> Posted by Kim Wilson, Fellow, Center for Emerging Market Enterprises and the Feinstein International Center, Tufts University
Embed from Getty Images“Everything should be as simple as it can be, but not simpler.” This aphorism credited to Albert Einstein inspires our call to Lean Research.
Two Fridays ago at MIT a group of 50 of us met to hash out some principles that, if followed, might generate better research in development and social science contexts. NGOs, universities, foundations, corporations, government, and multi-lateral agencies were represented in our group.
Our analogy of choice was Toyota. If “the Toyota way,” or lean manufacturing as it has come to be called, could cause profound and beneficial disruptions in production processes, might lean research cause equally profound and beneficial disruptions in research processes?
Much of our discussion that day traveled along four paths, which grew into a starter set of principles:
Lean Research is right-sized. The sample and method are perfectly suited to the research objectives. Unnecessary questions are cut, and necessary activities are added in. It also means much homework must be done to ensure the study is necessary at all and that the information could not have been gathered through secondary research or through key informants.
Lean Research is relevant. The information gathered must be understandable to a wide range of stakeholders, including the subjects researched. It must be accessible and actionable, able to drive changes in policy or practice. Again, this means much homework must be done before the research even begins. Researchers must identify stakeholders who will actually use the information.
Lean Research is respectful, even delightful. Human subjects might enjoy a six-hour interview during harvest or they may not. Enjoyment depends on the benefits they get from the research experience (in Lean Research, we assume every research is a treatment). How do we maximize the experience of the human subject to get better, more accurate results?
Lean Research is rigorous. As one of our colleagues from MIT’s JPAL said: the most respectful research is rigorous. And that is true. Research that uses faulty protocols is a complete waste of a subject’s time and undermines findings that could advance knowledge. Lean Research does not permit cutting corners. Samples must be representative of the population, and methods unbiased. The research must be reproducible.
Alex Counts of Grameen Foundation goes a step further to ask: how do we give subjects the chance to rebut or confirm findings from their perspective (see this for his ideas)? He also proposes an annual top ten examples of research that has driven practice and/or policy, which seems like a great idea.
What does this all have to do with financial inclusion? First, many folks in the room represented institutions that dealt directly with financial inclusion such as Bankable Frontier Associates, CGAP, The Bill & Melinda Gates Foundation, SEEP, The MasterCard Foundation, and MasterCard Worldwide.
Second, it was easy to see how Lean Research principles might guide intelligence gathering in financial inclusion. For example, if a large-sample survey produces roughly the same evidence as a series of financial diaries, then such a survey may be more right-sized (efficient), while still following a rigorous system of protocols. If a study team has the objective of helping an MNO and a bank to develop a successful mobile money platform, then the team is obliged to find the result categories that are actionable or relevant, before even conducting the study. In fact to inform the study design, they must map backward from the go/no-go decisions that suppliers would need in order to take their next step. In terms of respect of the human subject, how about this example?
A gift that people like – but rarely get – is a portrait. So during the second round of our survey, we took individual portraits of each grain trader (415 in all) and then distributed them during the debriefing phase. When I visited one trader he said “But how does this help me? I have four wives so I need one for me and one for each of them!” (And the researcher, Dr. Jenny Aker complied.)
The next steps of our initiative are emerging. Several participants volunteered to crystallize Lean Research principles and guidance in a clear, readable way, so that they could form a kind of checklist. Others wanted to push forward the idea of a prize for Lean Research. A few agreed to identify research that are starting soon so that they may press them through the Lean Research sieve, with peers commenting along the way. MIT and Tufts are interested in convening a process mapping of the experience of the human subject, just as one would map a consumer experience. We are confident that other expressions of Lean Research will bubble from our participants and we hope from well beyond those who could attend.
If you are reading this post and were not among us on August 1 but would like to be part future conversations, please contact us: .
Kim Wilson is a lecturer at The Fletcher School and a Fellow with the Center for Emerging Market Enterprises and the Feinstein International Center at Tufts University.
Have you read?
The Holy Grail of Microfinance Research?


Facebook
Twitter
6 comments
Comments feed for this article
August 12, 2014 at 2:21 am
Getaneh Gobezie
Thanks for this post
I would be happy to be part of such a conversation.
One of my work (”Empowerment of Women in Rural Ethiopia: A Review of Two Microfinance Models”) has already been published by Tufts University:
http://fletcher.tufts.edu/praxis/archives/xxv/XXV_FrontMatter.pdf
LinkedIn Profile:
http://www.linkedin.com/profile/view?id=47039565&trk=wvmp-profile
Regards
Getaneh
(getanehg2002@yahoo.com)
August 12, 2014 at 5:48 am
Chris Linder
Kim,
Wonderful post. I would only suggest we try very hard to take it further and make it so lean and simple that the MFIs/FIs can do it without the likes of academics (no offense) or the RCT shops, similar to what the PPI tries to do. Choose just the key information/outcomes wanted and focus on those for in-house studies. And definitely focus on practical needs of the MFI, like product development, etc. We can still leave the more complicated questions and designs to researchers…but if not integrated into the daily operations of the MFI, any study will still be costly and disruptive, no matter how lean. Have we done enough to teach the FIs on the basics of research?
August 12, 2014 at 9:55 pm
Kim Wilson
Dear Chris, I think and hope that we can apply our “starter” principles to any method, including RCTs and participatory research.
Personally, I have never been involved in research that exemplified the four Rs. Something or some things were always lacking. But at least now, I know what direction I want to move in. I completely agree on MFIs. If you want research that helps you make a decision on how to improve services or be more competitive, start with the decision points and then map backwards – what information will you need to make a decision?
Let us know if you want to be on our mailing list.
Best,
Kim
August 13, 2014 at 11:43 pm
Chris Dunford
The sentiments Kim expresses on behalf of the August 1st gathering are on the mark. It is encouraging that this meeting seeks to better understand problems and solutions in the necessary interface between researchers and practitioners. Yet I have a concern. I was not at the meeting, but Kim’s and Alex’s blog posts this week seem to emphasize only the relevance of research for practitioners.
Chris Linder’s comment further focuses on the practitioner’s perspective. As he writes, practitioners should use easy-to-manage tools and protocols to answer their own questions whenever possible. Moreover, they should incorporate data collection as much as possible into their regular operations, so as not to disrupt those operations to get the information they truly need for management decisions. On the other hand, lack of research expertise often leads practitioners to collect a great amount of data that is unused or useless. Elaborate, poorly conceived, internal “monitoring and evaluation systems” can be the major cause of unnecessary intrusion into the lives of clients.
To do lean research often means outsourcing to research experts and doing less information collection as part of standard operations. Data should be collected from as few clients as possible and from them as infrequently as possible. Such efficiency (right-sized) and effectiveness (rigorous) in research is most likely achieved by specialized researchers with considerably more training and experience than is typically found in practitioner organizations.
Chris Linder writes, “We can still leave the more complicated questions and designs to researchers,” which is always costly and, to some degree, disruptive of standard operations. However, the problem of relevance arises when practitioners “leave” the questions and designs to the researchers. Relevance is achieved through successful dialogue across the researcher-practitioner interface. This is not exclusively the responsibility of the researchers. I would argue that the responsibility for relevance lies more with the practitioners. After all, it is the practitioner whose operation will be disrupted and whose reputation will be put more or less at risk by the research findings. Even if the research costs are totally paid by a third party funder, the practitioner should be driving a hard but intelligent bargain to make sure the research justifies the disruption and risks by addressing questions important to the practitioner, not just to the researcher, and make sure the research design does not impose unnecessary burdens on either clients or staff.
The problem in the researcher-practitioner interface is that practitioners and researchers do not “speak the same language.” Their training, perspectives, incentives, concepts, vocabulary and even culture are quite different. There are positive examples of successful negotiation of the interface, but not many. The challenge is to develop an understanding of what it takes to be successful in this dialogue. There is need for more interlocutors on both sides, but the role will mostly be filled by researchers better attuned to development practice. Perhaps the August 1st gathering will encourage more systematic identification and training of such people who speak both languages to achieve right-sized, relevant, respectful and rigorous research that serves both the deepening of understanding and improved practice of development interventions.
August 14, 2014 at 11:04 am
Alex Counts
Reblogged this on Alex Counts' Blog about Fonkoze and Microfinance in Haiti and commented:
Here is a great blog written by Kim Wilson, who invited me to the “Lean Research” gathering at MIT on August 1.
August 14, 2014 at 2:51 pm
Anne Hastings
Congratulations to Kim and all the organizers and participants of this meeting. I only wish my schedule had allowed me to attend. As a former practitioner myself, this is an issue that is very near and dear to my heart. I think it is a terrific effort and love the simplicity of: right-sized, relevant, respectful and rigorous. Let’s keep this initiative moving forward!