Case Study: Patterns Found in Data Help Charity Fight Poverty

For the Family Independence Initiative, data isn’t just facts and figures that measure results. It is the fuel to power people’s climb out of poverty.

The charity believes that poor people know best what they need to improve their economic well-being. It recruits families in small groups and encourages them to set goals and depend on one another — instead of caseworkers — to achieve those goals. Participants can access matched savings programs, home and auto loans, and other resources through the charity.

Every month families report on their progress. Some of the data they provide is easy to quantify: income, savings, rent or mortgage payments, children’s school attendance, and whether the family has health insurance. Other information is more free-form: why the family’s income went up or down, the challenges they’re facing, and their long-term goals.

Stronger Grant Proposals

The Family Independence Initiative mines the information to find patterns. In Detroit, for example, people focus on transportation more often than in other cities because of the unreliable public transit system. The nonprofit uses findings like that to raise money.

A lot of nonprofits go to foundations and say they know what poor people need, explains David Henderson, the organization’s chief data officer. “We have a ton of data where people are saying, ‘This is what we want,’” he says. “We can go to funders and say, ‘Look, here’s the demand. Help us meet that demand.’”

Making Recommendations

Taking the data a step further, the Family Independence Initiative is taking a page out of Netflix’s playbook and building a recommendation engine.

The software will compare a family’s information to data from successful program participants who shared similar circumstances and goals. It will then recommend steps the family can take and identify other program participants who were in a similar position and can offer guidance. Right now, the charity is testing an early version of a recommendation engine and hopes to start using it this summer.

Initially the engine will make recommendations based on the monthly data families report, but in time, it will also draw on data from UpTogether, a private social network hosted by the nonprofit, which people use to share advice and support. Mr. Henderson says that capability will “supercharge” its recommendations.

“We think about the families as the experts, that we’re learning from them,” he says.

Tracking Progress

Mr. Henderson is also excited about another new data-driven project: quarterly email updates that will show each participant how his or her economic well-being has changed and how it compares with that of other people in the program as well as with local economic indicators. The first emails go out this month.

The idea is similar to exercise trackers like Fitbit that monitor users’ activity levels, says Mr. Henderson. The organization wants families to be able to track their progress and take pride in their accomplishments.

“They submitted the data,” he says. “We have the responsibility to take that data and not only make it work for them but to show it back to them.”