Today the House of Commons Library has published estimates of broadband coverage and speeds at constituency and ward level. These are based on our analysis of Ofcom’s Connected Nations open data.

You can read our full briefing on superfast broadband here, and download our data tables here.

This blog gives a summary of the areas of the UK have the best and worst superfast connectivity, and a summary of differences between urban and rural areas. The data is from June 2016, so things may have changed a little since then, but this still provides a useful snapshot of varying broadband coverage across the country.

Who’s getting superfast speeds?

The overall picture is of increasing superfast broadband coverage. In all but eight of the UK’s 650 constituencies, availability of superfast lines (those capable of receiving speeds of 30 Mb/s or greater) is over 50%. The highest and lowest constituencies for superfast availability are shown below.

However, not every connection capable of receiving superfast speeds is actually doing so. While 642 constituencies have over 50% availability, there are only 174 where over 50% of connections are actually receiving superfast speeds. The table below shows the highest and lowest constituencies on this measure. This reflects speeds actually being received, rather than the capability of the line, so it’s affected by factors including consumer decisions about which services to sign up for.

Urban and rural differences

Broadband provision is often thought of as having an urban/rural split, with urban areas receiving good coverage and rural areas receiving poor coverage. (The House of Commons recently debated this issue) There is some truth to this. In England and Wales, superfast availability averages around 93%-94% in urban areas, but falls to 89% in rural towns, 65% in rural villages, and 36% in rural hamlets. Similarly, the average download speed is 40-42 Mb/s in urban areas, 30 Mb/s in rural towns, 27 Mb/s in rural villages, and 20 Mb/s in rural hamlets.

However, it’s a bit more complicated than that. The centres of large cities often have a much lower proportion of superfast connections than the outskirts of cities. The maps below show this effect. These maps shows a single dot for each 100 people, meaning that built-up areas are highlighted and sparsely populated areas are shown only faintly. Purple dots show a higher percentage of superfast connections, with the darkest purple showing areas with more than two-thirds superfast connections. Orange dots show a lower percentage, with the darkest orange showing areas with 11% or less superfast connections.

As you can see, the centres of London, Manchester, Birmingham and Liverpool all have far lower percentages of connections actually receiving superfast speeds than the outer areas of those cities. A national dot map is available in our briefing paper downloads.

Thinking local: ward-level data

The data in the tables above gives a single figure for a whole constituency. This can hide a lot of variation within constituencies. For instance, in constituencies there are some areas with good coverage and some areas with poor coverage. In Central Suffolk and North Ipswich constituency, for example, superfast availability ranges from 99.2% in Castle Hill ward to 20.1% in Hoxne ward. Meanwhile average download speed ranges from 51 Mb/s in Castle Hill to 8.9 Mb/s in Helmingham & Coddenham ward.

To get a detailed picture, it’s best to look at ward-level data as well as constituency-level data. Data for all wards in Great Britain is available in our data table downloads.

As stated above, this data comes from our analysis of open data from Ofcom. Ofcom collected and analysed data from major fixed telecoms operators (BT, Virgin Media, Sky, Talk Talk, Vodafone and KCOM). The availability data also includes coverage information provided by alternative network providers (Gigaclear, Hyperoptic, IFNL, B4RN and Relish).

Due to variations in broadband performance over time, the data isn’t a definitive and fixed view of the UK’s fixed broadband infrastructure, but rather a snapshot of variation across the country. Due to privacy concerns Ofcom did not present the information in postcodes with less than four broadband connections. Our data tables include estimates for data completeness at constituency and ward level.

Ofcom’s ‘availability’ data is given in respect of premises in each postcode. In order to aggregate data these availability percentages are applied to the total number of connections in each postcode. This is a simplifying assumption but is necessary for aggregation. Postcodes where the number of connections is suppressed are included in this calculation by attributing them a nominal number of connections.

The UK averages given in our downloadable tables are those calculated on the basis of the postcode-level data (for comparability with the constituency/ward averages). They will not precisely match those found in Ofcom’s Connected Nations report.

Picture credit: Server Room by Spark Fun Electronics Creative Commons Attribution 2.0 Generic (CC BY 2.0)