How the figures are calculated
- Postcode speeds
- How the figures are calculated
Everything here is built from files published by Ofcom: the coverage and full-fibre take-up files from the Connected Nations update for Spring 2026, covering January 2026, and the fixed broadband performance files from Connected Nations 2025, covering July 2025. We accept one version of each, keep its bytes, and calculate from that copy.
Council-area figures are the percentages Ofcom publishes directly for that authority. We do not rebuild them from smaller rows. The one subtraction we publish is availability minus take-up, where both percentages are shares of all premises in the same release.
Postcode-district figures are equal-weight medians of the published index rows inside the district. They are medians of percentages, not percentages of all premises, and a district needs at least 25 rows before any of them appear.
Because Ofcom publishes rounded percentages, every difference carries a rounding allowance: half of the last printed digit on each side, added together. That allowance is printed beside the difference. It is not a confidence interval, and no binomial interval is attached to a percentage whose premises count we do not hold.
Speed bands keep their own averages. We never average band averages into a single speed, because the bands carry no weights.
Worked example, Leeds. Full-fibre availability 93.1% minus take-up 47.0% is 46.1 pp.
The headline average active-line speed for the United Kingdom cannot be checked against an unweighted aggregation of band figures, because the bands carry no weights.
Use a published page
A specific address is not checked here. Check an address with Ofcom.
Open the home pageCheck an address with OfcomOfcom Connected Nations update
| Measure | Result | Low | High |
|---|---|---|---|
| Gigabit availability | 96.0% | 95.9% | 96.1% |
| Premises below the service obligation | 0.0% | 0.0% | 0.1% |
| Unable to receive 2 Mbit/s | 0.0% | 0.0% | 0.1% |
| Unable to receive 5 Mbit/s | 0.1% | 0.0% | 0.2% |
| Unable to receive 10 Mbit/s | 0.2% | 0.1% | 0.3% |
| Unable to receive 30 Mbit/s | 1.4% | 1.3% | 1.5% |
| Superfast availability | 98.5% | 98.4% | 98.6% |
| Ultrafast availability | 96.2% | 96.1% | 96.3% |
| Full-fibre availability | 93.1% | 93.0% | 93.2% |
| Full-fibre take-up | 47.0% | 46.5% | 47.5% |
| Availability minus take-up | 46.1 pp | 45.5 pp | 46.7 pp |
| Rounding allowance | 0.6 pp | No interval | No interval |
What this does not mean
What this page does not show
since 2024 we no longer report on speeds at postcode level
the figures provided here can be compared with the ones reported in 2024 but are not directly comparable with the ones from earlier years
We only include output areas with speed information for at least five lines
Source
- Source
- Ofcom Connected Nations coverage and fixed broadband performance
- Coverage period
- January 2026
- Line performance period
- July 2025
- Publisher release
- Spring 2026 r2 and Connected Nations 2025 performance
- Coverage file SHA-256
- 0ddc33723e345d86ba25d7528d272f6fe12aef5d21d2e3adefdcaa53711c76f3
- Calculation release
- ofcom-2026-10-18-0ddc33723e34-09c7e7c1be2e
The headline average active-line speed for the United Kingdom is not compared with an unweighted aggregation of band figures, because the bands carry no weights.