A Tale of Three Homelessness Dashboards

The surest sign that someone new is in charge of the response to the homelessness crisis is that they publish a new “dashboard.” Both Seattle Mayor Bruce Harrell and the King County Regional Homeless Authority (KCRHA) have recently published new dashboards, each presenting their own unique spin on the crisis. The mere fact that Seattle and the KCRHA (which is largely funded by the City of Seattle) can’t even agree on a basic set of metrics for how we should be talking about homelessness speaks volumes about the state of debate on the issues and how deeply politicized it has become.

The Mayor’s new “One Seattle” homelessness dashboard (Harrell never misses an opportunity to rebrand something as
“One Seattle”) doubles as a pitch for his plan, making it challenging at times to discern the hard data from the aspirational plans. There is some data included that one could categorize as a “problem statement,” including the estimated number of tents and RVs in the city, the racial and ethnic composition of the homeless population, a breakdown of 911 calls related to homeless encampments,the number of shelter offers by city-funded staff and programs, and the combined number of emergency shelter and permanent supporting housing “PSH” units compared to Harrell’s promise of 2,000 by the end of 2022. It also includes a breakdown of the city’s annual budget for responding to the homelessness crisis (most of which now goes through the KCRHA).

The KCRHA dashboard is broken out by categories:

  • Counts of the number of households experiencing homelessness in King County, and the number of homeless households receiving services including shelter, housing referrals, job placement, and mental health and substance abuse treatment.
  • Crisis response: the number of households entering and exiting from the homeless system.
  • System performance: the number of households who are permanently housed, the average length of stay in temporary shelter, what percent of those in the system return to homelessness, and the utilization rate of the capacity in the system.
  • Rapid Rehousing program performance: the move-in-rate and time, the number of days assistance is provided, the percentage who are permanently housed, and the percentage that return to homelessness.

Last year the KCRHA rightly pointed out that the annual “Point in Time” count was a mess and didn’t provide accurate, useful numbers to help us quantify the scale of the homelessness crisis. Last December it issued a white paper with an updated methodology for counting the homeless, which integrated data from multiple sources to capture additional people who are accessing services but aren’t registered in the central HMIS tracking system. Unfortunately, the City of Seattle’s dashboard misstates the KCRHA’s count — and in fact the KCRHA itself sometimes misstates its own estimate.

The original white paper, which is very much worth reading, says that about 40,800 people in King County experienced homelessness at some point in time in 2020. The Seattle homelessness dashboard shortens this to “there are 40,000+ people experiencing homelessness in King County.”

This phrasing makes it sound like right now (or at any given moment in time) there are 40,000 homeless people in King County, which is a very different conclusion that is not supported by KCRHA’s analysis, and a significant overstatement. In fact, KCRHA also can’t seem to keep its own positioning straight, using the same language to overstate the number of homeless.

A closer look at the KCRHA data shows us why this is likely an overcount. In 2020, there were 14,300 entries into the homeless system, of which 73% were newly homeless in the past 24 months. There were also 16,800 exits from the homeless system. So there is clearly a lot of “churn”: people moving into, through and back out of the system.

How frequently does the homeless population churn? Again, according to the KCRHA data, the current average length of stay is 178 days, 54% exit to permanent housing and only 7% return to homelessness.

To be fair, the variance on length of time is large; according to a recent KCRHA budget presentation, the average length of stay is much longer for those in “tiny house” villages and transitional housing compared to those in emergency shelters.

Nevertheless, the average stay is less than six months, so RHA’s count of the number of people over the course of an entire year who are served by the homeless response system is much higher than the number of people who are homeless at any given time.


But the biggest problem with both of these dashboards is that neither answers the central question in defining the homeless response: how much service capacity do we need to provide in order to meet the need? How many emergency shelter beds? How many permanent supportive housing and transitional housing units? How many beds in mental health and substance abuse treatment programs? How much money in rapid rehousing programs? Knowing that over 40,000 people experience homelessness over the course of a year doesn’t tell us what we need to know in order to do real capacity planning; for that, we need to calculate the maximum number of people the system would need to serve simultaneously.

A third dashboard effort, led by Jeff Keenan with the support of the Third Door Coalition, is attempting to provide that “gap analysis,” beginning with a focus on permanent supportive housing (PSH): the recommended service for chronically homeless individuals who may never be able to live independently again. Jeff has been generating monthly reports that gather data on operational PSH units, new units in the pipeline, and the gap of additional units that would need to be created to fully meet the need for the estimated number of chronically homeless. He also estimates the capital costs to build out the additional PSH units, and the total ongoing operational costs for PSH units.

The easy part of modeling this is recognizing that there is almost no “churn”; PSH is intended to be a long-term housing+services solution for a targeted set of people who need that intervention, and many of its clients won’t ever “exit” to other forms of housing. The hard part, as Jeff has found, is tracking down information on the projects in the pipeline. And the scary part of what he has put together is the revelation there is essentially no new PSH in the pipeline beyond early 2023, despite a gap of almost 3,000 units.

To-date, there is no equivalent analysis for any of the other services targeted at the homeless population, and in particular the two we know are in desperately short supply: mental health and substance abuse services. This is by design: the government-published dashboards are political tools intended to make you believe that they are doing something meaningful. If they actually quantified the problem, then we might hold them responsible for putting forth a plan to solve it. Instead, to use Mayor Harrell’s favorite metaphor, they only explain to us how they will get the next first down, rather than how they will move the ball downfield and score a touchdown. Even overstating the total number of people experiencing homelessness serves their purpose, as it makes the problem seem so scary-large that any arbitrary amount of resources put towards homelessness must be a Good Thing; they can promote a budget request without an accompanying plan. Our elected officials’ words tell us that they want to solve homelessness, but their actions — and their dashboards — tell us otherwise: they want to do just enough to take credit for doing something, without drawing attention to the fact that it’s a drop in the bucket compared to what is really needed. And they have neither the plan nor the political capital to truly solve homelessness.

In an ideal world, we would have:

  • A single dashboard for Seattle and King County that presents a consensus view of the homelessness crisis: the need, the existing resources devoted to addressing that need, and the “gap” between the two. It would address emergency shelter, permanent supportive housing, transitional housing, rapid rehousing, and mental health and substance abuse treatment;
  • A plan from elected officials for funding, building, and operating resources and services sufficient to meet the full need;
  • A timeline for executing on the plan;
  • A funding proposal.

We will never get such a plan, because it would be bigger than our officials believe they can sell to voters and taxpayers, in terms of both the funding requirement and also the required trust that our local government could successfully execute on it. Unfortunately, that also means that we won’t get a real dashboard; it would call attention to their failure to pull together a real plan.


I hope you found this article valuable. If you did, please take a moment to make a contribution to support my ongoing work. Thanks!

Liked it? Take a second to support Kevin Schofield on Patreon!