> ## Documentation Index
> Fetch the complete documentation index at: https://docs.propilotapp.com/llms.txt
> Use this file to discover all available pages before exploring further.
# Neighborhood Benchmarks
> Census-tract market reference data - price per sqft, rents, sample counts, confidence scores, and tract demographics.
Neighborhood Benchmarks provides census-tract-level market reference data: median **\$/Sqft**,
average price, rent, sample counts, and a **Confidence** score per tract and property profile
(beds/baths). It's useful for understanding a local market beyond a single property's comps.
Each tract also carries **demographic data** (income, home values, education, and more) that
you can view right on a property.
**Key things to know**
* Benchmarks are area-level statistics; a property's [Comps](/features/comps) are a
deal-specific valuation - use both together.
* The page isn't in the main sidebar; open it at **`/neighborhood-benchmarks`** and bookmark
it.
* Editing samples or recalculating requires an organization **admin** role; viewing doesn't.
* The **Confidence** score reflects how many samples back a benchmark and how consistent
their prices are.
## Where to find it
Open it directly at **`/neighborhood-benchmarks`**. It isn't in the main sidebar today, so
bookmark the page.
## Browsing benchmarks
Use the search bar to filter by census tract, city, state, county, property type,
bedrooms, or bathrooms.
Toggle **Table** and **Map**. Clicking a tract on the map switches to the table filtered
to that tract. The info button next to the search bar explains the benchmark methodology.
Every column header sorts. Columns: **Census Tract**, **Beds/BA**, **Sqft**, **Garage**,
**Basement**, **\$/Sqft** (the headline number), **Avg Price**, **Samples**,
**Confidence**, **Rent**, and **Last Update**.
## Drilling into samples (admins)
Click a row to open the **Benchmark Samples** view for that tract and bed/bath profile. It
shows the individual sales behind the benchmark, with summary stats up top: how many samples
are included out of the total, how many outliers were auto-excluded, the average price, the
average \$/sqft, and the confidence score (green at 80%+, yellow at 60%+, red below).
Admins can:
* Adjust the **Outlier** threshold (default 20%): samples whose \$/sqft deviates from the
median by more than the threshold are excluded automatically. Changing it saves and
recalculates the benchmark.
* Exclude an individual sample with a reason, or re-include it later.
* Add samples manually (**+ Add Sample**) or pull them in with **Import Samples**, including
from properties your organization has saved.
* Fetch garage and basement details for samples to enrich the profile.
Excluding or including samples recalculates the benchmark's averages and confidence, so the
table reflects your judgment about which sales truly represent the area.
## Census tract demographics
Beyond market benchmarks, every census tract also has **demographic and socioeconomic data**
from the U.S. Census American Community Survey. This describes the people and housing in the
area, separate from the price and rent benchmarks above.
### Where to see it
You don't need to open this page to use demographics. On any property, click the
**Demographics** button next to its census tract to open the **Census Tract Demographics**
panel. Demographics show on a property no matter how it entered your pipeline: saved manually,
or added automatically when you message or email a listing agent or run a sequence.
### What's included
| Group | Metrics |
| ------------------- | ------------------------------------------------------- |
| Financial Metrics | Median Household Income, Median Home Value, Median Rent |
| Housing & Community | Owner-Occupied rate, Vacancy Rate |
| Demographics | Bachelor's Degree+, Poverty Rate, Unemployment Rate |
Each metric only appears when the Census publishes data for it in that tract. Areas without
published data won't show a Demographics panel.
## How it differs from a property's Comps
Benchmarks are **area-level** market statistics for a census tract: what similar homes in the
neighborhood sell and rent for in aggregate. A property's [Comps](/features/comps) are a
**deal-specific** valuation built from sales selected for one address. Use benchmarks to
sanity-check a comps value: if a property's comps-based $/sqft is far above the tract's
benchmark $/sqft, dig into which comparable sales drove it.
## Common questions
It combines how many samples back the benchmark with how consistent their prices per
square foot are. Few samples or widely scattered prices lower confidence - treat
low-confidence rows as rough guides, not gospel.
Benchmarks are kept per bed/bath profile within a tract (for example, 3 BR / 1 BA versus
3 BR / 2 BA), because those segments trade at different prices.
Coverage depends on having sample sales data for the tract. Not every tract has rows, and
rural tracts in particular may be sparse. Admins can add or import samples to build out a
tract they care about.
It was either excluded manually (with a recorded reason) or auto-excluded as an outlier
because its \$/sqft deviated from the tract median by more than the outlier threshold
(default 20%).
Anyone can browse. Adding, editing, excluding, importing, or deleting samples - and
changing the outlier threshold - requires an organization admin role.
## Limitations
* No sidebar link today - reach it by URL.
* Editing or recalculating benchmark data requires an organization **admin** role.
* Coverage depends on census-tract sample data; not every area has rows.
* Demographics only include metrics the Census publishes for that tract.