How do organizations glean the insights that the front-line staff (loan officers and tellers) have? They have the closest interaction with customers and often have the best ideas for improvements. This can be done through team and staff meetings but also perhaps through competitions with prizes for the best new ideas.

I am sure that this is just the tip of the iceberg. What are other methods (either quant or qual) that other providers use to measure the quality of their products and services and satisfaction of customers?

CGAP

Submitted by Nina Holle on

In my opinion, another important way to gauge customer satisfaction and performance of the institution is customer feedback. Bancamia in Colombia has been succesfully implementing a structured and transparent way of getting feedback from its clients. They also actively encourage their customers to provide feedback. I know from a field visit, that feeback was an important issue to Bancamia. And they noticed that the feedback culture among their clients tended to be very low, so they tried to change it.

CGAP

Submitted by Monique Cohen on

Many years ago I worked with ODEF in Hondurous on customer satisfaction. They developed an approach to using qualitiaitive data that made sense for them but probably would not pass the test with the purists. What did they do?
They trained their best loan officers to conduct FGDs with clients arround client satisfaction. this turned out to be a win-win for the staff, the clients and the MFI. But it too time for each to be confident enough to discusss openingly the strengths and weaknesses of the MFI.
The stafff learned that criticism does not lead to job loss but rather changes that can improve their productivity, Cients learned that they can openly and constructively crticize processes and bring about change. The result was staff and client empowerment in the context of the ability to effect change. The insitution made changes based on some of the recoomendations that can improve their bottom line. More Imformation can be found in the IDS Bulletin, Microfinance, Poverty and social Performance, Vol. 34, no 4, 2003. authors are Monique cohen and Katie Wright

Microfinance Opportunities

Submitted by Claudia McKay on

Hi Monique, it sounds like this was a very successful approach. I normally would be a little bit nervous to put loan officers in this role. On the one hand, they already have a good relationship with clients and they are experts on the products and services. However, I would imagine that for a variety of reasons clients would be reluctant to share openly with the loan officers (fear that they would not get repeat loans, or not wanting to hurt feelings etc.) Do you know how ODEF handled this? Have you seen examples where MFIs used other staff that are not outside research experts but not the loan officers themselves (HR officers or some similar position)?

CGAP

Submitted by Jasmina on

I agree with many of you who think that the research question should drive the methodology applied. In some cases, quantitative methods can provide useful insights, but sometimes, we need both methods.

I heard a presentation from Vijay Chandok (ICICI Bank) two years ago on the results of a very interesting research ICICI conducted to better understand the needs of their SME clients. They started with quantitative research but they decided to conduct qualitative research with focus groups as well, as some of the findings were not clear and, as they wanted to know more about the problems of SME managers. While the quantitative research indicated many issues (“what?”), the qualitative research helped them understand the reasons behind (“why?”). They found for example that the most important feature for SMEs was proximity and easy access to services as SME mangers/owners are often very short on time. They also found that SMEs valued the 5 five basic products they were offering and that no fundamental product innovation was needed. Based on these insights, ICICI focused its efforts on improving processes and efficiency, using technology.

CGAP

Submitted by Sarah Rotman on

It sounds like there is a general consensus that using a range of both qualitative and quantitative methods is the best way to source customer insights. However, this is very expensive. Currently, it seems that a lot of research is paid for by donors. Is this something that providers will gradually start to pay for themselves? Philip Kotler (economist and marketing guru) has commented that companies tend to spend half of one per cent of their marketing budgets on research and 99.5% on promotion. Should we be advocating a shift in priorities and asking providers to take on a larger share of the burden of paying for research?

CGAP

Submitted by Sarah Fathallah on

I think part of the reason it is as expensive it is because providers are still not seeing the need to develop research capacities in house and continue to look out for outside agencies / consultancies. The downfall of using outside capacity is that knowledge transfer does not always happen (and often times is not even intended). But then again, I am not sure how to convince providers that it could be beneficial for them to invest in internal research resources... Maybe if the donors make it as a requirement?

Reboot

Submitted by Getaneh Gobezie on

But the urge for doing research is context specific!

Where there is serious compeetition for clients, service providers have to do real research and identify areas where they can do better for their clients so that they can win competition. That means service providers would clearly see the value of doing the reasearch, and probably can set aside budget.....

However in situations where there there is little competion for clients, service providers can offer (virtually) any product, without worrying about meeting clients need, and can still sell it to cusomers who have no other choice (exept the traditional, expensive, individual money lenders, etc) and make a huge profit.... Under such circumstances, the research need to be pushed from outside in the interest of the poor client. Donors, governments, central banks should come in here.... Unfortunately many contexts at least in rural Africa faces this challenge...

Regards

Getaneh Gobezie
Independent Gender and Ruralfinance consultant
E-Mail:- getanehg2002@yahoo.com

Independent Gender and Ruralfinance consultant

Submitted by Sarah Fathallah on

Dear Akhand,

I agree with your point that qualitative and quantitative methods can be needed for distinct situations. You mentioned the need to use qualitative tools in an exploratory phase of research in order to better frame how to approach a problem. I couldn't agree more. Often times, providers want very specific *business* goals (customer acquisition, increasing market shares, etc.). But when providers lack a textured understanding of how customers use money and financial products now, and what sort of products these customers need/want in the future, service delivery (including product development, distribution, marketing, etc.) is often times geared to fill in gaps or opportunities seen in the market (e.g. untapped geographies, customer segments, etc.), as opposed to unmet needs from users. Now we hear a lot about using human-centered design methodologies to help substantiate that knowledge. Yet, service provider might not have the bandwidth and/or in-house capacity to carry out such research in a rigorous manner.

And even when they acquire this knowledge, translating consumer insights into successful product ideas poses a significant challenge, as it requires significant design efforts to make sure product ideas are extremely elaborate in terms of their business model, their interaction with users and user-facing agents, their communications strategy and how they integrate within the service provider’s existing offerings portfolio and overall market strategy.

To me, qualitative methods are a non-brainer when it comes to try to get nuanced understandings of often non-traditional consumer groups. Once we've gone beyond the one-size-fits-all approach (that often comes from once a market matures enough for players to compete against each other), starts the search for other opportunities. And user-centered design in particular is a great approach to conducting contextual research to uncover consumer and market insights, matching them to a provider's strengths, and designing tailored products, services, and communication strategies.

Reboot

Submitted by Claudia McKay on

Hi Sarah, thanks for bringing up human-centered design as one way to gain that textured understanding of clients. CGAP has worked with IDEO to go through this process and it was extremely qualitative. There were about 20 in-depth individual interviews (with lots of observations as well as different exercises to draw out the financial behaviors of the interviewees). There was no quantitative research and yet a variety of product concepts were developed from the interviews and the providers have tested these and are bringing some to market. Does Reboot use a similar approach? Can you explain more about the rationale behind doing a small number of in-depth interviews without a large-scale survey or other quantitative analysis?

CGAP

Submitted by Sarah Fathallah on

Hi Claudia,

I think it goes back to a fundamental reasoning behind what makes someone use a financial product. If it's only a matter of economics and incentives (e.g. cost, price), then we could design products in a lab without ever talking to customers. My opinion is that human psychology is still a very important factor in determining product use. And getting to a deep enough level of understanding does require spending time with clients. Even when you do spend significant periods of time together, it is still hard to assess all of their needs and choice ecosystems, especially since some needs will be latent (not expressed). Hence the need to go beyond surveys.

At Reboot, we do use a very similar approach, although sometimes we use quantitative data to substantiate some of our hypotheses.

At the end of the day, I think it's a trade-off between depth and breadth. There's only so many resources you can put towards conducting research. But I certainly feel there is a place for qualitative methods, in particular when you're talking about new product development.

Reboot

Submitted by Claudia McKay on

Thanks to all who participated in this session on combining qualitative and quantitative research methods. Although some participants are still holding out and strongly prefer one method over another, I think the majority seem to feel that each type of research plays an important role. Quantitative methods can tell us the 'what' while qualitative methods explore the 'how' and 'why.' We've heard some great examples of both methods being used in combination - in product development, in a provider seeking to understand the SME segment better, and others. As always, the most important starting point is to clearly define the research objective and then the specific research methods can be matched to answer the question(s) posed.

We've also started some discussions around how a provider can do this themselves and who should pay for this. I think this is a fantastic segway into the next session on making client insights part of your DNA. I hope you will all stick around and explore this topic further!

CGAP

Submitted by Robert Lule on

Came in late. But using quantitative or qualitative tools to obtain client insight- I believe the key word here is insight the underlying output from the study. Client insight is capturing the feel, the experience and understanding of clients on a specific subject, to be able to deduce perspectives. Am certain that the tool in a quantitative research may not be sufficient to draw this kind of information from a respondent. Nevertheless, the approach we use in research is conduct a qualitatve research first to help design a quantitative survey tool. Benefit segmentaion research is one area that might go through this kind of phased approach.

Ayani Inclusive Consultants

Submitted by Igbogo Peter on

Good day all! The idea of quantity and quality should not be underrated. If we focus on qualitative approach only, the effect will be deep but will lack the ability the satisfy: having few useful approaches to meet the needs. On the other hand, quantity without quality is like having more uselessness in useful vessels.
A combination of the two will mean, having more useful approaches than it is required to meet the entire needs.

Living Your Dreams Initiative