Point-in-Time (PIT) Count

What Is a Point-in-Time Count?

A point-in-time (PIT) count is a census of people experiencing homelessness on one night. It is carried out by each local Continuum of Care (CoC), the planning body that coordinates homeless services and applies for HUD’s Continuum of Care Program funding.

The count produces the most widely cited figures on homelessness in the United States. HUD compiles the local results into Part 1 of its Annual Homelessness Assessment Report (AHAR) to Congress. The AHAR has been published since 2007.

How It Works

The count happens during the last 10 days of January. It has two parts:

• Sheltered count. Communities count people staying in emergency shelters, transitional housing, and safe havens that night, drawing largely on records those programs already keep.
• Unsheltered count. Counters look for people sleeping in places not meant for habitation, such as sidewalks, vehicles, encampments, bus stations, and abandoned buildings. They often use short surveys to collect basic demographic information.

Under 24 CFR 578.7, a CoC must conduct a count at least every two years that covers both sheltered and unsheltered people. Most CoCs now count every year.

Alongside the PIT count, CoCs complete a Housing Inventory Count (HIC). It tallies the beds available that night in shelters, transitional housing, rapid rehousing, permanent supportive housing, and other permanent housing programs.

Example

According to HUD’s 2025 AHAR, published in May 2026, 745,652 people were homeless on a single night in January 2025. That was 25,828 fewer people, or about 3% less, than in 2024, when homelessness reached a record high. About 56% of those counted were in emergency shelters, nearly 36% were unsheltered, and 8% were in transitional housing.

Criticisms and Limitations

• A single night. The count captures who is homeless on one night, not how many people experience homelessness over a year. Annual figures come from separate reports based on local Homeless Management Information Systems (HMIS), the databases that homeless service providers use to record whom they serve.
• Undercounting. People sleeping outdoors or in vehicles are hard to find, and communities differ in their capacity to count them. HUD cautions that the actual number of unsheltered people could be larger than reported.
• Narrow definition. HUD’s count excludes people doubled up with others or staying in motels they pay for themselves. Schools use a broader definition of homelessness under the McKinney-Vento Act.
• Shifting conditions. Weather, local policies, and changes in counting methods can move the numbers from year to year. HUD notes, for example, that New York and Illinois reported their large 2024 increases were partly driven by immigration.

Even with these limits, the PIT count gives communities a consistent benchmark for planning and for measuring progress over time.