Documenting the undocumented: How one company counts the 'invisible poor'

Melanie Edwards knew there was a problem when market researchers had more information on the video game habits of American teenagers than on the populations of African nations. She launched Mobile Metrix, a market research company that gathers data on the "invisible poor."
Written by Christina Hernandez Sherwood, Contributing Writer

Melanie Edwards knew there was a problem when market researchers had more information on the video game habits of American teenagers than on the populations of African nations.

Drawing on her background from the Peace Corps, Wall Street investment banking and technology research, Edwards launched Mobile Metrix, a market research company that gathers data on the "invisible poor." We spoke recently about the project.

Why do we know so little about more than half of the population?

A lot of it is around access to these people, whether it be from the time they're born and not having them registered or just where they're living, whether it be remote rural areas or what they call "mega cities." The access to these is often difficult because of the labyrinth of buildings on top of each other in the case of the urban poor or because of violence, mainly drug trafficking, where people don't go in in order to get access to people and actually have them counted.

What are the consequences of knowing so little about so many?

I like to equate it to a family of seven. There are seven billion people in the world, that's the family of seven. Four of the members in our family, we don't know anything about. We don't know about their education levels. We don't know about their health conditions. You can imagine if four of your siblings were ill or illiterate or unemployed, that would have a big impact on the family as a whole.

How did you develop the idea for Mobile Metrix?

It became more obvious to me as I worked in developing countries and urban poor settings and realized how much we don't know about these people compared to when I was at the other extreme in corporate America. You could segment the market out the wazoo. I could tell you what the favorite video game is of the 17-year-old boy in the United States, but I couldn't tell you what the average age is of someone who lives in Chad. How could it be that over half the world's population we hardly know anything about? The other side of the coin is we spend billions of dollars -- whether it be from foundations or from places like the World Bank --to try to eradicate issues in this population without really knowing who they are and what they need. I think that's when it shook me the most: when I realized how much money we're spending when we're not even sure the issues and problems are. Another light bulb went off when I was working in one community where I was told there were 55,000 people, and yet I got estimates anywhere from 5,000 to 60,000. It was all across the board. What does that mean for how off we are on numbers around the world?

Talk about how Mobile Metrix works.

We're a market research company that's serving the base of the pyramid. Part of what's unique about that is how we do it. We hire local youth who come from these communities. They go door-to-door in the communities to collect demographic data using a handheld [device], whether it be a PDA or a mobile phone. That data is used by our clients, which could be governments, corporations, nonprofits or foundations, to more effectively channel their resources, their products and services, back into the community. It's a job creation scheme, especially for youth at risk, and it's also a way of collecting data about the community.

Where do you do this work?

We've launched in Brazil. From a ground operation standpoint, that's where we are at the moment because we're still in start-up mode. We're also consulting in other areas and we've also been asked to go to other countries, primarily in Africa and India. We've covered about six communities, roughly around 10,000 residents. These communities are much larger, but we'll do a random sample. About 50 mobile agents have been hired over the course of the last couple years. We're kind of focusing more on health care, but we're being asked to do it in other industries.

Talk more about how Mobile Metrix works in healthcare.

The first step is actually identifying how deep the need is. How bad is the diabetes? How bad is the dengue fever? In most communities we found that medical care is higher on their list as far as health care needs. That information gets back to the Ministry of Health and the community leaders. Accurate information is a fundamental first step to solving any social problems. From that information, it's working with the organizations that have the power to make changes.

What results have you seen so far?

The most prominent was during the dengue epidemic. We worked with Johnson & Johnson. We collected information on dengue fever and infection rates, which were around 26 percent in some communities. Dengue fever is not contagious, but it's deadly. The mobile agents who were collecting the data would educate the individuals on dengue fever by showing pictures of the dengue mosquitoes and they'd leave some anti-mosquito repellent on behalf of Johnson & Johnson. When we went back and did an impact assessment, we saw incredible changes in the knowledge levels. We really [increased] by about 40 percent how much people knew about dengue and the dengue mosquito. And once the mobile agent takes off their uniform, they're still the go-to in the community for anything about dengue. They're seen as the experts.

What's next for Mobile Metrix?

We're raising money to do a diabetes project which would be across 18 communities. We're looking to do that not only in Rio, but to bring that to other countries, as well. We've been asked to bring a Mobile Metrix lab to an inner city in the United States. We're narrowing down where that will be. The third one is the countries we're expanding to.

Photo: Melanie Edwards

This post was originally published on Smartplanet.com

Editorial standards