The tool everyone used wasn't working for everyone.
The VI-SPDAT — the Vulnerability Index-Service Prioritization Decision Assistance Tool — was the standard assessment instrument used by CoCs across the country, including TX-601. It produced a vulnerability score that was supposed to determine who got prioritized for housing. But when we looked at our own outcome data from June 2021 to May 2022, we found a troubling pattern: the households being housed did not reflect the demographics of the households experiencing homelessness.
The data raised a hard question. Was our assessment tool measuring genuine need — or was it measuring something else? Something correlated with race, trauma history, or the ability to present well in a structured interview? The VI-SPDAT wasn't broken in the sense that it stopped working. It was producing outcomes that didn't match our values or our community's actual population.
HUD's own guidance on Coordinated Entry made clear that communities should prioritize based on risk of continued homelessness and severity of need, using all available data. The VI-SPDAT, in practice, wasn't reliably producing that result in Tarrant County. The decision to build something locally was grounded in that evidence.
Built here, for here — by the people doing the work.
The Housing Assessment Tool was designed by and for the TX-601 community. Not adapted from a national template. Not purchased from a vendor. Built from practitioner knowledge, client feedback, and local outcome data — with one purpose: identify the most appropriate housing intervention for each household, surface system-wide resource gaps, and restore professional judgment to a process that had become over-mechanized.
The HAT is structured across six sections, each capturing a distinct dimension of a household's situation:
Assessment date, location, type, level, household type, and alternative housing needs.
A deliberate departure from deficit-only tools — the HAT documents what households bring alongside the barriers they face.
Ability and willingness to work; assessed capacity to maintain housing after 12–24 months without intensive support.
Disabling conditions and the level of ongoing support required to sustain housing stability.
Prior housing instability, system involvement, and longitudinal factors relevant to prioritization.
What the client actually wants — not just what the system has available. A field that rarely existed before.
"Assessors also documented client notes — qualitative observations that captured context no structured field could hold. That narrative layer was intentional. It returns professional discretion and problem-solving to a process that had become a scoring exercise."
1,998 assessments. A clearer picture of who we were really serving.
In the first seven months of HAT deployment, 1,998 assessments were completed across TX-601. For the first time, the CoC had population-level data disaggregated by household type — single adults, families with children, multi-adult households, and transition-age youth — each showing distinct patterns in intervention needs, barriers, and housing preferences.
The additional housing challenges documented across all HATs revealed the complexity of the population the system was serving:
| Challenge | % of All HATs |
|---|---|
| Four or more emergency room visits in the past year | 28% |
| Survivor of intimate partner violence | 24% |
| Survivor of family violence, sexual violence, or sex trafficking | 24% |
| One or more stays in a substance abuse treatment facility (lifetime) | 21% |
| Household member with a disabling chronic health condition | 17% |
| One or more stays in another type of residential facility | 17% |
| Acute care need at time of assessment | 14% |
| Intellectual or developmental disorder (IDD) | 8% |
This data didn't just describe the population — it immediately identified where system resources were mismatched to actual need, and where new resource development should be targeted.
What communities working on assessment reform should know.
- 1Assessment tools are not neutral. Every tool encodes assumptions about what vulnerability means and whose knowledge is authoritative. If your outcomes don't reflect your population, your tool is part of the problem.
- 2Local design requires local investment. The HAT worked because it was built by operations leadership, data staff, and outreach practitioners together — not handed down from above or imported from elsewhere.
- 3Strengths belong in assessment. A tool that only documents deficits produces a deficit-oriented system. Capturing what households bring to their housing situation changes how practitioners engage and how matches get made.
- 4Scoring and intervention-matching are different functions. A vulnerability score tells you how urgent someone's situation is. It doesn't tell you what kind of resource they need. The HAT was designed to do both.
- 5Population-level data is a resource development tool. The disaggregated HAT data gave the CoC, for the first time, the ability to make the case for specific resources tied to documented population need — not assumptions.