A team of developers has created a single system, called Masawa, to encourage development and delivery of information services to low-cost cell phones via mobile apps. Working with NGOs and local developers, the team will test the system in Kenya in 2011.
For millions of the world’s poor, cell phones are a primary source of information and an important means of communication. But cell phones still do not serve low-income users as well as they could. The cell phone market in many ACP countries is often divided between several network providers, and it is fragmented, with different users having different needs. Producers of cell phones and services rarely meet those needs adequately, as they tend to supply only what is already available, rather than tailoring their products to fit the customers’ requirements. The services that are available tend to be text-based applications that are of little use to people with low literacy skills.
Joshua Haynes, a graduate of The Fletcher School at Tufts University in the US, saw these limitations while testing an SMS-based application he had developed to deliver agricultural information to farmers in Niger. ‘The project worked quite well as it was part of a literacy programme,’ says Haynes, ‘so the people involved could read and write. But it wouldn’t be easy for other organisations to replicate, in either Niger or elsewhere, as most NGOs don’t have sufficient staff with technological expertise working in the field. NGOs are becoming increasingly aware of the issues, and may have ICT specialists working at their headquarters, but too many organisations are not yet capable of successfully applying cell phones to their project work. They often only use communications technology to keep in touch with the main office.’
It was in Haiti, however, that Haynes saw the real power, and problems, of working with cell phones. He had developed an SMS-based app, along with other Fletcher students, to help savings-led microfinance groups run savings accounts via cell phones. ‘The main problem, however, was that the accounting method limited the possibility for individuals to save different amounts. If one woman wanted to put in more money, the group treasurer would have to calculate proportional interest, which isn’t a straightforward calculation. So we developed an app to try to solve this issue, and became excited at the additional benefits of a savings group app. The groups then had a savings and lending history that could act as a credit history,’ says Haynes. ‘But after we started on-the-ground tests, we quickly realised that we shouldn’t have used SMS – it was too cumbersome.
‘When setting up a new local savings group with the SMS app, for example, someone would have to type out each new saver’s name and details, and every saver would require a separate SMS. To do that for 20 people takes a long time and is very prone to human error, and it’s not easy to correct those errors once the SMS is sent.’
The experience also showed the development team more general problems of working with SMS-based apps. ‘You have to be more than semi-literate to use SMS,’ says Haynes. ‘You have to be functionally literate, especially if you want to receive information by sending a specific code. The code has to be exact, or the system has to be programmed to handle the number of different possible input mistakes. If users repeatedly receive unexpected responses, they may quickly become dissuaded from using the service. Also, you can only send a very limited amount of information via SMS and it cannot be visual. You cannot send, for example, a short animation or video. SMSs are also relatively expensive. In general, the price to send an SMS has not gone down in the way the prices for voice calls have in most countries.’
Haynes points out that there can be privacy issues with SMS. Sent messages are stored on the cell phone, unless the user knows how to delete them. This could present problems for a woman reporting domestic violence, for example, or when someone sends details of voting irregularities during an election.
While testing the app in Haiti, and through their other work, the team realised that there were problems with other types of apps too, not only with those using SMS. ‘We saw that many organisations were creating apps, and these apps would be developed for a particular sector. But the apps only reach a certain number of people. Beneficiaries of a health organisation may have a health related app, and those served by microfinance organisations may have a finance app. It seemed common sense that people need more information than from just one health app, or one finance app, they need both, and much more.’
The team decided to create a system to develop and distribute apps easily to low-cost cell phones. After their experiences with SMS-based apps, the team now concentrates on apps written with the Java programming language, which means users would need a Java-enabled cell phone. ‘There are more than two billion Java-enabled devices in the world today,’ explains Haynes, ‘many of which are in developing countries, and the cost of handsets continues to come down. Plus, most programmers in the world know Java, so we wanted to bring the two together – the developers and the technology – to produce mobile apps for the specific needs of people in developing countries.’
The team set up a non-profit organisation, called Masawa (derived from the Arabic word for equality), that works with NGOs, international aid organisations and microfinance institutions (MFIs), to determine exactly what their target audiences need. They then contact local developers to create apps to meet those needs, and Masawa delivers these apps through a single system.
‘Currently, there is very little coordination between the projects developers are working on, or could be working on, and what people really need,’ says Haynes. ‘There is no way to monetise the process of app development and distribution, to make sure the people creating the apps get paid for what they do and to cover the costs to organisations who want to distribute information in this way.’
Masawa works with NGOs and technical schools to find developers to work on apps for the system. The team hopes it can encourage developers to concentrate more effort on apps for Java-enabled phones. Many developers are now working in Android and iPhone platforms, because they can earn money there. Masawa hopes to provide another source of income for local talent. NGOs and other organisations will pay the developers to create the app, while developers could also earn money by producing apps that people would pay for. Masawa would also provide the means to distribute those apps.
The distribution of mobile apps brings its own challenges. Users can normally download apps directly onto their phones via the cell phone network, but the networks in many countries are not yet capable of handling that process. Apps can also be downloaded from the internet and installed onto a phone, but too few people have access to the internet in the areas Masawa is trying to reach. Plus, many of the apps for Java-enabled phones offered on the internet are games, and not intended for delivering information services.
To get around these issues, Masawa uses the existing networks that their partner organisations already have in place. They provide initial training to field staff from the organisations on how to install the system onto phones and to help them train people to use the apps. When field staff travel in the normal course of their work, they take a small netbook computer and use it to download Masawa onto people’s cell phones. They can update the system on subsequent trips as new apps become available. Users then see a single icon on their phones through which they can access the various apps.
Local developers will produce most of the apps, working to a simple set of specifications so that each individual application fits into the overall system. This ensures that all the apps are distributed and updated easily onto each user’s cell phone. Masawa has already created a few apps that provide basic information, but also show their potential for low-cost Java-enabled phones. The health related app, for example, provides information through animations, and can show the correct way to breastfeed a baby, or how to mix oral rehydration salts.
Another app is an on-demand message board where people can receive information. Here, a health ministry, farming cooperative or local trader can pay to have messages appear on the board. They could, for example, announce a round of vaccinations for children in the area or give up to date advice on pest control or provide weather details for farmers.
There is also a microfinance loan and comparison app, and a children’s educational game, while another app provides agricultural information, such as how to tackle tomato blight. The information for this app is stored on the phone and it is not necessary to send or receive additional information. ‘These apps all show how we can deliver health, agricultural, educational and general information using a variety of different media,’ says Haynes. ‘They show how flexible the system is, and the many ways in which it can deliver information.’
Masawa is free and, depending on demand, it is also possible to include premium apps providing popular services, such as football results, for which people would pay. The cost of sending and receiving information via any of the apps depends on tariffs set by the cell phone networks in each country, but the cost of data transfer, which is how most of the apps would exchange information, is usually much cheaper than that of sending an SMS.
‘We will work with our partners in Kenya to test the system,’ says Haynes. ‘The team has worked closely with a number or people and organisations there, and there is an existing pool of creative, skilled developers to add new apps onto the system.’ Moreover, the data transfer and SMS prices are relatively cheap, compared to many other countries. Masawa and their partners will closely examine the progress of the pilot, aware from their previous experience how valuable it is to learn from mistakes.
‘The team will monitor the project very closely and methodically,’ says Haynes. ‘We really need to determine how people in rural communities consume information, how they use it, where they get it, and how effective that information is. Progress might be quite slow at this stage,’ he adds, ‘but only because we don’t want to expand in the future without understanding how to get information out to people.’
Even going ‘slowly’, the team estimates that their system will be available globally by 2013, and that by 2015 it could be installed on 10 million cell phones. Masawa has already seen success, winning the Social Entrepreneurship award in the 2010 Tufts University Business Plan Competition.
While Haynes is optimistic about the future of the system, there are, he says, many challenges ahead. ‘There are still some technology barriers; each series of phone, even if it’s from the same manufacturer, requires a slightly different piece of software, for example. Unless manufacturers address this issue, the only solution for developers is to invest a lot of effort into producing the different software for each phone, or base our work on a limited number of phones.’
Haynes is also realistic about the potential of Masawa, and the effect other mobile apps projects can have on the lives of people in developing nations. ‘There are very few stories of really successful app projects. I know many more that have failed than succeeded. The technology is often not the problem. It may be that the implementing organisation doesn’t have the staff or resources to properly incorporate ICTs into their projects. We’re not saying that Masawa is the answer or that it will solve those issues, but it does offer another opportunity. And we have to remember that technology alone will not solve poverty, it can only be part of the solution.’
Projet ABCs IMAC: Information sur les Marchés Agricoles par Cellulaire
As part of a literacy programme, this SMS-based app developed to deliver agricultural information helps farmers in Niger maintain their literacy skills.