Measuring PSH Performance Beyond Exits: Stability, Fidelity, and Operational Indicators That Matter

January 16, 2026

Permanent Supportive Housing is often evaluated using a narrow set of measures: occupancy, exits, length of stay, service contacts, emergency department use and returns to homelessness. These indicators can tell funders something about program activity, but they reveal far less about whether the operating model is protecting housing stability, honoring tenant rights or responding effectively when a tenancy begins to weaken.

Blunt performance measures can also drive the wrong behavior. A program may appear successful because it records few evictions, while tenants experience repeated lease warnings, avoidable crises, disengagement, unsafe property conditions or informal pressure to comply with services. Another program may report strong service-contact numbers even though those contacts are repetitive, poorly coordinated or disconnected from the issues placing housing at risk.

Effective performance frameworks must strengthen tenancy sustainment and housing stabilization while protecting the fidelity of Permanent Supportive Housing operations. The aim is not simply to produce more indicators. It is to make emerging risk visible, test whether Housing First principles remain active in daily practice and give leaders evidence they can use to improve the service.

The Housing Stability, Homelessness & Supportive Housing Knowledge Hub brings together wider guidance on tenancy sustainment, eviction prevention, coordinated entry, landlord relationships, housing-health integration and supportive housing operations. Within that wider system, PSH performance measurement should function as an early-warning and accountability structure rather than a retrospective reporting exercise.

A mature framework therefore asks more than whether someone remained housed. It examines whether the tenancy is becoming more secure, whether emerging difficulties receive a timely response, whether engagement remains voluntary, whether accommodations are considered, whether landlord concerns are managed proportionately and whether the provider can explain how operational decisions contributed to the outcome.

Why Traditional PSH Metrics Are Not Enough

Exit-focused measures tend to divide performance into two conditions: housed or no longer housed. Real tenancy stability is more complex. A person may technically remain housed while living through repeated utility shutoff threats, neighbor conflict, untreated health deterioration, rent arrears, property damage, isolation, exploitation or increasing avoidance of support.

Conversely, an exit does not always represent program failure. A tenant may move voluntarily to more independent housing, reunify with family, transfer to a more suitable accessible apartment or choose a different community. The meaning of the outcome depends on the circumstances, the person's preferences, the quality of transition planning and whether housing continuity was protected.

Traditional metrics also tend to emphasize events that are easy to count rather than the conditions that providers can influence. Evictions are measurable. The quality of early intervention is harder to capture. Service contacts are measurable. Whether those contacts address the tenant's own priorities is less visible. Emergency department visits are measurable. Whether the provider recognized and responded to deterioration before the visit requires stronger operational data.

A more useful framework separates:

final outcomes, such as sustained housing, planned move-on or return to homelessness;
leading indicators, such as arrears, missed contacts, complaints, isolation or repeated crisis use;
practice-fidelity indicators, such as voluntary engagement, accommodation and least coercive response;
operational reliability indicators, such as response time, action closure and property coordination;
tenant-experience indicators, such as trust, choice, safety and influence over support;
equity indicators, showing whether outcomes and interventions differ across populations; and
system-impact indicators, including emergency utilization, shelter returns and avoided crisis escalation.

This wider structure aligns measurement with outcomes in housing stability programs. It helps leaders distinguish between a service that records activity and one that reliably protects long-term housing.

The Central Measurement Question: Is the Tenancy Becoming More or Less Stable?

A PSH dashboard should help leaders answer a practical question: is the tenancy becoming more stable, remaining fragile or moving toward breakdown?

That requires indicators across several dimensions:

financial stability;
property and lease stability;
physical and behavioral health;
engagement and trust;
community relationships;
safety and exploitation risk;
service responsiveness;
landlord and property-management coordination;
tenant choice and rights; and
availability of flexible support when circumstances change.

No individual indicator should determine the provider's response. A missed visit, late rent payment or neighbor complaint may be isolated. The value comes from pattern recognition. Several low-level indicators occurring together may show that a tenancy requires earlier support even though no formal lease action has begun.

This is where PSH measurement becomes part of eviction prevention and early warning. The provider is not waiting for a notice to quit or formal eviction filing before recognizing instability. It is using operational intelligence to intervene while more options remain available.

Housing First Fidelity Must Be Measured in Everyday Decisions

Housing First fidelity cannot be inferred from the program name. It has to be visible in routine practice.

Providers should examine whether:

housing remains separate from participation in treatment or support;
service engagement is voluntary and non-coercive;
tenants are not threatened with tenancy consequences for declining support;
substance use or mental-health symptoms do not automatically trigger housing loss;
staff distinguish lease requirements from service preferences;
reasonable accommodations are identified and considered;
restrictions are lawful, necessary and proportionate;
tenant choice is documented in support planning;
harm-reduction approaches are available; and
termination or transfer decisions receive senior review.

Fidelity erosion often begins informally. Staff may describe engagement as voluntary while repeatedly linking cooperation with continued housing. Property concerns may be managed through broad behavioral expectations that exceed the lease. Tenants may be told that declining treatment will make it difficult for the program to continue supporting them.

These practices may not appear in formal policies. They become visible through supervision, tenant feedback, complaints, case sampling, accommodation records and review of tenancy-risk decisions.

Where providers need to evaluate whether policy, evidence and operational controls are sufficiently developed for external review, the Regulatory Readiness Gap Analyzer can help identify gaps between stated practice and demonstrable assurance.

A Balanced PSH Performance Framework

A cornerstone performance model should bring together eight connected domains.

1. Housing stability and tenancy sustainment

This domain measures whether tenants remain housed and whether the conditions supporting long-term stability are strengthening.

Possible indicators include:

housing retention at 6, 12, 24 and 36 months;
planned and unplanned exits;
returns to homelessness;
rent arrears and repayment-plan success;
lease-warning frequency;
eviction filings and outcomes;
emergency transfers;
temporary absences affecting tenancy;
utility-disconnection risk;
successful accommodation interventions; and
tenancies stabilized after an identified risk trigger.

2. Housing First and rights-based fidelity

This domain tests whether the program preserves voluntary support, tenant choice, accommodation and separation between housing and treatment.

Possible indicators include:

cases reviewed for potential coercion;
service refusal without tenancy consequence;
reasonable-accommodation requests and decisions;
tenant complaints involving rights or choice;
senior review of proposed termination;
use of harm-reduction planning;
documented tenant preferences;
appeals or grievance outcomes;
lease enforcement consistent with other tenants; and
staff supervision focused on fidelity drift.

3. Service responsiveness and engagement

Contact volume alone is weak evidence. This domain examines whether the service responds to changing need and sustains a trusting relationship without becoming intrusive.

Possible indicators include:

time from referral to first meaningful contact;
response time after a housing-risk alert;
successful re-engagement after missed contact;
tenant-defined goals progressed;
outreach intensity matched to risk;
follow-up after hospitalization or crisis;
availability outside normal business hours;
peer-support involvement;
language and accessibility support; and
tenant-reported trust in the support team.

4. Property and landlord coordination

Housing stability depends on how effectively support providers, landlords and property managers manage concerns without confusing their roles.

Possible indicators include:

maintenance-response time;
habitability concerns unresolved;
complaints routed through the agreed pathway;
landlord contact before formal lease action;
support responses to property concerns;
repeat damage or neighbor-conflict patterns;
use of landlord risk-mitigation resources;
disputes resolved without tenancy loss;
clarity between property and service responsibilities; and
tenant participation in resolving concerns.

5. Health, safety and crisis prevention

This domain measures whether the service recognizes deterioration and responds before it becomes avoidable emergency utilization or tenancy failure.

Possible indicators include:

emergency department use;
hospital admissions;
behavioral-health crisis contacts;
overdose events and reversals;
police or emergency-service involvement;
adult protective services referrals;
repeat crisis presentations;
post-crisis follow-up completion;
safety-plan review; and
tenancy stabilization following crisis.

6. Tenant experience, participation and quality of life

A tenant can remain housed while feeling unsafe, unheard or controlled. Experience measures test whether the model delivers meaningful security and autonomy.

Possible indicators include:

feeling safe at home;
control over daily life;
privacy and dignity;
trust in staff;
influence over support planning;
community connection;
loneliness and social isolation;
access to meaningful activity;
satisfaction with property conditions;
confidence in complaints routes; and
perceived risk of losing housing.

7. Equity and access

Aggregate results can hide unequal access, enforcement and outcomes. Providers should examine whether particular groups experience more exclusions, lease actions, crisis use, disengagement or involuntary exits.

Analysis may consider:

race and ethnicity;
gender;
age;
disability;
language;
sexual orientation and gender identity where appropriate and lawfully collected;
veteran status;
justice-system history;
behavioral-health need;
substance use;
family status;
geography; and
housing-site or landlord.

This should connect with health inequities and access barriers. A PSH program may report strong average retention while one population experiences substantially higher lease enforcement or lower access to accommodation.

8. Organizational and system reliability

This domain assesses whether the provider has the workforce, governance, data and partner capacity required to sustain the model.

Possible indicators include:

caseload size and acuity;
staff vacancies and turnover;
supervision compliance;
response coverage;
open safeguarding or quality actions;
unresolved landlord escalations;
data completeness;
corrective-action closure;
partner-response delays;
housing inventory and vacancy;
referral-to-housing time; and
contract or funding constraints affecting delivery.

Operational Example 1: A Stability Index That Captures Near Misses

What happens in day-to-day delivery

A PSH provider finds that quarterly eviction figures remain low, but unplanned exits and emergency transfers are increasing. Case reviews show that warning signs were visible months earlier: missed rent payments, neighbor complaints, declining contact, repeated emergency calls and unresolved maintenance concerns.

The provider creates a tenancy-stability index using a small set of weighted indicators:

new or increasing rent arrears;
formal or repeated informal lease concerns;
three or more failed contact attempts;
repeat emergency or crisis use;
significant change in health or functioning;
property damage or habitability concerns;
reported exploitation or safety concerns;
increasing isolation;
landlord request for urgent intervention; and
tenant request for transfer or indication that they no longer feel safe.

The score does not automate a decision. It places the tenancy into a routine, elevated or urgent review category. The assigned worker and supervisor review the pattern with the tenant, identify which factors matter and agree proportionate action.

Potential responses include benefits support, repayment planning, mediation, increased voluntary contact, accommodation review, health coordination, safety planning, housekeeping assistance, peer support, harm-reduction input or a property repair escalation.

Why the practice exists

Most tenancy breakdowns are preceded by smaller operational signals. Without a structured way to combine those signals, each concern may appear manageable in isolation until the cumulative position becomes much harder to stabilize.

What goes wrong if it is absent

Leaders are surprised by eviction filings and unplanned exits. Staff respond late, landlords lose confidence and the provider cannot show whether preventive options were explored before the tenancy failed.

What observable outcome it produces

The provider identifies more high-risk tenancies before formal lease action, increases the proportion stabilized through early intervention and reduces emergency transfers and preventable exits.

Required fields: stability indicators present, date identified, tenant perspective, current risk level, action owner, landlord or property coordination required, accommodation considered, review date and outcome.

Cannot proceed to formal tenancy-termination recommendation without: senior review of the stabilization actions attempted, tenant rights, accommodation duties, proportionality and available alternatives.

Auditable validation: records show that emerging indicators triggered review, actions were completed and the provider assessed whether those actions improved stability.

Providers can bring these measures into a structured oversight view using the Quality Dashboard Builder, combining final outcomes with early-warning, fidelity, equity and operational indicators.

Avoid Turning Early-Warning Metrics Into Surveillance

Early-warning systems must be designed carefully. Their purpose is to offer support earlier, not to increase monitoring, restrict tenant choice or create informal pathways toward eviction.

Providers should ensure that:

indicators trigger human review rather than automatic enforcement;
the tenant's account is included;
information is limited to what is necessary;
staff distinguish risk from lifestyle difference;
cultural and disability-related factors are considered;
support remains voluntary unless an immediate legal or safety duty applies;
the system does not penalize tenants for using crisis services;
landlord information is tested rather than accepted uncritically;
scores are reviewed for bias; and
tenants can challenge inaccurate information.

This reflects trust, transparency and ethical data use. A technically sophisticated performance system can still undermine Housing First if tenants experience it as surveillance or if indicators become evidence for punitive action rather than an offer of support.

Operational Example 2: Fidelity Monitoring Embedded in Supervision

What happens in day-to-day delivery

A provider reviews several unplanned exits and finds no single policy breach. However, case notes show gradual fidelity drift. Staff repeatedly described support as voluntary while telling tenants that continued refusal would make housing difficult to sustain. Some property complaints were managed through additional behavioral rules that were not part of the lease.

The provider introduces a monthly supervision sample focused on Housing First fidelity. Supervisors review:

whether housing and service participation were kept separate;
whether tenant refusal was respected;
whether outreach remained proportionate;
whether lease obligations were distinguished from staff preferences;
whether accommodations were considered;
whether language in records was neutral and rights-respecting;
whether harm-reduction options were offered;
whether the tenant's own goals were visible; and
whether proposed restrictions or termination received appropriate review.

Supervisors use examples from live cases for reflective discussion rather than treating fidelity as an annual training topic. Repeated themes are reported to quality governance.

Why the practice exists

Fidelity erosion is usually gradual. It develops through pressure, frustration, staff turnover, landlord demands and attempts to manage risk. Without routine supervision, the organization may not see the drift until complaints or tenancy losses occur.

What goes wrong if it is absent

The program moves toward a compliance-based model while continuing to describe itself as Housing First. Tenants experience coercion, disengagement rises and lease enforcement becomes entangled with service participation.

What observable outcome it produces

Records show clearer separation between tenancy and treatment, more accommodation consideration, fewer coercive statements and stronger tenant-reported trust.

Required fields: case sampled, fidelity question reviewed, concern identified, staff reflection, corrective guidance, accommodation issue, follow-up date and governance escalation where required.

Cannot proceed with a restrictive or tenancy-ending response without: evidence that rights, lease requirements, accommodation, voluntary engagement and less restrictive options were reviewed.

Auditable validation: repeated supervision sampling demonstrates whether fidelity concerns reduce and whether corrective guidance changes recorded practice.

Where a provider identifies significant drift, the Quality Improvement Action Plan Builder can translate findings into owned actions, milestones, evidence requirements and effectiveness checks.

Measure Engagement Without Rewarding Intrusion

Service-contact targets can distort practice if more contact is automatically treated as better performance. Some tenants require intensive engagement during instability. Others may be stable and prefer limited support. A high number of contacts may show responsive service, repeated failed outreach or an inability to resolve the same issue.

Stronger measures examine:

whether the tenant agrees with the contact plan;
whether contact intensity reflects current need;
whether the purpose of contact is clear;
whether action follows the contact;
whether repeated outreach is becoming intrusive;
whether staff use the tenant's preferred communication method;
whether missed contact triggers support rather than punishment;
whether engagement increases after a risk trigger;
whether contact decreases appropriately as stability improves; and
whether the tenant feels listened to and respected.

Engagement quality is more meaningful than activity volume. The metric should reflect whether the relationship helps the tenant sustain housing on terms that preserve dignity and autonomy.

Operational Example 3: Measuring Landlord and Property Coordination Without Shifting Responsibility Onto Tenants

What happens in day-to-day delivery

A scattered-site PSH provider receives frequent landlord complaints about noise, visitors, housekeeping and delayed rent. Internal reporting records the number of complaints but does not show how quickly concerns were reviewed, whether the tenant's perspective was obtained or whether property-management issues contributed to the dispute.

The provider introduces a landlord and property coordination register. Each concern records:

date and source of the concern;
lease provision potentially involved;
tenant perspective;
property or maintenance factors;
support needs identified;
reasonable accommodation considered;
immediate risk level;
response owner;
agreed action;
landlord follow-up date;
tenant follow-up date;
whether formal lease action was initiated;
resolution achieved; and
learning for future prevention.

Monthly review distinguishes isolated concerns from repeated patterns. Leaders can see whether particular landlords, buildings, teams or tenant groups experience higher rates of conflict or formal enforcement.

Why the practice exists

Landlord complaints can become the first step toward housing loss. However, not every complaint reflects tenant behavior alone. Poor repairs, unclear communication, inaccessible buildings, inconsistent rule enforcement and weak property-management practice may contribute to the issue.

What goes wrong if it is absent

Support teams react to landlord concerns without checking the lease, the tenant's account or accommodation needs. Informal warnings accumulate, tenants lose trust and formal action begins before the provider has explored proportionate alternatives.

What observable outcome it produces

The provider resolves more concerns before formal lease action, identifies recurring landlord or property issues and improves clarity about who owns each response.

Required fields: lease issue, tenant account, property contribution, accommodation review, response owner, action deadline, landlord communication and resolution status.

Cannot proceed to formal escalation without: confirmation that the concern has been checked against the lease, the tenant has had an opportunity to respond and accommodation or support alternatives have been considered.

Auditable validation: records show whether early coordination reduced repeat complaints, delayed formal action or prevented avoidable housing loss.

This supports stronger landlord engagement and risk mitigation. Good landlord relationships are not measured only by satisfaction. They are evidenced through timely communication, clear boundaries, proportionate problem-solving and reduced preventable tenancy failure.

Measure Property Conditions as a Housing Stability Issue

PSH performance frameworks often emphasize tenant behavior while giving less attention to property conditions. Habitability, accessibility, pest control, heating, water, security, noise and repair delays can directly affect health, engagement and tenancy stability.

Providers should track:

urgent repair response time;
routine repair completion time;
repeat repairs for the same issue;
habitability complaints;
accessibility modifications requested and completed;
pest or environmental-health concerns;
security incidents;
temporary relocation caused by property failure;
tenant satisfaction with repair communication;
landlord or property-manager responsiveness;
unresolved issues beyond target; and
tenancy instability associated with property conditions.

Property failures should not disappear into a maintenance system disconnected from support records. Where conditions affect health, safety or housing retention, the concern should become visible in the PSH assurance framework.

Operational Example 4: Connecting Property Defects With Health and Tenancy Risk

What happens in day-to-day delivery

A tenant with chronic respiratory illness reports recurring mold and ventilation problems. Maintenance tickets show repeated visits, but the PSH dashboard records only that each ticket was closed. The tenant's health worsens, appointments are missed and the person begins discussing leaving the apartment.

The provider changes its measurement approach. Repair closure is no longer treated as sufficient where a defect recurs or affects health, safety, accessibility or housing confidence.

The revised process requires:

confirmation of the underlying cause;
tenant verification that the issue is resolved;
health or accessibility impact;
temporary mitigation;
repeat occurrence tracking;
property-owner escalation;
reasonable accommodation consideration;
relocation review where necessary; and
post-resolution follow-up.

Why the practice exists

A closed maintenance ticket may record activity rather than resolution. Repeated property failures can undermine health, trust and willingness to remain housed.

What goes wrong if it is absent

Leaders receive false reassurance from high ticket-closure rates while tenants continue living with unresolved conditions. Health deterioration and tenancy dissatisfaction are treated as separate issues.

What observable outcome it produces

Repeat defects become visible, property escalation occurs earlier and the provider can show whether repair action restored safe and sustainable housing.

Required fields: defect, health or accessibility impact, interim control, responsible property contact, target date, tenant confirmation and recurrence status.

Cannot close without: evidence that the underlying condition has been addressed or an approved interim and escalation plan remains active.

Auditable validation: repeat repairs, tenant feedback and health-related escalation reduce after corrective action.

Measure Crisis Prevention, Not Only Crisis Use

Emergency department visits, psychiatric admissions, overdoses, police contacts and shelter returns are important indicators. However, final utilization counts do not show whether the provider recognized risk, offered alternatives or completed follow-up.

A stronger framework measures the full prevention and response pathway:

early warning sign identified;
risk reviewed;
tenant contacted;
preferred crisis response confirmed;
clinical or behavioral-health partner involved;
harm-reduction action offered;
after-hours support available;
emergency response avoided where safe;
post-crisis follow-up completed;
housing impact assessed;
support plan updated; and
repeat crisis pattern reviewed.

This links PSH performance to preventing system bounce-back. The aim is not to discourage appropriate emergency care. It is to determine whether repeated crisis use reflects a gap in housing support, health coordination, after-hours response or service accessibility.

Operational Example 5: A Post-Crisis Housing Stabilization Review

What happens in day-to-day delivery

A tenant returns home after a psychiatric hospitalization. The hospital discharge is recorded, but the PSH service has no standard process for assessing the effect on the tenancy. Medication has changed, rent remains unpaid and the landlord has raised concerns about property damage that occurred before admission.

The provider introduces a post-crisis housing stabilization review completed within a defined period after hospitalization, overdose, police involvement, emergency shelter use or another significant crisis.

The review covers:

current housing safety;
tenant's understanding of what happened;
changes in health or medication;
rent and benefit status;
property or neighbor concerns;
support preferences;
follow-up appointments;
crisis and safety-plan changes;
landlord communication;
reasonable accommodation;
re-entry or stabilization support; and
conditions that would trigger further escalation.

Why the practice exists

Crisis resolution in one system does not automatically restore housing stability. The person may return home to the same conditions that contributed to the crisis, with additional financial, health or property pressures.

What goes wrong if it is absent

The provider records the hospital discharge but misses the tenancy consequences. Follow-up fragments, lease concerns progress and the person may return quickly to crisis care or lose housing.

What observable outcome it produces

Post-crisis actions close more reliably, repeat emergency use declines for targeted tenants and tenancy risks are addressed alongside clinical follow-up.

Required fields: crisis type, discharge date, housing risk, medication or health changes, rent status, property concerns, tenant priorities, follow-up owner and review date.

Cannot close the episode without: confirmation that housing, health and practical follow-up actions have named owners.

Auditable validation: records show that post-crisis review occurred, required actions were completed and repeat crisis or tenancy escalation was monitored.

Tenant Experience Must Influence the Dashboard

Tenant feedback should not sit in a separate annual survey that has little effect on operational decisions. Experience data should be connected with tenancy stability, property quality, staff practice and service responsiveness.

Useful questions include:

Do you feel safe in your home?
Do staff respect your choices?
Do you understand your lease and rights?
Do you know who to contact about repairs?
Do staff respond when you ask for help?
Do you feel pressured to accept services?
Can you raise concerns without fear of losing housing?
Are communication and information accessible?
Do you feel involved in decisions about your support?
Do you feel connected to your community?
Are there conditions making it harder to remain housed?
What would make your housing feel more secure?

Providers should offer several feedback routes, including accessible surveys, conversations with people independent of the immediate support relationship, tenant meetings, peer-led engagement, complaint review and focused interviews after significant events.

Qualitative evidence can be organized through story, case studies and qualitative evidence. Individual stories should not replace quantitative assurance, but they can explain why a measure changed, how the service affected the tenant and where apparently positive data hides a weaker experience.

Operational Example 6: Turning Tenant Feedback Into a Measurable Improvement Loop

What happens in day-to-day delivery

Annual survey results show generally high satisfaction, but complaint records and tenant meetings reveal recurring concern about staff entering apartments without enough notice. The broad survey score has hidden a specific issue involving privacy, trust and role boundaries.

The provider introduces a focused improvement cycle. It reviews policy, lease provisions, emergency-entry rules, staff practice and documentation. Tenants help define what appropriate notice and communication should look like.

The provider then measures:

entries without standard notice;
reason for entry;
emergency versus routine entry;
tenant consent or notification;
complaints involving privacy;
staff understanding of access rules;
supervision findings; and
tenant confidence after the change.

Why the practice exists

High-level satisfaction measures can hide specific rights or practice concerns. Qualitative evidence often reveals the issue before aggregate scores move.

What goes wrong if it is absent

The provider reports strong satisfaction while tenants experience repeated privacy intrusions. Trust declines and engagement becomes more difficult.

What observable outcome it produces

Unplanned entries decrease, staff demonstrate clearer understanding and tenant feedback shows improved confidence in privacy and communication.

Required fields: feedback theme, evidence sources, affected policy or practice, action owner, tenant involvement, baseline, target and follow-up result.

Cannot close without: evidence that the change was tested with tenants and monitored after implementation.

Auditable validation: complaint patterns, audit results and tenant feedback show whether the issue improved.

The Community Impact Report Builder can help providers bring together tenant experience, housing outcomes, community participation, equity and avoided system use in a structured external narrative.

Measure Complaints as Quality Signals

Low complaint numbers do not necessarily indicate a strong service. They may show that tenants do not know how to complain, do not trust the process or fear consequences.

Complaint assurance should consider:

complaints by theme;
complaints by site, landlord or team;
accessibility of the complaints route;
time to acknowledgment and resolution;
tenant satisfaction with the process;
repeat complaints;
retaliation or fear concerns;
rights and accommodation themes;
property-management issues;
staff conduct or boundary concerns;
actions completed; and
evidence of service change.

This aligns with complaints as quality signals. The important measure is not whether the organization can minimize complaints. It is whether concerns become visible, are handled fairly and lead to learning.

Equity Analysis Must Go Beyond Overall Retention

A PSH provider may achieve a high overall housing-retention rate while producing unequal experiences and outcomes. Equity analysis should examine the full pathway from referral to long-term stability.

Useful equity questions include:

Who is referred and who is screened out?
Who waits longest for housing?
Who receives units in safer or better-resourced locations?
Who experiences more lease warnings?
Who receives accommodation review?
Who is more likely to be labeled noncompliant?
Who experiences police involvement?
Who has higher rates of involuntary exit?
Who receives intensive support?
Who reports lower trust or safety?
Who returns to homelessness?
Which landlords or buildings show unequal patterns?

Providers should interpret differences carefully. Small samples, incomplete demographic information and varying levels of need can affect results. However, uncertainty should prompt deeper review rather than justify ignoring a pattern.

Equity findings should connect with data-led equity planning. The aim is to move from disparity identification to owned operational action.

Operational Example 7: Identifying Unequal Lease Enforcement

What happens in day-to-day delivery

A provider reviews lease-warning data and finds that tenants with serious mental illness receive formal warnings more often than other tenants for noise, visitors and housekeeping concerns. Housing-retention rates remain similar, so the disparity has not appeared in headline outcomes.

The provider samples cases and examines:

the lease provision cited;
severity and frequency of the concern;
whether informal resolution was attempted;
whether disability-related need was considered;
whether accommodation was discussed;
differences between landlords;
staff language and assumptions;
tenant participation; and
final outcome.

The review finds inconsistent accommodation practice and substantial variation between property managers. The provider introduces a senior review threshold before repeated warnings progress toward termination.

Why the practice exists

Headline retention data can hide unequal exposure to enforcement, stress and risk. A tenancy may survive despite repeated avoidable escalation.

What goes wrong if it is absent

Disproportionate practice continues, trust declines and the provider remains unable to explain why one population experiences more formal action.

What observable outcome it produces

Accommodation consideration becomes more consistent, variation between landlords reduces and formal warnings are more clearly tied to evidenced lease concerns.

Required fields: population comparison, case-sample findings, landlord variation, accommodation evidence, corrective action, target and review date.

Cannot dismiss the disparity without: documented analysis of whether policy, landlord practice, staff judgment, accessibility or service design contributes to the difference.

Auditable validation: later reporting shows whether unequal warning and escalation patterns reduce.

Measure Access and Referral Pathway Reliability

PSH performance begins before move-in. Long waits, repeated document requests, inaccessible communication, unclear eligibility and weak coordination can cause people to disengage or remain homeless longer than necessary.

Access measures may include:

referral-to-contact time;
referral-to-eligibility decision time;
referral-to-housing time;
applications closed because contact was lost;
documentation barriers;
reasonable accommodation during application;
language support;
unit-offer acceptance and refusal reasons;
housing-location preferences;
people remaining in shelter or institutional settings while waiting;
referrals returned or rejected; and
equity differences across the pathway.

These measures support stronger coordinated entry and prioritization. A provider should be able to explain not only who entered housing, but where applicants were delayed or lost and what the system did to restore contact.

Measure the Quality of Move-In and Early Tenancy

The first 30, 60 and 90 days often determine whether a tenancy begins with trust and stability or immediate strain.

Early-tenancy indicators may include:

unit readiness at move-in;
furniture, utilities and essential supplies;
lease orientation completed accessibly;
tenant understanding of rights and responsibilities;
benefits and rent arrangements active;
health and medication continuity;
preferred contact plan agreed;
property and neighborhood orientation;
initial safety concerns;
first landlord or neighbor issue;
early missed contacts;
tenant-reported confidence; and
90-day housing stability.

Providers should avoid treating move-in as the endpoint of the referral pathway. It is the beginning of a high-risk transition that requires structured follow-through.

Operational Example 8: A Structured 90-Day Tenancy Stabilization Review

What happens in day-to-day delivery

A PSH provider finds that several unplanned exits occur within the first six months of tenancy. Case reviews show that move-in tasks were completed, but the service did not consistently reassess whether the tenant had settled, understood the lease, established essential routines or encountered barriers that were not visible during referral.

The provider introduces structured reviews at 30, 60 and 90 days. Each review is completed with the tenant and covers:

how safe and settled the tenant feels;
whether the apartment meets accessibility and practical needs;
rent, utilities and benefits;
understanding of lease rights and responsibilities;
contact preferences and support boundaries;
healthcare and medication continuity;
property or maintenance concerns;
relationships with neighbors and property staff;
community connection and isolation;
food, furnishings and essential household items;
emerging safety, exploitation or visitor concerns;
reasonable accommodation needs;
tenant-defined priorities; and
actions requiring follow-through.

Why the practice exists

Housing placement does not automatically produce housing stability. Early difficulties can remain hidden when providers focus on completed move-in tasks rather than the tenant's lived experience of settling into the home.

What goes wrong if it is absent

Benefits problems, unmet accessibility needs, landlord tension, loneliness and confusion about support arrangements can compound until the tenant disengages or the tenancy enters formal difficulty.

What observable outcome it produces

The provider identifies early tenancy risks sooner, resolves more practical barriers and reduces avoidable exits during the first year.

Required fields: review date, tenant perspective, housing confidence, lease understanding, financial position, property concerns, support preferences, accommodation needs, action owner and next review date.

Cannot close the early-tenancy pathway without: confirmation that outstanding housing, financial, health, property and support actions have been assigned and followed through.

Auditable validation: 30-, 60- and 90-day review completion is tracked alongside early lease warnings, crisis use and first-year retention.

Measure Workforce Capacity Against Tenant Need

PSH performance cannot be understood without workforce context. Deteriorating response times, missed follow-up and weak engagement may reflect individual practice, but they may also show that caseloads, vacancies, travel demands or supervision capacity no longer match tenant need.

Workforce measures should include:

caseload size by worker;
caseload acuity and complexity;
vacancy rate;
staff turnover;
use of temporary or agency staff;
supervision completion;
time spent on travel and administration;
after-hours coverage;
unallocated or partially allocated tenancies;
overdue reviews;
missed or delayed risk follow-up;
staff competence and training assurance;
worker continuity for tenants;
sickness and burnout indicators; and
time from vacancy to stable replacement.

These measures should not be interpreted separately from tenant outcomes. Leaders should examine whether higher caseloads are associated with slower responses, reduced engagement, more landlord escalation or increased crisis use.

This connects with workforce data and capacity planning. A provider should be able to show how staffing assumptions relate to the intensity and complexity of the housing-support model.

Operational Example 9: Linking Caseload Pressure to Housing Instability

What happens in day-to-day delivery

A multi-site PSH provider sees rising staff turnover and increasing unresolved landlord concerns. The headline caseload average remains within contract expectations, but some workers hold more high-risk tenancies, cover large geographic areas and provide substantial post-crisis follow-up.

The provider redesigns its caseload analysis. Instead of counting every tenancy equally, it considers:

current housing-risk level;
recent crisis or hospitalization;
frequency of landlord or property intervention;
behavioral and physical health complexity;
communication and accessibility needs;
travel burden;
new-tenancy intensity;
legal or accommodation activity;
required partner coordination; and
current level of tenant engagement.

The weighted view shows that several workers carry substantially higher operational demand despite similar numeric caseloads. Managers rebalance assignments, add temporary supervisory support and prioritize recruitment for the most pressured locality.

Why the practice exists

Numeric caseload limits can hide substantial variation in complexity. Equal numbers do not necessarily represent equal workload or equal risk.

What goes wrong if it is absent

High-risk tenancies receive delayed follow-up, staff become reactive and leaders attribute performance decline to individual workers rather than structural capacity mismatch.

What observable outcome it produces

Response times improve, overdue actions reduce and staff turnover begins to stabilize in the targeted teams.

Required fields: worker, numeric caseload, weighted demand, travel burden, high-risk tenancies, overdue actions, supervision needs and management response.

Cannot conclude that performance is an individual competence issue without: review of caseload demand, vacancies, supervision, systems and workload distribution.

Auditable validation: workforce changes are compared with response times, action closure, tenant experience and housing-stability indicators.

Providers seeking to model how workforce capacity, service demand and quality pressures may interact can use the Digital Twin Scenario Modeler to test alternative staffing, caseload and service-design assumptions.

Measure Partnership Reliability, Not Merely Partnership Activity

PSH depends on coordination with landlords, property managers, behavioral-health services, primary care, hospitals, benefits agencies, crisis teams, substance use services, legal support and community organizations.

Meeting counts and referral volumes do not show whether those partnerships work when a tenant needs help.

Partnership measures may include:

referral acceptance rate;
time to partner response;
closed-loop confirmation;
shared action completion;
failed or returned referrals;
information-sharing delays;
attendance at joint case reviews;
post-discharge coordination;
repeat escalation caused by partner delay;
availability of urgent consultation;
tenant experience of coordinated support;
unresolved responsibility disputes; and
service gaps affecting housing stability.

This is particularly important within housing-health partnerships and care integration. A referral is not successful because it was sent. It is successful when the receiving service accepts responsibility, the tenant can access support and the required action is completed.

Operational Example 10: Closed-Loop Behavioral Health Referral Assurance

What happens in day-to-day delivery

A tenant's mental health begins to deteriorate. The PSH worker sends a behavioral-health referral, records it as completed and assumes follow-up will occur. Several weeks later, the tenant enters crisis. Review shows that the referral was returned because information was incomplete, but no one in the housing team saw the message.

The provider introduces a closed-loop referral standard. Every priority referral must record:

reason for referral;
tenant consent and preferences;
receiving service;
date sent;
acceptance confirmed;
appointment or next action;
barrier identified;
responsible follow-up worker;
escalation date if no response;
tenant informed; and
outcome confirmed.

Why the practice exists

Referral transmission is not the same as care coordination. Without acknowledgment and follow-through, people can remain unsupported while each organization assumes another team is acting.

What goes wrong if it is absent

Referrals disappear, deterioration continues and the provider cannot explain whether access was secured or why escalation did not occur.

What observable outcome it produces

Referral acceptance becomes more visible, failed pathways are escalated earlier and fewer tenants remain without follow-up after an identified need.

Required fields: referral purpose, consent, receiving service, acceptance, appointment, owner, escalation threshold and outcome.

Cannot close the referral without: evidence of acceptance, an agreed alternative or documented escalation where access remains unavailable.

Auditable validation: referral closure, response time and subsequent crisis outcomes are reviewed together.

Measure Avoided Costs Carefully and Credibly

PSH can reduce shelter use, hospitalization, emergency care, incarceration and crisis-system demand. These system effects are important, but providers should avoid presenting every reduction as a directly attributable financial saving.

Strong value analysis distinguishes between:

actual cashable savings, where expenditure genuinely reduces;
avoided costs, where a likely future cost does not occur;
reduced utilization, where service use falls but fixed costs remain;
capacity released, where another person can use the resource;
cost shifting, where expenditure moves between agencies;
improved outcomes, which may justify investment even without immediate savings; and
long-term social value, such as improved safety, autonomy and community participation.

Providers should state assumptions clearly, avoid unsupported counterfactual claims and include the cost of delivering PSH when presenting return on investment.

This aligns with return on investment and value for money. Credible value evidence is more persuasive than an inflated claim that cannot withstand payer or funder review.

Operational Example 11: Building a Defensible Avoided-Utilization Analysis

What happens in day-to-day delivery

A provider wants to demonstrate the effect of PSH on emergency utilization. Initial analysis compares emergency department use during the year before housing with the year after move-in. Results show a substantial reduction, but leaders recognize that changes in health, insurance status, local service access and incomplete pre-housing records may affect interpretation.

The provider strengthens the methodology by documenting:

the population included;
the pre- and post-housing periods;
data-source completeness;
people excluded and why;
changes in health or eligibility;
whether utilization shifted to another service;
actual service costs where available;
limits on causal interpretation;
tenant outcomes alongside utilization; and
subgroup differences.

Why the practice exists

Simple before-and-after comparisons can be useful but may overstate direct causation. Transparent methodology protects credibility.

What goes wrong if it is absent

The provider publishes a large savings claim that funders cannot reproduce. Confidence in the wider performance framework declines.

What observable outcome it produces

The resulting report presents a more defensible account of reduced utilization, housing stability and wider community impact.

Required fields: cohort definition, observation period, data source, exclusions, assumptions, limitations, utilization result and housing outcome.

Cannot describe reduced utilization as a cashable saving without: evidence that expenditure genuinely reduced or capacity was released in a financially meaningful way.

Auditable validation: calculations can be reproduced from the defined source data and assumptions.

Build a Dashboard That Supports Decisions

A PSH dashboard should not display every available metric. It should organize information according to the decisions leaders and teams need to make.

A practical dashboard may contain five layers.

Layer 1: Executive outcomes

housing retention;
planned and unplanned exits;
returns to homelessness;
eviction filings;
emergency transfers;
tenant-reported safety and trust;
crisis utilization; and
equity differences.

Layer 2: Leading stability indicators

new arrears;
repeat complaints;
failed contact patterns;
property issues;
hospital discharge;
benefit interruption;
landlord escalation;
accommodation requests; and
tenant-reported insecurity.

Layer 3: Operational reliability

response time;
action closure;
30-, 60- and 90-day reviews;
post-crisis review;
closed-loop referrals;
repair resolution;
complaint handling;
case-review compliance; and
partner responsiveness.

Layer 4: Fidelity and rights

service refusal without tenancy consequence;
accommodation consideration;
termination reviews;
tenant complaints about coercion;
lease and service-role separation;
privacy concerns;
appeals; and
supervision findings.

Layer 5: Delivery capacity

vacancy and turnover;
weighted caseload;
supervision;
after-hours coverage;
overdue actions;
data completeness;
open improvement actions; and
contract or funding risks.

Every dashboard measure should have an owner, definition, data source, reporting frequency, tolerance, escalation threshold and expected management response.

Set Thresholds That Trigger Action

A dashboard becomes useful when leaders know what happens after a measure changes.

Examples of action thresholds include:

two consecutive months of rising arrears trigger financial-stability review;
repeat landlord complaints trigger supervisor case review;
post-crisis review below target triggers management escalation;
a disparity in lease warnings triggers equity sampling;
repair recurrence triggers property-owner escalation;
high weighted caseload triggers workforce rebalancing;
repeat unplanned exits from one building trigger service and landlord review;
declining tenant trust triggers focused engagement;
overdue corrective actions trigger executive oversight; and
data completeness below tolerance prevents external publication without qualification.

Thresholds should not create automatic punitive responses. They should activate review, professional judgment and proportionate intervention.

Use Statistical Context Without Losing Operational Meaning

Small PSH programs can experience substantial percentage changes from only one or two events. Leaders should therefore consider:

numerator and denominator;
rolling averages;
longer-term trends;
site and population differences;
seasonal variation;
data completeness;
changes in program size;
changes in tenant acuity;
outliers; and
qualitative explanation.

A change from one eviction to two is a 100% increase, but the operational interpretation depends on the circumstances. Conversely, a stable percentage can conceal a recurring pattern affecting the same population or landlord.

Data Quality Is Part of PSH Performance Assurance

Performance conclusions are only as reliable as the underlying information. Providers should define controls for:

consistent exit reasons;
planned versus unplanned moves;
returns to homelessness;
lease-warning categories;
accommodation records;
complaint themes;
service-contact purpose;
crisis events;
property and repair data;
tenant demographics;
landlord identifiers;
referral status;
duplicate records;
missing data; and
changes to measure definitions.

This connects with data collection and data quality. A provider should be able to explain how each material indicator is defined, captured, checked and corrected.

Operational Example 12: Correcting Misleading Exit Data

What happens in day-to-day delivery

A PSH provider reports a rising rate of voluntary exits. Record sampling finds that several tenants left after escalating landlord conflict, informal pressure or repeated unresolved property problems. Staff selected “voluntary move” because the tenant technically chose to leave.

The provider revises its exit taxonomy to distinguish:

planned move to greater independence;
planned transfer to more suitable housing;
tenant-led relocation for preference;
move following unresolved property conditions;
move during escalating lease action;
abandonment;
formal eviction;
institutional admission;
incarceration;
death;
return to homelessness; and
unknown or unconfirmed outcome.

Supervisors review complex exit coding before final submission.

Why the practice exists

Broad categories can make pressured or preventable exits appear positive. Accurate coding is essential for meaningful learning.

What goes wrong if it is absent

Leadership underestimates tenancy failure, improvement priorities remain hidden and external reports present an overly favorable picture.

What observable outcome it produces

Exit reporting becomes more transparent, root causes are easier to identify and corrective action targets the real pathways into housing loss.

Required fields: exit type, tenant choice, housing destination, lease status, property concerns, support actions, accommodation review and supervisor approval.

Cannot code an exit as positive move-on without: evidence that the move reflected informed tenant choice and resulted in suitable, stable housing.

Auditable validation: exit categories reconcile with case records, destination evidence and tenant circumstances.

Create an Operating Rhythm Around the Data

PSH measurement should operate at several levels.

Daily and weekly operational review

Teams review urgent tenancy risks, crisis events, landlord escalations, failed contact, hospital discharge and unresolved safety or property concerns.

Monthly service review

Managers review leading indicators, action closure, early-tenancy performance, caseload pressure, complaints, repair issues and emerging equity concerns.

Quarterly quality and governance review

Leaders examine trends, fidelity, tenant experience, workforce, partner performance, corrective action and system impact. They test whether improvement activity changed outcomes.

Annual strategic review

The organization reviews longer-term retention, returns to homelessness, contract expectations, housing inventory, landlord network health, funding sustainability and service-model development.

This reflects a disciplined dashboard operating rhythm and performance cadence. The purpose is to place the right evidence before the right decision-makers at the right time.

Operational Example 13: A Quarterly Learning Loop That Changes Delivery

What happens in day-to-day delivery

Quarterly analysis shows that repeat lease warnings are concentrated in two buildings. Staff initially attribute this to higher tenant complexity. Deeper review finds that the buildings share the same property manager, have slower repair response and use more formal communication than other sites.

The provider convenes a focused review with tenants, support staff, property representatives and leadership. It agrees actions covering:

repair escalation;
communication standards;
reasonable accommodation;
pre-warning case discussion;
tenant information;
staff supervision;
landlord training;
complaint routes; and
monthly monitoring.

Why the practice exists

Performance patterns require structured inquiry. Without it, organizations may attribute variation to tenants or frontline staff when the real cause is a building, landlord or operating-system issue.

What goes wrong if it is absent

The same pattern continues, formal enforcement rises and the provider produces reports without changing the conditions behind the result.

What observable outcome it produces

Repair response improves, lease warnings decline and tenants report clearer communication in the affected buildings.

Required fields: pattern identified, evidence reviewed, tenant input, root cause, actions, owners, deadlines, outcome measures and verification date.

Cannot close the improvement cycle without: evidence that the agreed actions were implemented and the relevant indicators were remeasured.

Auditable validation: before-and-after data demonstrates whether the intervention changed the identified pattern.

Distinguish Assurance From Performance Management

Performance management asks whether targets were met. Assurance asks whether leaders can trust the result and whether the underlying system remains safe, lawful and effective.

PSH assurance should therefore test:

whether measures are correctly defined;
whether data is complete;
whether positive outcomes conceal fidelity concerns;
whether known limitations are disclosed;
whether actions were completed;
whether actions improved the result;
whether disparities were examined;
whether tenant voice influenced decisions;
whether workforce capacity is sufficient;
whether landlord and partner responsibilities are clear;
whether risks are escalated; and
whether the board receives an honest picture.

The Governance Maturity Assessment can help organizations examine whether decision rights, assurance lines and board oversight are strong enough to govern complex housing performance information.

Build Evidence Packs for Funders and Regulators

A strong external evidence pack should connect performance claims with operational proof.

It may include:

approved performance framework;
measure definitions;
housing-retention trends;
exit analysis;
early-warning indicators;
fidelity audit findings;
tenant-experience evidence;
equity analysis;
workforce and caseload information;
property and landlord performance;
crisis and utilization analysis;
complaint themes;
corrective-action plans;
verification results;
case studies illustrating pathways;
data-quality checks;
known limitations; and
governance oversight records.

This aligns with evidence packs for funders and regulators. External audiences should be able to see not only the result, but how the provider knows the result is reliable and what it did when performance weakened.

Translate Performance Findings Into Corrective Action

Measurement only improves PSH when weak performance leads to owned, time-bound and verified action. A dashboard that repeatedly highlights the same concern without changing practice becomes a reporting burden rather than a management control.

Corrective action should begin when evidence shows:

housing-retention performance is deteriorating;
unplanned exits are concentrated in a site, landlord or population;
lease enforcement is inconsistent;
reasonable accommodations are delayed or poorly documented;
tenant trust or safety is declining;
property defects repeatedly affect housing stability;
post-crisis actions are not completed;
referrals are not closing reliably;
staffing capacity no longer matches tenant need;
complaint themes recur;
equity disparities remain unexplained;
fidelity audits identify coercive or compliance-led practice;
data quality prevents reliable oversight; or
funding or contract expectations are at risk.

A strong corrective-action record identifies the problem, evidence source, root cause, immediate control, longer-term action, accountable owner, deadline, expected result and verification method.

The Quality Improvement Action Plan Builder can support providers to convert audit, tenant feedback, performance variation and contract-monitoring findings into structured improvement plans with clear ownership and follow-through.

Operational Example 14: Corrective Action After Rising Unplanned Exits

What happens in day-to-day delivery

A provider's quarterly dashboard shows that unplanned exits have increased across one scattered-site program. Initial discussion focuses on tenant complexity, but a structured review identifies several contributing factors:

high staff turnover;
delayed assignment of new workers;
inconsistent landlord communication;
weak post-hospital follow-up;
limited after-hours support;
poor documentation of reasonable accommodation;
repair delays; and
no senior review before tenancy termination activity progressed.

The provider creates a corrective-action plan covering workforce stabilization, landlord escalation, early-warning reviews, accommodation practice, post-crisis follow-up and termination governance.

Each action includes a baseline, target and evidence source. Examples include:

reduce unallocated tenancies to zero;
complete post-crisis housing reviews within the approved timeframe;
introduce senior review before formal termination support is withdrawn;
reduce overdue repairs affecting health or safety;
audit accommodation consideration in all sampled high-risk cases;
improve landlord-response time;
restore supervision completion; and
remeasure tenant trust after implementation.

Why the practice exists

Unplanned exits rarely result from one isolated failure. They often reflect several operational weaknesses interacting over time. A narrow response may address the final event while leaving the wider system unchanged.

What goes wrong if it is absent

Leaders discuss the increase but assign no clear actions. The same causes remain active, further tenancies fail and staff begin to see reporting as disconnected from real improvement.

What observable outcome it produces

The provider can demonstrate which controls changed, whether implementation occurred and whether unplanned exits, warnings and tenant-reported insecurity reduced afterward.

Required fields: performance concern, affected population or location, evidence, root cause, interim control, action, owner, deadline, target and verification date.

Cannot close the corrective action without: evidence that implementation occurred and the relevant performance measures were reviewed again.

Auditable validation: before-and-after evidence shows whether the action changed the identified risk rather than merely completing an administrative task.

Use Regulatory Readiness Reviews to Test the Whole Framework

PSH providers often hold substantial evidence but cannot retrieve it quickly or connect it into a coherent assurance story. Regulatory and funder readiness should test whether leaders can demonstrate how performance, rights, safety, tenancy stability and improvement fit together.

A readiness review should examine whether the provider can evidence:

the approved PSH operating model;
Housing First principles and local implementation;
separation between tenancy and voluntary services;
tenant rights and complaint routes;
reasonable accommodation processes;
risk and crisis escalation;
lease-warning and termination governance;
housing-retention and exit outcomes;
property and repair controls;
workforce competence and supervision;
partner and referral reliability;
tenant experience and involvement;
equity analysis;
data definitions and quality checks;
corrective-action completion;
board and executive oversight; and
evidence that learning changed delivery.

The Regulatory Readiness Gap Analyzer can help organizations identify where policies, records, operational evidence and leadership assurance do not yet align.

Operational Example 15: Testing Whether the Evidence Tells One Consistent Story

What happens in day-to-day delivery

A provider preparing for contract monitoring has strong headline retention results. However, a readiness exercise compares dashboards, case records, complaints, supervision notes and landlord logs.

The exercise identifies several inconsistencies:

the dashboard reports low eviction activity, but informal landlord warnings are not centrally recorded;
reasonable accommodations appear in narrative notes but not in a searchable register;
tenant satisfaction is high, but complaint records show recurring concerns about staff access to apartments;
post-crisis follow-up is described in policy but not routinely evidenced;
exit reasons differ between case records and external reports; and
board reports do not include workforce or landlord-performance risks.

The provider creates a readiness improvement plan that aligns definitions, strengthens evidence capture and revises governance reporting before the external review.

Why the practice exists

Oversight credibility depends on consistency across evidence sources. A strong headline result can be weakened when supporting records tell a different story.

What goes wrong if it is absent

Leaders enter monitoring meetings relying on aggregate results but cannot explain operational variation, missing records or contradictory evidence.

What observable outcome it produces

The provider presents a more coherent assurance account, retrieves evidence more quickly and identifies weaknesses before they become formal findings.

Required fields: evidence source, inconsistency identified, risk created, corrective action, owner, deadline and validation result.

Cannot describe the service as inspection-ready without: testing whether policy, records, dashboards, tenant feedback and governance reports align.

Auditable validation: follow-up sampling confirms that previously inconsistent evidence now reconciles.

Use Dashboards to Support Board-Level Accountability

Boards and executive teams do not need every operational measure, but they do need enough evidence to understand whether the model remains stable, rights-based and financially sustainable.

A board-level PSH scorecard should normally include:

housing retention and returns to homelessness;
unplanned exits and eviction activity;
tenant safety, trust and complaint themes;
Housing First fidelity;
reasonable accommodation and rights concerns;
significant property risks;
crisis utilization and post-crisis follow-up;
equity disparities;
workforce vacancies, turnover and weighted caseload pressure;
landlord and partner reliability;
open high-risk corrective actions;
data-quality limitations;
contract compliance;
financial sustainability;
housing inventory and landlord-network risks; and
decisions required from leadership.

The board should receive commentary explaining why performance changed, what action is underway and when assurance will be restored. Green ratings should not be used where evidence quality is insufficient or material corrective actions remain overdue.

The Quality Dashboard Builder can support providers to organize indicators, tolerances, ownership and escalation routes into a clearer governance structure.

Operational Example 16: Escalating a Green Dashboard That Conceals Material Risk

What happens in day-to-day delivery

A PSH program remains rated green because annual housing retention exceeds the contractual target. However, operational evidence shows:

rising lease warnings;
two unresolved habitability concerns;
high workforce turnover;
declining tenant trust;
several overdue post-crisis reviews;
increasing landlord dissatisfaction; and
an unexplained disparity in formal enforcement.

Leadership changes the scorecard methodology. Overall retention remains visible, but the program cannot remain green where leading indicators and control failures exceed defined tolerances.

Why the practice exists

Lagging outcomes can remain positive for a period after the underlying system begins to weaken. Waiting for retention to fall means intervention begins after avoidable harm has occurred.

What goes wrong if it is absent

The board receives false reassurance, management action is delayed and the service eventually experiences a cluster of exits that appeared sudden but was visible in earlier indicators.

What observable outcome it produces

Leadership intervenes earlier, corrective actions receive clearer executive attention and board assurance becomes more realistic.

Required fields: headline outcome, leading risks, tolerance breached, assurance rating, executive action and recovery criteria.

Cannot retain a green assurance rating without: evidence that material control failures are either resolved or subject to effective, timely mitigation.

Auditable validation: the revised scorecard demonstrates that assurance ratings respond to both outcomes and underlying control strength.

A Practical PSH Performance Framework

Providers can organize their approach around ten connected domains.

1. Housing stability

Retention, returns to homelessness, eviction activity, planned and unplanned exits, and housing destination.

2. Early warning and prevention

Arrears, failed contact, complaints, landlord concerns, crisis indicators, benefit disruption and property risk.

3. Housing First fidelity and rights

Voluntary services, non-coercive engagement, lease separation, reasonable accommodation, privacy, choice and due process.

4. Tenant experience and participation

Safety, trust, responsiveness, involvement, accessibility, complaints and tenant-led improvement.

5. Property and landlord reliability

Repair response, habitability, landlord communication, lease-warning patterns and building-level variation.

6. Health, crisis and partnership coordination

Post-discharge follow-up, crisis prevention, closed-loop referrals, medication continuity and partner response.

7. Workforce capacity and competence

Weighted caseload, vacancies, turnover, supervision, continuity, training and after-hours coverage.

8. Equity and access

Referral progression, waiting time, accommodation, enforcement, exit patterns and tenant experience by population.

9. Data quality and evidence

Definitions, completeness, reconciliation, source reliability, limitations and audit trails.

10. Governance and improvement

Thresholds, corrective action, verification, executive oversight, board assurance and external reporting.

No single domain is sufficient. Strong retention with weak rights protection is not a mature model. High tenant satisfaction with poor property conditions is not reliable assurance. Low crisis utilization with inaccessible referral routes may reflect unmet need rather than success.

Questions Leaders Should Ask

Senior leaders, boards and funders should be able to answer:

Do we know who is at risk of losing housing before formal action begins?
Can tenants decline services without jeopardizing tenancy?
Are reasonable accommodations considered consistently?
Do landlord and property concerns receive proportionate, timely responses?
Can we distinguish planned move-on from pressured or preventable exit?
Are property conditions included in housing-stability assurance?
Do post-crisis and post-discharge actions close reliably?
Are referrals confirmed rather than merely sent?
Does workforce capacity match tenant complexity?
Do tenant complaints and qualitative evidence influence decisions?
Are disparities visible and acted upon?
Can every material dashboard measure be traced to a definition and source?
Do corrective actions remain open until impact is verified?
Does the board see leading risks as well as final outcomes?
Can our external claims be reproduced and defended?
Does the performance framework strengthen Housing First practice rather than distort it?

Conclusion

Permanent Supportive Housing cannot be measured well through exits, occupancy and service-contact volume alone. These figures provide only a partial view and may reward the wrong behavior if they are disconnected from stability, rights, fidelity, tenant experience and operational reliability.

A stronger PSH performance framework measures whether people remain safely housed, whether warning signs are identified early, whether landlords and property partners respond effectively, whether crises lead to closed-loop follow-up and whether tenants experience choice, dignity and meaningful control.

It also tests the systems behind the outcome: workforce capacity, data quality, reasonable accommodation, referral reliability, complaint handling, equity, corrective action and governance oversight.

For funders and regulators, this creates a more defensible picture of performance. For leaders, it provides earlier visibility of risk. For frontline teams, it connects data with practical decisions. Most importantly, for tenants, it helps ensure that performance pressure strengthens housing stability rather than encouraging premature exits, coercive engagement or quiet erosion of Housing First principles.

The most credible PSH programs do not simply report that people remained housed. They can show how stability was protected, how rights were upheld, how emerging risks were controlled and how evidence changed the way the service operated.

Return to Knowledge Hub Index