Blog | August 18, 2015 | Jen Williams

Over the past twelve months WaterAid, Itad and IRC have been conducting desk research to better understand why some ICT initiatives improving water supply in rural areas succeed where others don’t.

In this second blog of the series, Jen Williams discusses the designs of the eight different ICT initiatives from the study, and explains how they work.

Making sure breakdowns are dealt with quickly

We looked at two rural initiatives which relied on community crowdsourcing through SMS for reporting of breakdowns of water points.

  • The first, Mobile Phones for Water (M4W), was implemented in eight rural districts of Uganda between 2011 and 2014 and was designed so that, if there is a problem with a water source, a community water-user sends a text message to the system to report this. This SMS is automatically sent to the relevant hand pump mechanic’s phone and the district water office. The hand pump mechanic is responsible for supporting the community with the repair and the district water officers are responsible for keeping track of the reports made and checking if repairs have been carried out.
SNV World 2013: Mobile Phones for Water (M4W)

SNV World 2013: Mobile Phones for Water (M4W)

  • The second project, Maji Matone in Tanzania, was a six month pilot in 2010/11 where text messages were sent to the project team before being forwarded to the mobile phone of the district water engineer, and to a local journalist. Follow up messages would be sent over the next three weeks but if a report was not acted on by the  in that time, journalists were informed that actions were not being taken. The aim was to pressure local governments through the media to act quickly to help the communities with breakdowns. Maji Matone was discontinued and declared as a failure due to only 53 text messages being received in 6 months but the project was actually quite successful in turning reports into action to fix the problems.
  • The final rural initiative we examined, Smart Handpumps in Kenya, was quite different. This initiative does allow users to report a breakdown via mobile phone but largely relies instead on a GMS transmitter fitted inside the handle of the pump which automatically and periodically sends text messages about use of the pump and functionality to the project teams. The one year pilot has shown good results with a ten-fold reduction in handpump downtime and a functionality level of 98%.

Urban crowd sourcing initiatives

We also looked at three crowd-sourcing initiatives in Kenya, Tanzania and India which help to show how crowd sourcing can work successfully in an urban environment.

  • Maji Voice in Kenya has similarities to the first two initiatives above, as it relies on crowdsourcing with water users reporting when there’s a problem. However, the key difference is that it has been introduced in urban rather than rural areas, including Nairobi and Nakuru. This means that the project focuses primarily on paying customers with accounts rather than on community water points. The water supply company perhaps has more incentive to deal with paying customers’ feedback efficiently, whilst customers are probably more likely to complain to their service provider if they are paying for their service.

    Maji Voice also gives water users a choice of ways to report – including SMS, phone call, website and visiting the office directly – with many still preferring to visit the office directly rather than rely on ICT, even in urban areas.

    -Jen Williams, WaterAid

  • Human Sensor Web in Zanzibar also relied on crowdsourcing and asked community users to send text messages to the water authority if they experienced water shortages. However, this didn’t focus on customers who pay regularly and get water delivered directly to their premises, but instead on urban stand pipes and water kiosks used by communities. In practice, follow up from Zanzibar Water Authority did not get embedded in the initiative and very few text messages were received because there was little trust among communities that the water authority would respond. Arguably, there was less incentive than with Maji Voice because it didn’t focus on paying customers.
  • The Next Drop initiative in India was also introduced in cities but information flows the opposite way – Next Drop was contracted by the water board to keep customers informed about the specific times they should expect to have water (many users only have supply for two hours per day). The water board’s valvemen give information to Next Drop about the times at which the valves will be turned on and off by calling a toll free number or entering it through a smartphone app; Next Drop then uses SMS to share this information with the water users.

Improving planning through regular reporting

The final two initiatives are different as they don’t focus on dealing with specific incidents but instead focus on periodic reporting of coverage and functionality for the local and national governments to use in planning of their work.

  • The Water and Sanitation Information System (SIBS) in Timor Leste uses local government staff to collect information in rural communities, including on water source functionality, water quality and time taken to collect water, every three months by sending a form through a text message to a national system. The aim is that this data is then processed at the national level and sent to local governments via CD (because internet connection is poor) to use for planning purposes.
  • Re-imagining Reporting is similar to this system but relies more on the staff of the INGO Water for People for processing and use of data. The initiative is currently being implemented in nine countries but, for this research, the focus was on Bolivia where the project is being implemented in six rural municipalities. There is a Water for People staff member who supports each team of data collectors (students, community members and a private company have all been involved) and brings the data back to the office to be uploaded. Once the data has been collected, it is Water for People staff that carry out the data cleaning and analysis before encouraging sector stakeholders to use it for planning purposes. If the municipal governments were to lead the data collection in future, Water for People would also need to build the capacity of government to analyse this data in-house.


About the author

Jen Williams is an international development consultant

About this blog

On 8 September, IRC will host a webinar to discuss the initiatives and research results in depth so please email Joseph Pearce ( ) if you are interested in joining, asking questions, and sharing your thoughts on what works and what doesn’t.