Between 2011 and 2017, homelessness in Los Angeles increased 75%, to an estimated 55,000 individuals. The Los Angeles Homeless Services Authority is the entity primarily responsible for contracting for homeless services in the Los Angeles region, managing millions of dollars of services such as diversion/prevention, interim housing, rapid rehousing, and permanent supportive housing.
The Harvard Kennedy School Government Performance Lab spent thirteen months working with the Los Angeles Homeless Services Authority to implement Active Contract Management across 88 service providers and 400+ contracts. ACM allowed LAHSA to engage in high-frequency reviews of performance and outcomes data with providers and to work collaboratively to improve results.
LAHSA is the first organization to restructure an entire department in order to standardize ACM as a formal practice and represents the GPL's largest ACM project to date. LAHSA created a new department for performance management including dedicated staff for implementing ACM. Dashboards, reports, data analysis, regular meetings, and provider engagement are overseen by analytics and management teams that substantially increase the agency’s ability to identify and respond to data-driven insights into provider performance.
See Full Project Description
Between 2011 and 2017, homelessness in Los Angeles had increased 75% to an estimated 55,000. To combat this epidemic, voters in the region passed Measure H and Proposition HHH, which will provide $1.2 billion in bond revenue for new housing for the homeless and $355 million annually in sales tax revenue for homeless services such as diversion/prevention, interim housing, rapid re-housing, and permanent supportive housing. The Los Angeles Homeless Services Authority (LAHSA) is an independent joint powers authority whose role is to coordinate the use of federal and local funding for homeless services across Los Angeles County. Measure H and Proposition HHH brought a large influx of resources to an agency that already managed over $243 million annually in federal, state, county, and city funds. The increase in resources highlighted the urgency for accountability and demonstrable progress in ending homelessness in Los Angeles. To help them accomplish this goal, GPL helped LAHSA set up a performance management system called Active Contract Management (ACM) that allows LAHSA to engage in high-frequency reviews of performance and outcomes data with providers and to work collaboratively to improve results.
Need for ACM at LAHSA:
With the passing of Measure H, LAHSA realized its need for the tools and capacity to review data and manage homeless service providers during its contract cycle. Prior to the passage of Measure H, LAHSA’s contract management strategy focused heavily on monitoring and compliance. LAHSA would produce quarterly and annual performance reports to track progress on contractually obligated performance targets. However, the discussion around these reports tended to focus on report accuracy rather than performance and opportunities for improvement, and providers that failed to meet their performance targets rarely faced any real consequences. Outside of these reports, homeless services providers found it difficult to pull and review their own data from the Homeless Management Information System (HMIS), the tool which LAHSA uses to gather data on homelessness across all providers. Some were tracking client progress in separate databases, which led to inaccurate data in HMIS and duplicative work.
When issues arose with a provider, the agency had an Interagency Support Team (IAST) to provide technical assistance for individual contracts. Meetings were called on an ad hoc basis, primarily to discuss administrative challenges such as invoicing, compliance, and data quality. They rarely focused on program outcomes or progress toward performance targets and typically happened after problems had already materialized. LAHSA recognized the need to review performance outcomes on a regular basis, yet the agency had neither the tools nor the staff capacity to do so. At the end of every contract term, the monitoring department produced a report of findings for each individual contract. This report was utilized to retroactively engage providers on fiscal, compliance, and performance issues from the previous contract term. Homeless System Analysts used the reports to assist those agencies with a significant number of findings. However, this process did not lead to tangible improvements in the outcomes of homeless services contracts. Increased funding from Measure H provided the opportunity and the resources to significantly alter LAHSA’s contract management strategy.
ACM at LAHSA:
ACM is a set of strategies developed by the Harvard Kennedy School Government Performance Lab (GPL) in partnership with state and local governments that apply high-frequency use of data and strategic agency-service provider engagement to improve outcomes from contracted services. ACM consists of regular, data-informed meetings between government agencies and service providers that focus on performance outcomes. This enables agencies and service providers to troubleshoot problems, solve operational challenges, and identify best practices for broader systems transformation.
The GPL has been piloting ACM across the country and adapting the model to fit agency need. In Rhode Island, the Department of Children, Youth, and Family (DCYF) organizes monthly ACM meetings with leadership from its four family preservation service providers. Issues raised in these meetings are addressed through ad hoc working groups consisting of agency and provider staff. In Seattle, the city’s Human Services Department (HSD) tracks progress on six key homelessness metrics. Its ACM strategy uses three types of meetings: monthly data sharing and contract check-ins between HSD and service providers, quarterly internal ACM meetings to discuss trends, and quarterly executive meetings between HSD and service providers to review performance.
LAHSA is implementing ACM across all 88 direct providers and 400+ homeless services contracts in Los Angeles County. Given the scale of the project, the ACM model has been adapted to accommodate LAHSA’s needs. In January 2018, LAHSA restructured its organization to build a department dedicated to performance management. The Performance Management Division hired a team of thirteen, with ten analysts, two supervisors, and one manager responsible for implementing ACM at LAHSA. LAHSA has developed individual dashboards for the three primary program components: crisis housing, bridge housing, and rapid rehousing. The dashboards are populated by the Performance Analysis and Reporting (PAR) team and utilized by the Performance Management team (PM).
LAHSA’s ACM strategy consists of the following features:
LAHSA uses Tableau to create dashboards that are used to review trends by provider and program component. For each program component (crisis housing, bridge housing, and rapid rehousing), there is one dashboard that portrays aggregated data as well as individual provider performance. Dashboards are updated by the PAR team monthly, allowing for the PM team to review trends on a short time horizon.
2. Performance Summaries
The PM team utilizes the dashboard to develop individual monthly summaries. Monthly summaries include enrollment and exit figures, as well as progress on outcomes targets. These summaries are sent to individual providers along with an overall system summary for comparison. The summaries serve as a starting point for troubleshooting agency performance.
3. Data Analysis
Based on initial trends observed in the dashboards, the PM team conducts deeper dive analyses every month, focusing on a specific topic to present at the ACM internal meeting. Topics may center on priority scores, length of stay, or progress towards performance targets, among other topics. This process helps to uncover agencies which may need additional support or which are performing well outside of contractual performance targets.
4. Monthly ACM Internal Meetings
Internal staff meet monthly to review the dashboards and determine which providers to prioritize for engagement. The meetings are cross-departmental, engaging members from the finance, data, policy and systems, and performance management departments.
5. Monthly Provider Engagement
Based on the findings from the ACM internal meeting, PM conducts provider engagement with both high-performing and low-performing agencies. The objectives of these meetings are to gather best practices and collaborate with providers to define strategies for performance improvement.
6. Quarterly Learning Communities
Learning Communities bring together providers by program component to look at data, discuss trends, and share best practices to improve performance. The learning communities allow program managers and direct line staff to learn from each other and identify ways to more effectively deliver services. Learning communities are a place to develop tools, share effective practices, and identify training needs. LAHSA, along with U.S. Department of Housing and Urban Development technical assistance providers, facilitates these meetings and supports agencies with needs that are identified in the meetings.
Within the first few months of implementation, ACM already began to show benefits. By utilizing the dashboards and other information generated from the ACM process, LAHSA was able to conduct a comprehensive system analysis for the Family Coordinated Entry System at mid-year of FY17-18. This analysis uncovered budgetary shortfalls within the family system that would have caused some providers to run out of funding before the end of the fiscal year. By identifying this problem early, LAHSA and the County Executive Office (CEO) were able to shift resources and change practices to ensure that funding would remain available to keep families stably housed and off the streets. If LAHSA had not been able to provide this level of analysis and to react in a timely manner, many agencies serving families would have had to turn them away from interim housing, which would have led to an increase in families experiencing unsheltered homelessness.
In addition, an in-depth rapid re-housing analysis was conducted to identify agencies in the adult and youth systems in imminent need of technical assistance. Within the first three months of the launch of ACM, LAHSA conducted 31 technical assistance/provider engagement visits to address the issues identified in the analysis. The results from this analysis were an integral part of the performance management division’s first rapid re-housing Learning Community which gave providers the opportunity to receive a detailed, guided exploration of the data while engaging in peer-to-peer learning.
LAHSA’s ACM model is the first attempt at implementing this strategy across an entire agency. The process has helped to uncover important insights in service delivery, leading to intensive technical assistance plans for agencies, as well as County-wide trainings on service delivery and data integrity.