By Steve Smith -
By almost any honest measure, the construction industry is finally getting better at jobsite safety. The data tells the story:
- 1,034 construction workers died on the job in 2024, the most recent year for which data is available, according to the Bureau of Labor Statistics — 41 fewer than in 2023.
- The fatal injury rate fell to 9.2 per 100,000 full-time equivalent workers, the lowest rate in over a decade — but still nearly three times the rate across all private industry.
- Federal OSHA fatal investigations dropped 11% in fiscal year 2024, with trench-collapse deaths down nearly 70% since 2022 and fall-related deaths down close to 20% in areas of focused enforcement.
That progress is real, and it's worth naming. Compliance, training and enforcement — the levers the industry has pulled hardest over the past decade — are working. But the rate is still stubbornly high, and the easiest gains have been made.
The harder question, and the one Construction Safety Week's new five-year vision is really pointing at, is what comes next. From where I sit, the next step forward in construction safety won't come from another regulation or another standalone tool. It will come from something less visible: how the technologies we already deploy on a jobsite connect to each other — and to the people who need to act on what those technologies see.
The fragmentation problem
Spend any time on a complex jobsite in 2026 and you'll see the future and the past stacked on top of each other.
Drones perform morning scans. Mobile lidar units capture as-built conditions twice a week. AI-powered cameras flag missing PPE in real time. Wearable sensors track fatigue and posture. Reality-capture platforms compare scheduled progress against actual site conditions. BIM models drive coordination across trades. Digital markup and collaboration tools — including the kind we make at Bluebeam — keep field crews and office teams aligned on RFIs, submittals, and drawing revisions.
Any one of those technologies, deployed well, can move the safety needle. The issue is not whether the tools exist, or whether they work. The issue is whether they connect.
Industry research has been pointing at this for a while. Around 37% of construction companies use four or more applications on their projects, and many of those applications are not properly integrated. That is not a tooling problem so much as a workflow problem. And in safety, workflow problems get people hurt.
This is what fragmentation looks like in practice: A drone scan identifies a trench edge encroaching into a pedestrian work zone, but the data lives in a reality-capture platform the superintendent does not open daily. An AI camera flags repeated unsafe behavior near a loading dock, but the alert routes to a safety dashboard nobody reviews until the weekly meeting. A foreman annotates a hazard on a mobile markup, but the change does not sync into the project management system until the next file refresh — by which point the relevant trade has rotated off-shift. None of those scenarios involves a failure of technology. Each tool did its job. The failure was at the seams.
"All In Together" is a technology argument, too
The 2026 Safety Week framework — Recognize, Respond, Respect — is fundamentally about collective ownership. It is a call to engineer safety across every phase of a project, with shared terminology, shared responsibility and aligned action across owners, designers, contractors and craft. It is a cultural argument.
But every cultural argument in construction is, eventually, an integration argument. Shared responsibility only works if the underlying data is shared and easily accessible. A foreman cannot act on a hazard flagged by a system he does not have access to. A safety officer cannot drive accountability for a behavior pattern she cannot see across crews. A project executive cannot allocate resources to high-risk work zones if reality-capture data, near-miss reports and incident logs live in three different platforms with no connection between them.
The conversation has spent years on which new tool to deploy. The more useful question, especially for the next five years, is which existing tools need to start working together, and what the data flow looks like when they do.
Where integration earns its keep
Field studies on hazard recognition show that during typical pre-task plan briefings, construction workers identify only about 45% of the hazards present. Structured tools like the Energy Wheel model can improve recognition by roughly 30%. The hazards workers miss are not random; they tend to be the ones requiring more cognitive effort to spot, like pressure, mechanical or chemical hazards, as opposed to obvious gravity or motion hazards.
That gap is exactly where integrated technology earns its keep. Reality capture handles the cognitive load humans are not built for: volumetric comparisons between yesterday's site and today's, millimeter-level deviations in a structural member, trench-edge geometry that is hard to assess by eye. Integration is what gets that information to the foreman in the morning huddle, in the same view as the crew assignments and the day's drawings, not buried three logins deep.
For our part at Bluebeam, this is what the partnership work has been about. Bluebeam Studio gives field and office teams a shared markup environment, but its real value emerges through the integrations layer: connecting Bluebeam to construction management platforms like Procore, to document control systems, to design tools across the broader Nemetschek ecosystem, and to safety-management workflows from sister brands like GoCanvas and SiteDocs.
A hazard surfaced on one tool can be routed, tracked and resolved across the others. That is not a feature pitch but a description of what is becoming table-stakes for construction technology in 2026.
Integration is not just a software problem
There is one more piece of this that I want to be clear about. Connecting tools is necessary, but not sufficient. The people using those connected tools have to know how to interpret what comes out of them, and that is increasingly a workforce question.
Bluebeam CEO Usman Shuja has talked about the importance of "dual athletes" in construction: professionals who combine years-long industry experience with the technical fluency to deploy modern tools effectively. Safety is one of the clearest places that idea matters. A risk dashboard nobody knows how to read is just a screen full of red. A predicted hazard means nothing if the foreman who needs to act on it cannot connect the prediction to what is happening at his cut line. And a 25-year veteran's instincts add the most value when his observations can be captured, shared and acted on across the project, not when they live only in his head and disappear when his shift ends.
Tribal knowledge plus technical fluency plus connected tools. That is the workforce equation behind "All In Together."
The work ahead
Safety Week is a useful pause. The banners, the toolbox talks and the executive walkarounds matter. But the work that drives the next decade of fatality reduction will happen in the architectural decisions general contractors, owners and trade partners make about which tools their crews use, how those tools fit together and how data moves between them.
The 2024 numbers are evidence the industry can move. The next move — getting from a 9.2 fatality rate to something durably lower — will require a different kind of effort. Not more tools. Better connections between the tools we already have, and a workforce that can use those connections well. "All In Together" applies to the systems we ask our people to work with, not just to the people themselves.
The faster the industry recognizes that, the faster the next leap in construction safety will arrive.
