The Story so Far (about my actual work)

This post was transferred from my old blog. See the original post (with comments) at:


Hey Everyone,

Sorry I’ve been a bit intermittent with the blog posts lately. It’s probably no surprise to anyone who knows EWB, but recently I have been busy (italicized for emphasis). Plus quality internet access has been pretty sparse.

Just as a random note, the internet situation has recently taken a turn for the better. At our last EWB team meeting, we discussed for the umpteenth time how communication challenges really hurt our effectiveness. Magically, however, this time we decided to do something about it. As a result, I’m now the proud owner of a fancy little USB internet thing – meaning I can now get wireless internet anywhere in Malawi that I can get cell phone service.

Now, I’m sure some of you are thinking: “how on earth can someone get widespread mobile internet access while working in a fairly small, fairly rural district of one of the least developed countries in the world?

To you I say: welcome to modern Africa. Few things exemplify the contrasts in modern Malawi more than readily available wireless internet signals being broadcast through villages of mud and thatch huts. Modernity exists everywhere here; it’s access that’s the problem. But I digress.

My intention with this post is to share a little more about my actual work here. For the past 2.5 months, I’ve been stationed in Thyolo, partnered with the Thyolo District Water Office.

Thyolo is one of two districts that benefitted from COMWASH, a $11.9 million Canadian funded project that worked on water access, sanitation, and health in Thyolo and Phalombe districts from 2001-2007.

In Thyolo I am working with four gravity fed water systems (GFS) that were installed or rehabilitated by COMWASH. My goal here is to learn about challenges being faced by the GFS management committees, and make recommendations to improve the schemes’ long-term sustainability.

The challenges facing the schemes are formidable. Just over 2 years after the end of COMWASH, there is definitely a mixed picture of scheme functionality. The four schemes I work on are called: Limphangwe, Didi, Mvomuni, and Sankhulani. Each of the schemes has between 80-110 water taps, connected to small springs and rivers in the surrounding mountains.

2009.06.03 GFS TapA water tap at the Sankhulani GFS. The water comes all the way from the forested mountain in the background, powered only by gravity (hence the term “Gravity Fed Scheme”). 

Since the project finished, the Limphangwe scheme has been working consistently at 80-90% functionality (e.g. 80-90% of the taps are working at any given time). Meanwhile, the Didi scheme has been the polar opposite, falling to around 20% functionality. Numbers for Mvomuni and Sankhulani are harder to come by, but I believe they are both somewhere in the 50-70% range. (Note: the above numbers are based on my field observations and conversations with scheme management, not on full surveys – they are not meant as official, but I believe they are fairly accurate).

What I am trying to do is to learn what factors make the schemes succeed or fail, and translate that learning into concrete recommendations to improve scheme management.

One thing I can say right now is that managing these schemes is not easy. Most of the work is done by volunteers, and the work is hard. I spent a week shadowing repair teams at the Sankhulani scheme, and I swear I have never sweat that much in my life. Almost every day that week the scheme broke down, or we had to do routine maintenance, sometimes necessitating 20km+ walks through the mountains. The work was exhausting, but worse still, it never seemed to be finished.

SDC13868Volunteers and paid staff fixing a pipe at a river crossing on the Sankhulani GFS. It rained heavily the night before and the subsequent flood washed a tree down the river, knocking out the pipe. In the village where I was staying we woke up without water.

On my last day with the repair teams we fixed a leak at a road crossing (the pipe had been hit by a tractor grading the road). After half a day of work the pipe was fixed, and I assumed it was “problem solved”. However, I have since learned that same pipe became clogged later that day, and since that time (almost 5 weeks ago) no taps downstream have had water. This kind of thing is frustrating for both repair teams and water users, but it happens all too often – illustrating how challenging it is to maintain a GFS system.

SDC13989Mr. Alizeo, a paid repair team member (and all around nice guy) shoveling out mud at the main water intake for the Sankhulani GFS. A dirty job preceded by a half day of uphill walking to get on-site.

One thing I am working on directly is helping the District Water Office do improved monitoring of the GFS performance. To that end I have worked with District staff to create a financial auditing program, and to implement a first pilot over the last few weeks. The system we are using is very simple, and I believe is quite similar to something that the COMWASH program attempted to implement.

SDC14009Just for fun (e.g. unrelated to most the content above or below): me standing with Amayi ndi Abambo Mauritia, the couple who hosted me while I was at the Sankhulani system. The house in the background is where I stayed for the two weeks I was there. (Coinciding with the two weeks I had no internet and almost no phone, for those who were trying to contact me).

GFS management committees are supposed to hire revenue collectors, who will travel around the entire system collecting money from taps each month. The households who use each tap are supposed to collectively raise 100 Kwacha monthly (about 80 cents) which will be used for spare parts and other expenses related to maintaining the scheme.

Based on past work by COMWASH, and our recent work at the office, we are thinking of re-implementing a very simple form for tracking revenue collection and system performance, with a few small changes. The form looks something like this:

Tap Name Revenue Collector Jan Feb Mar Apr Etc

The following may only appeal to monitoring and evaluation geeks, but I think it’s pretty interesting. Each revenue collector gets a copy of this form, with her/his tap names already filled in. For each tap, they keep monthly collection records. If they manage to collect 100 Kwacha from the tap, they write “100” in the box for that month. If they don’t manage to visit the tap that month, they write “DV” (didn’t visit). If the tap-users refuse to pay because the tap was broken down that month, the revenue collector writes “BD” (broken down). If the tap-users refuse to pay for any other reason, the revenue collector writes “RP” (refused payment).

This seems very simple (because it is), but in fact it puts a lot of information in one place – information we don’t currently have. Currently we document revenue collection with receipts. This means to figure out total money collected for a month, a year, etc., you have to sort through dozens or hundreds or receipts, copying them out and totaling them up by date and tap name. With this new system you’d just need to quickly look at a one-page form.

This new system will also give information that the receipt system doesn’t. For instance, if there’s no receipt for a given month, there’s no way to know whether the tap was broken, the community refused to pay, or the revenue collector didn’t visit. With the new system, we can get the following:

  • Total amount of money collected for any period
  • Functionality statistics for the system for any period (because the form records breakdowns)
  • Information on the performance of the revenue collectors (e.g. are they visiting all their taps?)
  • Information on community payment (so the management committee can shut off taps that are routinely refusing to pay)

Best of all, the system doesn’t require any use of computers or outside staff. You don’t have to take all the forms and amalgamate them (although it might be useful), you don’t have to put it into Excel – each management committee has the information right in front of their eyes. Further, the information is displayed in a very clear and useful form – it’s easy to see if a tap has been broken for months (so you can fix it), or if a committee hasn’t been paying (so you can shut off their tap), or if a revenue collector isn’t managing to visit all their taps (so you can get them some help). The system will hopefully enable decisions to be made easily at the field level.

Finally, the system will allow the District Water Office to do easier audits of financial records and performance for the schemes. The form would make it easy to get figures for total revenue collection (although we’d still need to check some receipts, just for verification). The figures for total revenue collection can then be checked against expenditure receipts, bank records, etc., to ensure that all the money’s ending up in the right place. Finally, the form would give easy information on scheme functionality, making course-correcting faster if a scheme starts to show signs of failure.

Of course, all of this is still a work in progress. The form will likely go through 1-2 more iterations, and will definitely need to be translated into Chichewa. Following this, I still have to get final buy-in for it, and then the biggest challenge: implementation.

So, like all of EWB’s work overseas, no guarantee this goes anywhere. I think it has a fairly good chance though. There’s good buy-in form people at our office, and from the management committees we work with. Many of the community volunteers are extremely dedicated to making these schemes work, and see the value in this information as much as we do.

I leave Thyolo in 3 short weeks, onto the next phase of my placement (still to be determined). By that time, I will have this system finished and in the field (most of the legwork is already done). My co-workers are very competent, and I believe they will take things from there.

I will also be submitting my final report on general recommendations, which I think will have some good ideas, and which I know will be taken seriously. I’ll definitely provide some updates as things move forward.

As always, thanks for reading.




  1. Thanks for interesting reading! I have a question: Who is volunteering to be managing and repairing these gravity fed schemes? Are they community leaders who are sort of expected to participate, or is it individuals who are stepping up? What types of jobs and/or families do they have? (is there a certain demographic that is more typical? Like, more well off, less well off, type of job, etc.?) Based on absolutely nothing, other than your pictures I guess, I'm inclined to assume they are mostly men, is that correct?

    Happy transitioning to the next phase of placement!

  2. Hey Owen,

    It's interesting to see how small M&E contributions can have such a snowball effect on the whole program in terms of management.

    I have a few questions regarding the storage of information. Are all the revenue collectors based at the District Water Office ? Where and how are the forms going to be stored ? I've seen so many written paper reports and forms being lost, meaning that information is lost forever.

    It would be interesting to have a plan where the forms are photocopied periodically, or you can hire a young intern maybe a month a year who would be in charge of typing them up in Excel.

    On another note, do you think GFS being repaired at the community level like this on a volunteer basis represent a sustainable solution to water supply ? In Burkina Faso, water supply management committies that actually work are very very very rare, because of many many different reasons. The national strategy seems to be aiming in a more « private sector » perspective, at least in terms of repairing.


  3. hey Owen. good writing, helped me a lot in writing an assignment about Sankhulani.
    i have one question though: who does the Sankhulani (and the other gfs's) belong to?
    thank you and good luck in yor next placement.