Technology offers tremendous potential to improve public sector services in developing countries. However technology interventions often fail to have the desired impact. We’ve collated experiences from a number of experts involved in the design and implementation of tech solutions in developing countries and describe four key pointers that could be the difference between success and failure.

IMG_8013.JPG

1.    Address the real problem

Sometimes great technology interventions are aimed at the wrong target. It's important to be very clear about the specific problem your tech intervention is addressing, gain a good understanding of the root causes driving this problem and be sure that this is a central problem to the system in general. We have seen a number of tech interventions which have been lifted from one environment to another without enough thought on whether they are addressing the central problems of this new system. At its worst, tech interventions can be a distraction from improving the system where they’re aiming at the wrong target.
 

2.    Always consider the end user

Designing effective tech interventions for developing nations requires a comprehensive understanding of who the end users are.  In particular, their technology skill set and working environment. The mismatch between people who drive intervention design (such as central policy makers) and the end users is often substantial. Where multiple end users exist, targeting information and features to the relevant user (e.g. managers, administrators, central policy makers or general users) can improve navigability and boost effectiveness.
 

Common challenges of end users in developing countries

Common challenges of end users in developing countries Design considerations
Limited access to hardware
  • What hardware do the end users have access to?
  • How do we make technology that works on that hardware?
Limited access to internet and mobile services
  • How do we reduce the data requirement?
  • How can the technology work offline or via a local copy?
Limited experience of end users with mobile technology
  • How do we make the interface more intuitive?
  • How can we simplify the interface?
  • How can we reduce the learning curve for the user?
  • How do we design effective training range of skill sets?
Limited ability to charge devices
  • How do we reduce the power requirements?
  • Can we increase battery size?
  • Are mobile charging devices required?
Language barriers
  • Is it possible to provide language options?

For example, in work on android based vaccinator tracking in Pakistan, the first version simply required users to open the app and push a button once a day to send their geographic location. As users became more comfortable using the software it was possible to add more features to gather more detailed data on vaccinator activities. Bombarding users with complexities in the first version would have caused significant compliance issues.
 

3.       Anticipate how the tech will operate for end users in day-to-day work

Every tech intervention requires systems and procedures to complement the implementation of the tech. Carefully examining how the technology will operate in the field throughout development can identify problems before they cause issues in the messy environment of real-world implementation. For example, applications which collect patient data but take time to input can inadvertently increase patient wait-times and reduce clinic effectiveness, or applications which require constant remote database access may fail at scale where mobile data coverage is patchy.  
 

4.       Consider complementary non-tech solutions

Complementary non-tech solutions will often be the best strategy to increase impact. Throughout the design, there are likely to be many obstacles which technology could solve, however it’s important to consider other realities such as the time, cost and complexity when deciding how best to solve them. For example, often paper based visuals can be more powerful than dashboards in environments where internet speeds are slow or logbooks can be used for more effective data validation than tech systems
 


About the Author(s) 


Emma Hannay is Director of Health at Acasus and has extensive experience working on public health projects in developing countries. Fenton Whelan founded Acasus and has more than a decade of experience in public health and education development.