Current context
City broadband rankings need clean city samples
Full fibre and gigabit-capable networks are expanding across UK cities, but a city result is only useful when the dataset is large enough, recent enough and filtered carefully. Package choice, router limits, Wi‑Fi conditions and device speed can all distort a raw speed table.
Use this page as the ranking framework, then open the linked city guides for local rollout, provider and building-access context before comparing your exact address.
Data-building hub
City broadband speed table — data building
The table below is not yet a live ranking. It is the city tracking framework LinkSpeed will use once enough anonymised tests are available. Until then, each city is clearly marked as data building rather than ranked.
| City |
Ranking status |
Why it may perform well |
What to test |
| London | Data building | Dense competition from national networks and London-focused fibre providers in many areas. | Test both Wi‑Fi and Ethernet because flats and dense Wi‑Fi can distort results. |
| Manchester | Data building | Major urban market with strong provider competition and growing full fibre availability. | Compare city-centre flats, suburbs and home-office areas separately. |
| Birmingham | Data building | Large city with strong national provider availability and continuing fibre expansion. | Check whether slow tests are package limits, Wi‑Fi problems or line performance. |
| Leeds | Data building | Large northern city with active ultrafast and full fibre network expansion. | Track upload speed and latency as well as headline download speed. |
| Glasgow | Data building | Growing full fibre and gigabit-capable coverage with postcode-level variation. | Separate wired line tests from Wi‑Fi tests in older buildings. |
| Liverpool | Data building | Urban provider competition and ultrafast availability in many postcodes. | Test multiple times of day to spot congestion or household load. |
| Bristol | Data building | Good mix of national and alternative network availability in many areas. | Compare central, suburban and student-heavy areas carefully. |
| Cardiff | Data building | Full fibre and gigabit availability varies by street and provider footprint. | Check exact address availability before assuming citywide performance. |
| Edinburgh | Data building | Strong urban broadband options, but building type and Wi‑Fi density can affect tests. | Compare Ethernet results against Wi‑Fi in tenements and flats. |
| Belfast | Data building | Northern Ireland has been a strong full fibre market compared with many UK regions. | Track whether available full fibre is actually being used by testers. |
| Newcastle | Data building | Urban competition and ongoing fibre rollout can produce strong results in many areas. | Watch upload, jitter and evening performance. |
| Sheffield | Data building | Large city with mixed legacy and fibre availability by postcode. | Separate router/Wi‑Fi problems from provider speed limits. |
| Nottingham | Data building | Strong student and city-centre demand can drive provider upgrades. | Test both busy evening periods and quiet periods. |
| Southampton | Data building | Dense residential and student areas can create mixed results despite good availability. | Compare wired versus wireless and check packet loss. |
| Cambridge | Data building | High digital demand and provider competition, with local variation. | Measure latency and upload for remote work and research-heavy usage. |
Data-building mode: this page is not currently intended to claim a fastest city. It is prepared for live city rankings once LinkSpeed has enough anonymised city-level tests for a meaningful comparison.
Interactive checker
City ranking readiness checker
Use this to decide whether a city has enough tests for a stable ranking. This can later be replaced with live sample counts from the LinkSpeed recent-results database.
Early signal only
42 tests can show early patterns, but the city should not be treated as a reliable ranking yet.
Methodology
How LinkSpeed city rankings should work
City broadband rankings are easy to distort. A fair ranking needs enough tests, recent data, sensible filtering and separate metrics for speed and responsiveness. LinkSpeed should also use provider and ASN signals where available to separate Openreach-based routes, Virgin Media HFC or nexfibre, CityFibre and other altnets before summarising a city.
Use anonymised tests only
No personal address-level data should appear in the public city table.
Set a minimum sample size
A city should not be ranked from just a handful of tests.
Prefer medians
Median download and upload speed reduce the effect of extreme outliers.
Show latency metrics
Ping and jitter are essential for gaming, calls and remote work.
Filter errors and route distortion
Exclude incomplete tests, browser failures and impossible results. Where available, use ISP, ASN and provider-route signals to avoid one access layer hiding another.
Keep time periods clear
Show whether a ranking covers 30 days, 90 days, 12 months or all-time results.
Recommended ranking metrics
| Metric | Why it matters | How to display it |
|---|
| Median download speed | Shows typical broadband speed for streaming, updates and downloads. | Mbps, ranked highest first. |
| Median upload speed | Important for video calls, cloud backups, cameras and working from home. | Mbps, ranked highest first. |
| Median ping | Measures responsiveness for gaming, calls and cloud tools. | ms, ranked lowest first. |
| Median jitter | Shows connection stability and lag spikes. | ms, ranked lowest first. |
| Sample count | Shows whether the ranking is statistically meaningful. | Visible next to every city. |
| Last updated | Prevents old city rankings being mistaken for current performance. | Date or rolling period. |
Why city rankings can be misleading
Wi‑Fi bottlenecks
Many users test over Wi‑Fi, so weak signal can make a fast city look slower.
Package choices
A city with gigabit availability may still test slowly if many users buy cheaper packages.
Device limits
Old laptops, phones and routers can cap speed test results below the line speed.
Test server distance
Ping and speed can vary depending on where the test server is located.
Time of day
Evening tests can reveal congestion, but they can also reflect household load.
Small samples
A city with very few tests should be marked as “data building”, not ranked.
What makes a city fast?
The fastest broadband cities usually have a combination of strong full fibre coverage, cable or altnet competition, modern router adoption, shorter internal Wi‑Fi distances, better upload availability and enough users testing on genuinely fast packages.
Network competition
Openreach, Virgin Media, CityFibre and local altnets can overlap in competitive areas.
Full fibre uptake
Coverage only helps rankings when households actually upgrade to faster packages.
Router quality
Modern routers and mesh systems help high-speed packages perform over Wi‑Fi.
Upload speed
Symmetric or higher-upload networks can stand out for home working and creators.
Low latency routes
Gaming and calls depend on ping, jitter and routing, not just download speed.
Address-level availability
A citywide result hides street-by-street differences, especially in flats and new builds.
Useful source links
External sources are useful for national rollout context and availability checks. LinkSpeed city rankings should still be based on the site’s own anonymised speed test dataset once it is large enough.
