Three Lessons from Bangladesh

Minhaj Chowdhury
Minhaj Chowdhury is the CEO of Drinkwell, a clean water start-up based in Boston, MA.
IDDG Guest Speaker on 10/29/2013

Excited couldn’t even begin to describe my enthusiasm. I had returned! Two years after implementing a university-funded clean water project aptly titled “Clean Water for Peace“, I was now a Fulbright Fellow, ready to evaluate the scalability of a clean water intervention. And then it happened. I arrived at the project site and was greeted by Fatima, a short but powerful mother who said:

Why wasn’t I involved when you planned this?

You said the free water, food, and education I would receive would help us move beyond poverty and into prosperity. But all you’ve done is create dependency. Your free water filters were made of raw materials that could be sold in the bazaar for more than what my family makes in two months. The free food that eventually came from another NGO ran my uncle’s food stall out of business. The NGO-run school shut down because we didn’t have enough teachers. You didn’t help us. I actually think you made things worse.

All I want now is the dignity of work. Can you at least give me that?

As soon as Fatima had finished, Habib – a longtime friend and Manager with a partner NGO – immediately took me aside to his offices. I didn’t see Fatima the rest of the day. So, what happened?! I asked Habib. Why did Fatima greet me with such an impassioned monologue?

For starters, let’s talk about three: the number of filters being used two years after distributing 100 household filters in Golaidanga in the summer of 2009. John, a Chemical Engineer of German descent and college friend/partner-in-crime for the project, and I thought we had everything right. Buy-in from local stakeholders? Check. Employ a local caretaker to ensure uptake? Check. Upon further inquiry however, I learned Shikha, the employed female caretaker, no longer lived in the village. When visiting her home, I saw a picture of John and Shikha from the previous summer. John was the first bideshi, or foreigner, villagers had seen. Shikha’s picture with him had cemented her social status as one who mingles with bideshis. Five months later, this led her parents to wed their only daughter to a wealthy family. We failed to account for the cultural implications of a new face entering an established community. Consequently, there was no caretaker to ensure uptake.

Bangla1Local schoolchildren pose for a picture with John and Shikha [R] upon performing a play dispelling myths regarding arsenic.

We came to call this the “John Effect.” As a strong believer in the adage of “mistakes are okay as long as they are new ones,” I couldn’t help but think of just how many times the “John Effect” impacts an intervention.

As my focus for the next nine months was to understand villager perceptions of us, the well-intentioned development practitioners who just couldn’t seem to get things right, I found myself joining a movement alongside local NGOs wishing to add more of the beneficiaries voice to the bland monthly scorecards and dashboards. While well-intentioned at first, many of these qualitative discovery trips took form in the following manner: a Bideshi student or aid worker would select a village to conduct interviews regarding the progress of a specific intervention. Armed with a voice recorder, translator, and perhaps worse, a camcorder to “put a face” to the beneficiaries, the interviewer arrives in a remote village accompanied by local field staff. A focus group discussion is organized where the interviewer asks beneficiaries how a program is progressing. The beneficiary then responds to such questions surrounded by flashing lights (especially if it’s the time of year to publish annual reports) as well as community members. In my case, as the interviewer asked how many households used a clean water filter the beneficiary, a quiet and reserved mother, kept looking across the group, waiting for cues from an elder female before responding to each question. This leads to my first lesson from working in Rural Bangladesh:

Lesson 1 – The Bideshi Mirage

I was accompanying a pair of Swedish interns who were conducting a one-week assessment of microfinance programs in northern Bangladesh with a local NGO. Accompanied by a translator, the students immediately drew a crowd with their Swedish looks and Macbook Pros. As a Bengali-American, I looked a lot more like a local than they did, so I introduced myself as a local college student whose parents’ village was nearby. As the students began their interview with the mother and daughter of the family, I took notice of the father, Farooq, and took him aside to ask “Is your family going to tell the truth about the NGO’s work?” Farooq chuckled:

If we say how things continue to fail, the bideshis won’t return. They would be disheartened. If, however, we lie and paint a rosier picture, the bideshi visits that serve as a highlight of our otherwise monotonous day will continue for us! Better yet, they will invest more money into trying to help us. Why stop such a thing?

Bangla2 Local high schoolers crowd around a tubewell with camera phones ready as they await the arrival of bideshis.

Farooq’s point of view made perfect sense – why stop an innocent effort to help, especially when it meant mingling with people he can only see on television? He quipped, if we are to truly effect change, we must involve Bangladeshis, not bideshis. Such is the reality for a country that is more reliant on donor-funded NGOs than on its own government.

Lesson 2 – How can you ensure disruptive innovation with non-disruptive adoption?

There are many alleged innovative “low cost, high quality, game changing” products out there. In late 2012, I found myself mouthing these words at a now commonplace venue for social innovation, a hackathon. Boston Startup Weekend was the scene, and I, frustrated with the inefficiencies of gaining honest insights using a pen and paper monitoring and evaluation system, thought I had a solution all figured out. The name of the start-up was Ashalytics, an “early warning system for the developing world.” The idea was simple. Field agents across Bangladesh were going door-to-door and amassing 8 million single-page surveys of data for Bangladesh alone. By the time they are tabulated by hand, the data is outdated.

Since a majority of people in Bangladesh have cell phones, however, collecting and analyzing data in real time via mobile phones was low-hanging fruit, and something that could drive out the inefficient pen and paper reporting process. This mobile platform was Ashalytics. Organizations could subscribe to Ashalytics’ Software-as-a-Service. Villagers and Field agents use mobile phones to upload data directly to the cloud. This means managers can access information in seconds, instead of waiting for a month with the current paper-based system. Data collection costs are drastically reduced. Accuracy is improved. Armed with timely and actionable data, engineers and public health practitioners are better equipped to fix broken water systems, educate the populace, and deliver cleaner water faster. Furthermore, this data could be shared with donors, thereby eliminating the need for bideshis to visit the field and encounter mirage dynamic altogether.


[L] A typical NGO’s monthly dashboard. Time from survey collection to publishing results on the dashboard: 30-45 days. [R] Ashalytics’ real-time dashboard. Time from survey collection to publishing results on Ashalytics’ mobile platform: 1-2 minutes.

But, this came at an expense. It turned out that organizations embracing Ashalytics would let go of community employees whose sole job was to collect, sort, and tabulate the heaps of paper. While eliminating such inefficiency is a no-brainer in the private sector, such a move muddles the reputation of NGOs and social sector organizations whose cost-cutting moves can be at odds with its social mission of creating opportunity in impoverished communities. Furthermore, certain organizations would have to restructure their performance evaluation systems, as instant reporting meant instant knowledge of broken water points and flawed programs. Beforehand, field agents would have over a month to determine a mitigation plan and patch issues before management became involved. Real-time notifications meant managers would be inundated with problems that have no immediate mitigation plans, thereby changing their perception of their employees ability to solve problems. These adoption pains proved too much for our initial organizations, and outweighed the gains of cost-savings, streamlined donor reporting, and real-time monitoring. Put simply, our product was too painful to adopt.

During one of our weekly team meetings, we had to ask ourselves if we wanted to be in the business of displacing local enumerators. The alternative? Shift our strategy and align our platform with organizations who wouldn’t face such difficult tradeoffs. This meant venturing into the nascent world of social enterprise to see if Ashalytics would have a less painful adoption for smaller, more private-sector focused enterprises. And so away we went. Away from the donor-funded organizations whose large budgets wooed us into thinking we could secure sky-high valuations, and towards the volatile but exciting world of social entrepreneurship.

Lesson 3 – It’s job creation, stupid.

Today, Ashalytics is being designed for use with Drinkwell, a clean water enterprise that establishes franchise-owned water filtration systems primarily in arsenic and fluoride-affected areas across South Asia. Drinkwell provides clean water through a “Select, Build, Sell, Collect” model. In the Select stage, a Drinkwell Committee comprised of local government officials, NGOs, business partners, and school and religious officials is formed. The Committee selects an entrepreneur who owns a contaminated tubewell. In the Build stage, an $8,000 system is installed over one month using local materials and workers. The franchisee also hires 2 drivers and 1 caretaker to run the plant. Franchisees then Sell clean drinking water for $3 a month and keep a portion of monthly gross profits. Households can purchase a “Drinkwell Card” in local shops that have 30 punch holes for redeeming 20L jugs of water daily. Each system can, at a minimum, serve 600 households on a daily basis (capacity can be easily increased). Finally, Drinkwell “collects” customer data using Ashalytics (liters dispensed, household demographics, and other output data) to fuel customer growth.

As Drinkwell creates 3 jobs for each system opened, caretakers and franchisees use Ashalytics to monitor system usage. Instead of displacing community enumerators, Ashalytics can now be used as a tool to create jobs as the seamless reporting enables faster growth driving human capital needs of franchisees. As the endgame goal of most development interventions is to stamp a sustainable seal of permanence, it’s taken me five years to arrive at a simple concept to ensure true success: it’s job creation, stupid.


Leave a Reply

Fill in your details below or click an icon to log in: Logo

You are commenting using your account. Log Out / Change )

Twitter picture

You are commenting using your Twitter account. Log Out / Change )

Facebook photo

You are commenting using your Facebook account. Log Out / Change )

Google+ photo

You are commenting using your Google+ account. Log Out / Change )

Connecting to %s