In much of the developing world, small and medium enterprises are a staple of the economy—providing goods, employment, and income for local communities. But small-scale shop owners tend to keep only limited records of their profits and inventories, which can make it tough to manage a business. This lack of data makes it even harder to understand why some microenterprises succeed and others fail. 

Enter Radio frequency identification technology (RFID). RFID tags are used by major retailers in the United States to record inventory, without having to individually scan each item via barcode. These tags have a small, radio-wave emitting microchip that allows them to store and transmit inventory data for tagged items. When a product with a RFID tag is within range of a reader, the tag reflects a signal back to the reader, which then counts the item.

Researchers from the World Bank and the University of Peradeniya in Sri Lanka decided to test RFID technology in partnership with 20 microenterprises in Kandy, Sri Lanka— with the goal of generating better data for business management. But it didn’t work. 

When innovation outsteps practicality

There’s a lot we can learn from this failure. The research team learned that RFID tags are not only difficult to use, but ill-suited for a microenterprise environment. That’s because the tags fail to work with commonly sold products, such as food and beauty items, that have high water content or metal packaging. Both water and metal can block the signal headed to the RFID reader. Moreover, given space constraints of a small storefront, it is common practice to stack products high or behind each other— making it ever more difficult to get a reliable count with RFID.

One textile shop owner in Daulagala, Sri Lanka observed, “If clothes are stacked on top of each other, or if they are placed on hangers behind each other… the machine has trouble reading properly… the project team has to disturb the arrangement of clothes in the shop. I am quite certain this technology is not useful for my business.”

This project highlights an instance where innovation outsteps practicality. RFID doesn’t work with products commonly sold by small business owners, it was complicated and costly to set up, and it wasn’t an improvement over the status quo: physical counting of inventory (or simply surveying business owners) can generate useful and more accurate information. Most of us agree technology is a critical driver of economic development, but it’s also not a silver bullet.

Failing forward

What makes this story unique? This failure was published, providing an opportunity for others to learn. The full story can be found in Development Engineering “Dev Eng”, a new open access research publication that highlights development innovations—including failures. The journal was launched with support from the Higher Education Solutions Network (HESN) at USAID's U.S. Global Development Lab, and is providing evidence on what works (and what doesn't) in the field of technology for international development.

RFID Project leads David Mckenzie, Lead Economist at the World Bank, and Suresh de Mel of the University of Peradeniya share their perspective on why they published their findings,

“It was important to us that we share this story because measuring microenterprise turnover is notoriously difficult, and we had high hopes that this new technology would provide an objective and new way to measure this. The trial dashed our hopes, but also gave us more confidence in standard survey measures, and we have already heard from others contemplating using this technology who are now considering alternative options as a result of us sharing our failed experience.”

Dev Eng is just one of many outlets encouraging dissemination of null results and failures. Other initiatives include the All Trials Campaign which calls for results from all clinical trials to be published. Recently, notable organizations like the World Health Organization have similarly advocated for the public disclosure of both positive and negative clinical trial results. Other outlets include an annual failure report released by Engineers without Borders – Canada and platforms like Admittingfailure.org, which allow civil society organizations a venue to admit, write-up, and publically share their failures.

“We see major benefits from publishing studies that find weak or no impacts. In global development, there should be no silent failures; there is inherent value in learning from interventions that fail to achieve their intended impact,” note the Dev Eng Editorial Leadership, Temina Madon, Ashok Gadgil, and Paul Gertler, in the publication’s introductory article.

Development Engineering "Dev Eng" is a global open access, interdisciplinary journal applying engineering and economics research to the problems of poverty. Dev Eng is driven by the hypothesis that early insights from economics and the social sciences in the innovation process will yield solutions more likely to achieve development impact and scale. Dev Eng was incubated by the Development Impact Lab at UC Berkeley with support from the U.S. Global Development Lab’s Higher Education Solutions Network.

About the Author

Sarah White is a Senior Program Associate at the Center for Effective Global Action, a hub for research on global development at the University of California, Berkeley.

Related Links: