CONSTRUCTION TIMING / CENSUS EVIDENCE
Put a real history behind the handover date.
See how long comparable construction categories took to finish, then translate that history into an explicit planning window.
2025 completed-home benchmarks. Published Census and HUD data for the United States and four broad regions. These describe buildings that finished; they are not local completion probabilities or a prediction for your site. Timing begins at construction start, not land purchase or permit application.
01 / CHOOSE THE EVIDENCE
Match the way you build
Builder constructs a house and sells it with its land as a package. United States; not a city-level estimate. Each observation represents a building with one housing unit. A 10-home development is not treated as one 10-unit building.
02 / YOUR PLANNING ASSUMPTIONS
Place it on your calendar
Planning inputs are saved in this browser.
03 / HISTORICAL DISTRIBUTION
The average does not show the whole spread
United States · Built for sale · completed in 2025. Each bar is the published share of completed one-unit buildings in a duration band.
Percentages are rounded and may not total 100. When rounding could move a percentile across adjacent bands, the displayed interval includes both bands. These are estimated historical percentiles, not statistical confidence intervals.
04 / CALENDAR ILLUSTRATION
A starting point for the schedule conversation
Apply the historical month bands to a Oct 2026 start, plus your 0-month planning buffer.
This illustration cannot establish when your project will finish. Site work, crew capacity, approvals, utility delivery and procurement can differ from the broad historical cohort. Add those dated project facts before committing to a completion or closing date.
05 / RECENT CONTEXT
How the published average changed
| Completed in | Average months |
|---|---|
| 2020 | 5.9 |
| 2021 | 6.5 |
| 2022 | 7.6 |
| 2023 | 7.5 |
| 2024 | 6.3 |
| 2025 | 6.1 |
06 / GEOGRAPHY CHECK
Compare the same category across regions
| Geography | Months | Mean RSE |
|---|---|---|
| United States | 6.1 | 3% |
| Northeast | 9.9 | 10% |
| Midwest | 6.5 | 3% |
| South | 5.5 | 4% |
| West | 6.4 | 4% |
RSE is the published relative standard error of the estimated mean. It describes sampling uncertainty, not how much an individual build might vary.
SOURCE READER
Know exactly what is behind the numbers
U.S. Census Bureau and U.S. Department of Housing and Urban Development, Survey of Construction. Privately owned buildings with one housing unit completed in 2025 in permit-issuing places. Retrieved Sep 6, 2026. This is a reviewed annual snapshot; reopening the page does not refresh its observation year.
Coverage, interpretation and original workbooks
- Published aggregate tables are used directly. No microdata records are pooled or reweighted.
- The selected built-for-sale, contractor-built and owner-built columns are separate published categories. The total-one-unit column, which includes built-for-rent units, is not used.
- The denominator is buildings completed in the observation year. Unfinished buildings are not counted as completed; these data do not estimate the chance that every new start will finish.
- These are historical duration benchmarks, not forecasts of a particular project, approval time, sale date or closing date.
- Distribution shares are rounded to whole percentages and may not sum to exactly 100. Grouped percentile ranges allow for this rounding without interpolating within a bin.
- Source timing is month-based. Same-month starts and completions are assigned one month; the next month is 1.5 and the second following month is 2.5, with month differences used from the third following month onward.
- The 13-months-or-more category has no upper bound. No upper completion date is invented.
- Relative standard error measures sampling uncertainty in the published estimate; it is not the variation in individual construction duration or a project confidence interval.
The published aggregate categories are used directly, so this tool does not pool repeated microdata cases or attempt to convert unfinished cases into completion probabilities. It does not estimate abandonment risk. Preconstruction, entitlement, utility connection before start and selling time remain separate questions.
Selected mean: United States!C61; previous mean: United States!C60; sampling error: United States!C62. Exports include every selected distribution cell and the SHA-256 fingerprints of the downloaded workbooks.
Your project inputs stay in this browser unless you explicitly export or share them. Public source requests do not include your budgets.