by Nyein Sein, Marketing Manager

Aggregated searches for housing assistance on auntbertha.com in the past 365 days.

Until recently, homelessness in Los Angeles was concentrated—for the most part—in Skid Row, a downtown district just a few blocks south of City Hall. Starting about five years ago, though, a sharp rise in homelessness in Los Angeles meant that LA’s homeless population could no longer be confined to four square miles.

This heat map shows searches for housing assistance on Aunt Bertha’s public platform at https://www.auntbertha.com/ across Los Angeles and Orange Counties. Search trends on our free platform are a proxy for where people are falling behind in a community and what their most critical needs are. The search density spectrum moves from purple (low concentration of searches) to blue (medium concentration) to green (high concentration).

Twenty years ago, it’s likely that this map would have been a smaller sea of purple with one prominent, green dot overlaid on Downtown Los Angeles. Today, we see housing searches spread far and wide across both counties, with high search density in areas that have not been traditionally associated with homelessness.

We saw the highest volume of housing related searches in Santa Monica, which has experienced a growing number of homeless individuals living in its beaches and parks. The cities of Malibu and Pacific Palisades, previously isolated enclaves of affluence, have witnessed a migration of the homeless population northward along its beaches. (The other hotspot north of Santa Monica reflects the famous 90210 ZIP Code often entered by users exploring our platform for the first time).

It’s much easier for someone to become homeless in Los Angeles compared to other cities, due to the worrisome combination of skyrocketing rents and lack of affordable housing. The housing mix in Los Angeles, with its low ratio of multifamily housing to single family homes, has exacerbated both factors.

It’s worth noting that we observed substantial search volume by people in need for housing assistance in regions without a visible homeless population, such as Pasadena and the San Fernando Valley. Our data on search patterns can be useful for pinpointing communities where people may be at risk for homelessness in the near future, based on real-time spikes in housing assistance searches. Our data was consistent with the notion that the rising number of Los Angeles natives experiencing homelessness for the first time (a 16 percent increase between 2017 and 2018 according to Bloomberg) may have been foreshadowed by the high search density samples on Aunt Bertha’s platform.

An innovative way that health clinics, government agencies, and Community Based Organizations have helped to tackle homelessness is to identify at-risk clients as part of their intake process. Questions such as the following can be incorporated into intake applications:

  • Do you have any overdue utility or housing bills?
  • In the last three years, how many times have you been without steady housing?
  • How long has it been since you’ve had a steady place to live?
  • Are you worried about losing your housing in the future?
  • Do you have a proof of address, such as a signed lease or utility bill in your name?

A client who meets criteria would then be assisted by a social services navigator who would refer them to resources in the community for help with their situation.

For example, a health system with clinics in the San Fernando Valley may monitor housing assistance searches in real-time to determine whether additional care coordination is needed at those locations. Such interventions benefit healthcare organizations as social risk factors such as housing insecurity often interfere with a patient’s health. Screening questions may help identify relationships such as patients coming in for stress-related illness who also cannot keep up with rising rents. A screening questionnaire would identify such at-risk patients and direct them towards help to pay for housing before further detriment to their well being.