Ideas to Action:

Independent research for global prosperity


My dishwasher died over the holidays.

The first thing I did was go to Consumer Reports to find out what their experts considered the best bet for a replacement.  I was on the verge of ordering one of their top-ranked models when I decided to click on the "User Reviews" tab.  I was shocked to see that the model was ranked only 2.5 out of 5 stars by actual users.  Consumers had a wide range of complaints, describing how hard it was to load and the length of the wash cycle, and others complained the thing broke down too often.  So I kept going down the list of recommended models until I found one that the experts liked and the users loved, and then I went online and ordered it.  It was simple, and it took less than an hour.

What if governments - and even regular citizens – had a similar tool to use in development projects?  Or better yet - why don’t they?  When I went on my first "mission" to Indonesia in 1987 as a young economist at the World Bank (something I will talk about in a future blog post), I was struck by the fact that there was almost no way for farmers to tell us “experts” what types of agriculture projects they wanted.  Nor could they tell us how they felt about our existing projects.

So when Mari Kuraishi and I left the Bank in late 2000 to launch GlobalGiving, one of our first design principles was that anyone in the world should be able to comment on the projects listed on the site.  A couple of years ago, the Rockefeller Foundation encouraged (and supported) us to experiment with a new way of getting beneficiary views. The result is the Storytelling Project, and it’s showing great promise.

When I stepped back from day-to-day operations at GlobalGiving, I started thinking more broadly about feedback loops in development.  When I looked around, I was pleasantly surprised to see a lot of exciting experiments under way.   The work that GreatNonProfits is doing to mobilize user reviews of NGOs around the world is potentially pathbreaking.  At a project level, the Danish Refugee Council has launched an SMS feedback system that leverages the Ushahidi platform to display beneficiary views on their projects.

Official aid donors are also getting in on the act. The Mapping for Results initiative spearheaded by the World Bank Institute and Development Gateway aims to increase accountability by making transparent exactly where the Bank’s money is supposed to be going, and why.  As my colleague Owen Barder recently noted, DfID has recently committed to “publishing feedback from those directly affected by aid,” and it recently hosted a meeting together with the Omidyar Network on how technology might enable such feedback. Progressive government agencies are taking the lead in some cases.  The Philippines Department of Education and the Affiliated Network of Social Accountability (ANSA) launched to help monitor school resources and allow parents, educators, and students to note when books were missing and when toilets needed to be fixed, substantially reducing the time it takes to address these problems.

Companies like Google are getting involved as well, applying their technology acumen by giving grants that promote real-time monitoring of things like clean water.  The nonprofit Splash (previously A Child’s Right) created its own tool, Proving It, to achieve total transparency and to give constituents a way to hold Splash accountable for performance over time.

These examples are just the tip of the iceberg and could mark the beginning of a new chapter in development that allows better matching between what people really want and what they get.  For the full potential of feedback loops to be realized, however, there are many key conceptual, design, and logistical questions that must be answered:

  • How do government agencies, donors, and citizen groups provide incentives for broad-based feedback?
  • How do they know that feedback is representative of the entire population?
  • How do they combine the wisdom of the crowds with the broad perspective and experience of experts?
  • And, perhaps most important, how do they ensure that the feedback mechanisms are broadly adopted and actually lead to positive changes in aid projects?

I am thrilled to be working with colleagues at the Center for Global Development to try to answer these questions, and I would like to invite you to contribute.  Are these the right questions to be asking?  Do you have additional examples of feedback experiments that you believe are particularly promising (or particularly unsuccessful)?   We welcome your feedback and ideas; please put them in the comments section below, send them to us directly at, or tweet them to @DennisWhittle using the hashtag #feedbacklabs.