clock menu more-arrow no yes

Why aren’t we harnessing technology to end homelessness?

For starters, why not implement predictive analytics for people at risk of being homeless?

Bill O’Leary / The Washington Post via Getty Images

This is a contributed article by Joel John Roberts, CEO of PATH and PATH Ventures, a homeless housing and services agency that serves more than 120 cities across California.


Back in pre-iPhone and pre-internet days, the old solutions to addressing homelessness were building shelters and transitional housing. But in the last few years, the tools for addressing homelessness have become increasingly more sophisticated.

Today, the new approach is what we at Los Angeles-based PATH, for People Assisting the Homeless, call “housing first,” where we provide people with an apartment and support them with employment, health and other services. Studies around the country have shown that “housing first” developments have a 98 percent retention rate, are more cost effective than shelter programs and increase access to services for tenants.

We now perform annual point-in-time counts that tell us how many people are living on our streets — 4,350 people in San Jose were counted in early 2017 — and where they are. We cannot solve a social issue like homelessness unless we know the full extent of the problem.

This month, PATH will break ground on a supportive housing development for the homeless in San Jose. For those of us in the world of housing and services, this type of development is a cutting-edge solution to addressing homelessness. It directly moves people from living on the streets to living in their own apartment.

We also have a highly technical triage tool that assesses a person’s physical and mental health status, length of time on the streets and other personal barriers that cause homelessness. The tool scores people’s “vulnerability,” places them on an index and prioritizes housing for people who are most vulnerable, or most at risk of dying on the streets.

Communities are also adopting a system of coordination that utilizes this triage tool for prioritization, and links all the services and housing in a region to this triage index. Previously, shelters and services helped whoever could walk into their front doors. Today, we find the most vulnerable people on our streets and help them access available housing and services.

You would think that with all this sophistication, homelessness would dramatically decrease, yet communities are still struggling with growing encampments, RVs and tents. The last San Jose count in 2017 found that homelessness increased by 3.7 percent since 2009.

Although today’s approach toward homelessness has made significant strides, we are still in need of new technology to disrupt the world of homelessness, similar to how the worlds of shopping, music, taxis and hotels were disrupted by Amazon, Apple, Uber and Airbnb.

A few years ago, Amazon filed for the predictive analytics patent, “anticipatory shopping.” Based on personal shopping data and on what websites a shopper visited, Amazon is able to predict what shoppers are going to buy in the future and make sure that product is within their neighborhood for same-day delivery.

So why not implement predictive analytics for people at risk of being homeless?

If those of us who provide homeless services could look into this technological crystal ball, we would be able to locate families and individuals who are on the verge of becoming homeless and provide the services or help they need in order to stay in their homes.

For people who are already on our streets, why do we have to perform a homeless count only once per year when we know that people become homeless 365 days per year? Why don’t we have the technology to know how many people are on the streets, where they are sleeping, why they are homeless every single day? That way, every day, outreach teams can locate and serve them.

How about building a new type of housing that is cost efficient, quickly built and doesn’t take up expensive land? Tiny homes and container homes are the first generation of this new technology. But our next generation of architects are going to have to disrupt their old approach to housing the masses.

Finally, funding for homeless solutions needs a new paradigm. Crowdsourcing is a next-generation tool that needs to be refined. Charities like PATH need new, simpler tools that would allow a donor to see who they are supporting, how they are helping and how they are funding much-needed non-program costs.

Are these suggestions simply utopian? I don’t think so.

I remember using a payphone on the streets. Today I have a personal telephone the size of my wallet.

So too, disrupting the way we solve homelessness may not be so utopian after all.


Joel John Roberts is the CEO of PATH and PATH Ventures, a homeless housing and services agency that serves more than 120 cities across California. When he joined PATH in September 1996, it was a local one-site project in West Los Angeles. Under Roberts’s leadership, the agency has grown into a multi-site, nationally recognized leader in social services, known for its innovative co-location approach. Roberts’s professional background is in international relief and development; he holds a master’s degree in cross-cultural Studies from Fuller Seminary, and a bachelor’s degree in communications from California State University, Long Beach. Reach him @joeljohnroberts.


This article originally appeared on Recode.net.