Britain's road risk, mapped street by street
We process the official collision, crime and injury records the government publishes and turn them into maps and tables anyone can use. Free for the public. Commissioned reports for the people who need them for work.
Every figure traceable to its source file. Method published.
Official sources
DfT, police.uk, DVSA, HSE
125 areas covered
Towns and cities across Britain
Method published
Every rule open to checking
Free to reuse
Open Government Licence
Raw government files, turned into something usable.
We take the official files
Half a million police-reported collision records, monthly street-level crime feeds, 32 million MOT tests, workplace injury returns. All published by government, none of it easy to read.
We clean and join them
Records get matched to places, checked against boundary data, filtered for the known traps, and every processing rule gets written down on our method page.
We publish it free
Maps, tables and downloads for 125 areas, updated automatically. Councils, consultants and campaigners commission the deeper cuts.
Six datasets nobody else puts in a usable form
Check the roads where you live
Enter a postcode to see the nearest collision hotspots — the junctions where people have been hurt more than once — drawn from five years of official DfT records. No email, no sign-up, and the underlying data is downloadable.
Open the postcode checker →Everything here can be checked
Anyone can publish a chart. These are the standards we hold ourselves to instead.
Named sources on every figure
Each dataset page lists the exact government file it came from, with the publication date. If you want to reproduce a number, you can.
The method is published
Every filter, exclusion and matching rule is written down on our methodology page, including the awkward ones that change results.
We publish the gaps
Where a force's data is incomplete or a dataset can't answer a question, we say so on the page rather than quietly filling it in.
Updated automatically
Pipelines re-run as new government data is released, so the maps reflect the current release rather than a snapshot from launch.
Free to reuse
Everything is derived from Open Government Licence data and can be reproduced with attribution, including by people who disagree with us.
Corrections welcome
If a figure looks wrong, tell us. We check it, fix it if needed, and note what changed.
Built for people who have to justify a number
The figures are free, sourced and reproducible. That matters most to the people who will be challenged on them.
Journalists
A local casualty figure with the source file named, and a custom cut by ward or corridor turned round the same day, free.
Campaigners
Evidence for a junction or a School Street that a council cannot wave away, because the method is published alongside it.
Councils and partnerships
Comparison against genuinely similar authorities, normalised by traffic rather than by population.
Researchers and developers
A free JSON API with source provenance on every endpoint, and CSV export on every table.
Open in full, and checkable
Everything here is published in full, free, with the source files and method attached, because numbers that get used in public need to survive being checked. The ward-by-ward analysis of Birmingham covers every fatal and serious collision over five years, mapped to the ward it happened in and ranked against the rest of the city. That is the standard we hold the whole site to.
Accidents are common. The official record shows how common.
UK road, crime and injury data, summarised from official sources.
128,272
people were casualties in reported road collisions in Great Britain in 2024
29,467
were killed or seriously injured on the roads in 2024
55%
of all road casualties are car occupants — drivers and passengers
1,633
people were killed on Britain's roads in 2024 — around four every day
Where the UK’s crashes cluster
The junctions with repeat collisions five years running, mapped from official DfT data. Click a circle to explore.
Sources: Department for Transport, Department for Transport. Open Government Licence v3.
Common questions.
What is Swift Assist?+
A UK road risk data publisher. We process the official government records of road collisions, casualties and vehicle crime — files that are public but published in formats almost nobody can use — and turn them into maps, rankings and downloadable datasets covering every local authority in Great Britain. Everything is free to read and free to reuse with attribution.
Where does the data come from?+
Named government sources, cited on every page. Road collisions and casualties come from DfT STATS19, the record police complete at the scene of every injury collision. Traffic volumes come from DfT's road traffic estimates, vehicle crime from police.uk street-level records, and MOT and driving test figures from DVSA. All are published under the Open Government Licence v3. We create no data of our own; what we add is the processing.
How current is it?+
The collision data covers the most recent five years published by DfT, refreshed automatically when a new release lands. Vehicle crime updates monthly. Each page shows the date it was last rebuilt and names the source file behind it, so you can check the figures against the original.
Can I reuse these figures in an article or report?+
Yes, free of charge, including commercially. We ask for a credit and a link to the page the figure came from. Journalists, campaign groups, councils and academics do not need to ask permission, and we will usually cut a figure a different way — by ward, corridor or junction — free and same-day. See our licensing page for the detail.
How is road risk measured here?+
By killed and seriously injured casualties per billion vehicle miles driven, rather than per resident. Nearly every published comparison divides by population, which flatters places nobody drives through and penalises places absorbing other people's traffic. DfT publishes traffic volumes per authority, so we use them — and we publish the per-head figure alongside, because the two disagree and that disagreement is often the story.
What are the limits of this data?+
STATS19 counts collisions police attended and recorded, so collisions never reported never appear, and under-reporting is worst for cyclists and for slight injuries. Police Scotland does not publish street-level crime, so Scottish vehicle crime is absent. Where a dataset cannot answer a question we say so on the page rather than filling the gap. Every processing rule and exclusion is written down on our methodology page.
Need this data for your own work?
Site and junction reports from £95, next working day. Free for community groups, campaigners and journalists.