What share of U.S. households have a broadband subscription?
In the United States, during the 2020–2024 ACS survey period, 91.0% of households reported a broadband internet subscription of any source-defined type.
Estimated numerator: 117,573,073 households. Denominator: 129,227,496.
Across the 50 states, estimates ranged from 93.6% in Washington to 84.1% in Mississippi. Find any state or county in the data explorer.
Source: U.S. Census Bureau · ACS five-year, B28002 · percentage. Derived percentage; 90% margins of error shown below.
Show accessible data table
| Area | Value (percent) | Numerator | Denominator | 90% MOE |
|---|---|---|---|---|
| United States | 90.9815% | 117,573,073 | 129,227,496 | ± 0.1 pp |
| Texas | 91.3504% | 10,041,985 | 10,992,816 | ± 0.1 pp |
| California | 93.4582% | 12,661,798 | 13,548,091 | ± 0.1 pp |
| New York | 90.8465% | 7,015,754 | 7,722,646 | ± 0.2 pp |
| Florida | 91.6148% | 8,018,865 | 8,752,810 | ± 0.2 pp |
What the answer means
The ACS definition includes cellular data plans. A subscription measures household use of a service; FCC availability measures what providers report offering at locations. These two measures should not be combined as though they were the same thing.
Includes cellular data plans. Household subscriptions differ from provider-reported availability, speed, and performance.
How to check it
The chart, answer sentence, indicator and download use one approved observation, rather than separately maintained numbers. The displayed percentage is rounded; the table and export preserve the calculation’s precision. Reference period, population universe and source version are part of the figure.
Inspect the indicator definition, formula and state table . The source is the U.S. Census Bureau, ACS 2024 five-year detailed tables, B28002, released January 29, 2026. The source files were acquired September 30, 2026; this snapshot, which adds derived margins of error, was published October 7, 2026.
Before comparing places
Compare the same measure and survey product with the same definition. A difference between two estimates is a descriptive observation, not proof of a meaningful change, a policy effect or an individual outcome. The national percentage has a derived 90% margin of error of ±0.1 pp. Compare places with the same measure and treat differences as meaningful only when they exceed the combined margins of error.