Britain’s roads claimed 1,602 lives in 2024, with someone killed or seriously injured every eighteen minutes. Against that backdrop, the government's Road Safety Strategy, published in January 2026, represents the most comprehensive policy commitment to road safety in over a decade. It sets ambitious targets: a 65% reduction in people killed or seriously injured by 2035, and 70% for children under sixteen. Achieving those numbers will require the whole system to perform, from legislation and enforcement to vehicle design and, critically, infrastructure.

A recent Project Edward conversation with Chris Spinks, Managing Director of Westcotec and a former Norfolk and Suffolk roads policing chief inspector, offered a practitioner's-eye view of where that system is currently leaving value on the table, and what might be done about it. The discussion centred on rural roads, smart roadside technology, and the role of real-time data in closing the gap between what government knows and what is actually happening on the roads.

The conclusions could have been alarming but, instead, they are encouraging. The technology already exists. The evidence base is growing. And the strategy itself, in several places, creates the precise openings needed to accelerate deployment. The question is how quickly those openings can be converted into coordinated action.

The rural road problem: Well-documented, imperfectly addressed

The strategy rightly identifies rural roads as a disproportionate source of fatalities. In 2024, 60% of all road deaths occurred on rural roads, despite those roads carrying only 45% of traffic. The reasons are well understood: higher speeds, limited separation of opposing traffic, poorer surface conditions, and a greater prevalence of unexpected hazards – from agricultural machinery to wildlife to concealed junctions.

The strategy's response to this challenge, set out under Theme 3 (Ensuring Infrastructure is Safe), focuses principally on updated speed limit guidance, rural road segmentation research, and expanded trials of Project PRIME – a targeted initiative using road markings and signage to guide motorcyclists through sharp bends. These are sensible and evidence-based measures.

What the strategy does not yet fully address is the potential of intelligent roadside technology to complement these measures at a fraction of the cost of hard engineering, and at a pace that hard engineering cannot match. That gap represents an opportunity rather than a failing, and it is one that the strategy's own data architecture commitments could help close.

What smart roadside technology can do… and what it has already done

The Westcotec conversation provided a striking illustration of what is already achievable. At a staggered crossroads in County Durham, a thermal detection system was installed to identify vehicles approaching from a minor road and trigger large vehicle-activated LED warning signs on the main road. The cost was approximately £13,000. The result: in the five years following installation, there were no killed or seriously injured (KSI) collisions at a location that had recorded twelve in the preceding four years.

“Over a three or four year period there were twelve killed and major serious injury collisions on this staggered crossroads. After installation, in the subsequent five years, there was not a single KSI.”

Chris Spinks, Westcotec

This is not an isolated example. Westcotec now has close to 18,000 installations across the country, covering vehicle-activated speed signs, junction warning systems, flood depth indicators, and specialist applications including equestrian crossing alerts, developed in partnership with The Jockey Club, at Lambourn and Newmarket. Each installation has the capability to be cloud-connected, logging speed data, vehicle counts, and temporal patterns that can inform future safety investment decisions.

The technology has also evolved significantly and continues to do so. Systems now in development include in-vehicle verbal warnings delivered through Android Auto when a hazard, such as an over-height vehicle approaching a low bridge, a pedestrian at a crossing, a flooded road, is detected. Temperature-sensitive activation thresholds adjust sign behaviour automatically when conditions approach freezing. And acoustic near-miss detection, capturing tyre noise and impact sounds to identify dangerous incidents before they escalate to recorded collisions, is moving from concept toward production.

These capabilities are directly relevant to the strategy's stated priorities. But the path from available technology to deployed infrastructure remains more complicated than it needs to be.

Where the strategy creates openings

The strategy is, in several respects, well-designed to accommodate the integration of smart roadside technology. Those designing implementation plans should look carefully at the specific commitments that create room to act.

Theme 2's commitment to establish a data-led Road Safety Investigation Branch (RSIB) for Great Britain is particularly significant. The RSIB is explicitly designed to move beyond reactive, post-collision analysis toward pattern recognition and predictive safety intervention. Its core data sources will include police-recorded STATS19 data and the Collision Recording and SHaring system (CRaSH), however, both of which are inherently lagging indicators, with data being generated only once an incident has occurred and been reported.

It is widely acknowledged that road collisions are significantly under-reported at sub-KSI severity. Near-miss incidents and damage-only collisions are almost entirely invisible to current data systems. Yet it is precisely these events that carry the most predictive value for identifying where the next serious collision is most likely to occur.

This is where the 18,000 cloud-connected installations operated by companies like Westcotec become relevant as a policy asset rather than simply a commercial product. That data – covering speed profiles, vehicle classifications, approach patterns by time of day, and activation frequencies at known hazard locations – represents a rich layer of pre-collision intelligence that currently sits largely outside the national analytical picture. Exploring how it might be made accessible, in aggregated or anonymised form, to feed the RSIB's thematic analysis would be a natural and constructive next step.

Similarly, the strategy's commitment under Theme 3 to explore rural road segmentation at a local level, and to assess the potential for national application of rural road categories, points toward exactly the kind of location-specific risk characterisation that smart roadside data can support. Speed profiles and vehicle type distributions from existing installations could provide a richer evidence base for that segmentation work than road geometry data alone.

The approval and deployment challenge

One of the most candid moments in the Westcotec conversation came when Chris Spinks was asked what he would ask of the roads minister. His answer was direct: "to take the brakes off research, development, and deployment by streamlining the Home Office type approval process for new roadside safety products".

The concern is understandable and warrants constructive engagement. The current approval framework was designed primarily around prosecution-grade enforcement technology, such as speed cameras and ANPR systems, where evidential standards are rightly rigorous. But, a significant proportion of the most promising roadside safety innovations are not enforcement tools at all. They are information tools: systems that warn, inform, and prompt drivers to make better decisions, with no prosecutorial function whatsoever.

Applying the same regulatory pathway to a thermal junction warning system as to a speed enforcement camera creates a mismatch between the scrutiny applied and the risk profile of the technology in question. It also imposes delays that, in a context where a dangerous rural junction may be claiming lives each year, carry a real human cost.

The strategy's ambition to reduce KSIs by 65% in under a decade requires the system to move faster than it has historically. One practical way to support that ambition would be to develop a proportionate, tiered approval pathway for non-prosecution roadside safety technology – one that maintains appropriate standards while recognising that information-only systems present a fundamentally different risk profile than enforcement equipment. This need not require primary legislation and could be progressed through the new RSIB's work programme as a relatively early deliverable.

Data sharing and the collaboration imperative

The strategy is explicit that road safety is a shared responsibility. It calls for partnership between government, local authorities, emergency services, and the wider road safety sector. The RSIB should embody that partnership at a strategic level.

At an operational level, however, the mechanisms for translating good local practice into national learning are still underdeveloped. As Simon Turner of ARRM and Project Edward noted during the Westcotec discussion, authorities that have achieved significant casualty reductions through smart roadside technology often have the data to demonstrate that impact. The problem is that there is currently no systematic process for sharing those case studies with other authorities considering similar investments.

The result is a pattern of reinvention. An authority in the north of England installs a junction warning system, documents a dramatic reduction in KSIs, and the evidence sits in that authority's Westconnect dashboard. Meanwhile, a highway engineer in the south faces a similar junction problem and is unaware that it has already been solved, cost-effectively, elsewhere.

The strategy's commitment to the RSIB and to linked road safety data offers a framework within which this kind of structured learning could be formalised. A practical initiative – perhaps facilitated through Project Edward in the near term, and potentially incorporated into the RSIB's work programme over time – would be to develop a searchable, publicly accessible repository of evaluated smart roadside interventions, with before-and-after KSI data where available. This would support local authorities in building investment cases, give technology providers a clearer pathway to evidencing impact, and help the RSIB build a richer evidence base for its thematic recommendations.

Funding, proportionality, and the business case

A recurring theme in discussions of rural road safety technology is the tension between the scale of the safety benefit and the perceived cost of intervention. The strategy's £24 billion capital commitment for road maintenance between 2026 and 2030 is a significant signal of intent, and the record £1.6 billion investment in local road maintenance for 2025-26 provides real resource for local authorities to work with.

The proportionality argument for smart roadside technology is, however, compelling on its own terms. The government estimates the cost of a fatal collision at approximately £2.6 million across the responding agencies. The Durham junction system cost £13,000 and eliminated KSIs at that location for at least five years of monitored performance. Even a conservative assessment of that outcome produces a benefit-cost ratio that most infrastructure investments would struggle to match.

The challenge is that local authority decision-making processes are not always structured to capture that calculation readily. Road safety budget lines compete with maintenance, pothole repair, and active travel investment. The case for smart technology investment is strongest when it can be expressed in terms that align with the metrics decision-makers are already using. The strategy's Safety Performance Indicators, which will be monitored annually, provide exactly that common currency.

Encouraging local authorities to include smart roadside technology assessment within their road safety planning, as a standard step before committing to more expensive hard engineering, would help ensure that cost-effective options are systematically considered. This does not require mandating any particular technology; it simply asks that the question be asked.

Recommendations: Practical next steps

The following suggestions are offered in a constructive spirit, as contributions to the implementation planning that will now follow the strategy's publication. They are designed to be proportionate, deliverable within existing structures, and aligned with the strategy's stated commitments.

• Explore a structured data partnership between the RSIB and providers of cloud-connected roadside infrastructure, to assess whether aggregated, anonymised speed and activation data from existing installations can contribute to the RSIB’s thematic analysis and pre-collision evidence base.
• Task the RSIB with reviewing the type approval pathway for non-prosecution roadside safety technology, with a view to developing a proportionate, tiered framework that maintains appropriate standards while reducing unnecessary delays in deployment.
• Develop a structured mechanism, potentially through Project Edward, for systematically capturing and sharing evaluated case studies of smart roadside interventions, with before-and-after KSI data where available.
• Encourage local authorities to include an assessment of smart roadside technology options as a standard step in road safety scheme appraisal, prior to committing to hard engineering solutions, particularly at identified rural collision hotspots.
• Consider whether representation from the intelligent traffic systems sector should be included within the Expert Advisory Panel, given the growing role of that sector in generating both safety outcomes and road safety data.

Conclusion

The government’s Road Safety Strategy is an ambitious and well-evidenced document. It is right to situate road safety within a Safe System framework, right to acknowledge that progress has stalled, and right to set specific, measurable targets for 2035. The commitment to a data-led Road Safety Investigation Branch is particularly significant and has the potential to transform the analytical basis on which road safety investment decisions are made.

What the Westcotec conversation illustrates is that the technology to support that ambition is not waiting to be invented. It is already deployed, already generating data, and already producing outcomes – in some cases dramatic ones – at costs that compare very favourably with traditional engineering approaches. The task is not to create something new, but to connect what already exists more effectively to the national policy framework that the strategy has now put in place.

The 65% target is achievable. Closing the gap between available technology and systematic deployment is one of the more tractable steps toward it.