What our shelter software doesn't know about childhood

Our intake forms start at age 18. For most of the people we serve, the story started much earlier.

A numbered children's game drawn in sidewalk chalk on schoolyard asphalt.

Our intake form starts with a date of birth and a set of questions about the last twelve months. By the time someone answers them at our front desk, they are an adult, and the story has usually been running for years.

Last spring we started asking one more question, in conversation and only when it felt right: When was the first time you didn’t have a place to stay? The answers moved our attention back by decades.

Fig. 1
47%
of people surveyed first became homeless before age 25, including 14% under age 13.

Source: Everyone Counts 2024, Government of Canada.

That number isn’t ours. It comes from the national point-in-time count, and it matches what our frontline staff told us long before we had data to show it.1 It also says something uncomfortable about our software: almost everything it records happens after the moment that mattered most.

What we built

We didn’t build anything clever. We changed one question and one report, and we put both in the hands of the people who already know Community Members by name.

From the front line · Intake, 2026

What we built

An optional “first time” question, asked by staff in conversation. A monthly count of Community Members who first became homeless before 25, shared with our youth-serving partners.

What we learned

People answered more fully when staff explained why we were asking. The report gave us a reason to start a conversation with a school board partner.

Every adult in our shelter was once a child. Our tools should be able to see that far back.

The second number is harder to sit with. More than half of young people experiencing homelessness in one national survey had been involved with child welfare.2

Fig. 2
57.8%
of homeless youth reported child welfare involvement.

Source: Child Welfare and Youth Homelessness in Canada, Canadian Observatory on Homelessness, 2017.

What failed

Our first version put the childhood question on a self-serve tablet in the lobby. Fewer people answered, and two Community Members told staff it felt like a test. We pulled it after three weeks.

What we changed: the question only gets asked inside a relationship, by a person, with an explanation.

An open, blank notebook with a pen and pencils on a wooden desk.

Noticing earlier

Prevention isn’t a program a shelter can run alone. It’s a habit of attention we share with schools, youth agencies and child welfare partners, and our data only helps if it travels to them.

I come to this as a Christian. Following the Chalmers Center, I understand poverty as broken relationships with God, self, others and creation.3 A child who loses a home usually loses people first. Our tools should help us see that, and help someone show up sooner.

Sources

  1. Everyone Counts 2024, Government of Canada. ↩

  2. Child Welfare and Youth Homelessness in Canada, Canadian Observatory on Homelessness, 2017. ↩

  3. Steve Corbett and Brian Fikkert, When Helping Hurts, Chalmers Center. ↩

Chad Verity

Chad builds frontline technology at Hope Mission in Edmonton, Alberta, and writes about what the work has taught, including what failed.

Views are my own and don't necessarily represent Hope Mission.

Get the next one in Upstream

One letter a month on building technology for people experiencing homelessness.