That is Saul Griffith, a wunderkind inventor who created a new way to make eyeglasses on an inexpensive device in developing countries. For that invention and others, Griffith won a MacArthur genius award. The eyeglass machine was a great invention; unfortunately, the real constraint turns out to be testing eyes and writing accurate prescriptions rather than making the lenses."It turned out we were solving the wrong problem."
Griffiths' story is highly relevant to the aid business, especially these days as new foundations and people with experience in technology are trying to help address challenges faced by the world's poorest. A lot of smart people are trying to help, and since their expertise is in software or technology, they (naturally) try to use those tools first.
But as Griffith says in the article, "The speed with which software-based activities and web innovations catch on - text messaging, eBay, Twitter - has encouraged public perception that transformative technological change takes place almost instantaneously." Unfortunately, this is not the case with most development challenges, as in the case of the new eyeglass machine.
A more fundamental problem is that the inventors don't have a good way to assess demand. So they invent all sorts of things that *seem* to make sense, but turn out not to be used or adopted by the intended beneficiaries. Over at the CGD blog, April Harding has written about the "product pileup" in the health sector, where new agencies and foundations have created products such as Malaria Rapid Diagnostic Kits and bednets. She cites a book by Laura Frost and Michael Reich that discusses why these innovations are not being adopted, and what can be done to reduce this mismatch between supply and demand in the health sector.
With support from the Rockefeller Foundation, GlobalGiving and Innocentive have teamed up on a pilot to reverse the sequence of innovation. First, we ask the communities what they need. And then we advertise that need to Innocentive's virtual teams of inventors. This pilot is described here.
Illustration source: http://www.buzzle.com/articles/small-business-financing-is-missing-the-target.html
2 comments:
Love this post Dennis. And love the pilot too.
I am looking forward to see what can happen when you ask communities about their needs.
Another important way to make sure you are solving the right problem is to monitor implementation well. An example, in Nicaragua a few years back, a program was developed aiming to increase use of health services in rural areas. Conditional cash transfers were, and are, very fashionable for increasing use of health services. So, that's what they did in Nicaragua. But, guess what. It didn't work. Because there weren't sufficient services (in the public sector) available. Because they were monitoring program impact closely, they were able to detect the lack of impact, and took steps to expand supply. They did this by contracting health services from the private sector. And, voila. They got their increase in services.
Later, due to lack of funds, the CCT element was discontinued. And service use didn't decline.
So, by doing good program monitoring over time, they were able to go from solving the wrong problem, to solving the right one.
Here is a link to the paper on the Nicaragua experience. http://www.cgdev.org/doc/books/PBI/11_CGD_Eichler_Levine-Ch11.pdf
April,
I could not agree more about the power of ongoing implementation monitoring. Doing it in real time is 100x better than the "batch" method we often use, where we evaluate a project ex-post, when it is too late to do anything about it. Here is a post about what Scott Guggenheim did in Indonesia, which seems so simple yet was so powerful: http://denniswhittle.blogspot.com/2010/03/on-shoulders-of-giants-scott-guggenheim.html
And we have started designing pilots that will allow beneficiaries to text their answer to questions such as "How well is that health clinic serving your needs?" As Paul Cadario has pointed out, the best results happen when there is real-time bottom up feedback combined with top-down expert evaluations and audits. I think that getting the two to work together in near real time is the next frontier.
Post a Comment