Walk the floor at CONEXPO and the story tells itself.
Hundreds of thousands of professionals — contractors, engineers, project owners, technology vendors — converging to do what this industry has always done: find better ways to build. The energy is real. The investment is real.
And the numbers behind it are staggering.
Trillions of dollars flow through the sector annually — a number that grows every year as urbanization, infrastructure renewal and the energy transition keep driving demand.
The people building at this scale are experienced, resourceful and increasingly tech-savvy. Technology adoption in construction is getting better, and it’s not because consultants told the industry to change, but because the people doing the work found tools that made hard jobs more manageable.
The question is not whether construction can embrace technology. It already has. The question is whether the tools the industry has adopted are talking to one another, and what it costs when they do not.
There is an old story about a tower whose builders could not coordinate because they no longer shared a common language. Construction has built its own version of that tower — not out of ambition, but out of necessity. Each generation of software solved a specific problem well: estimating platforms, BIM tools, field reporting apps, project management suites, finance systems.
Each one is genuinely useful. The problem is that they were never designed to speak to one another.
Before I ever sold a construction tool, I taught German at a university for four years. I have spent most of my career thinking about what happens when people — or systems — do not share a language.
Consider what a project manager on a mid-size commercial job navigates today. There is a design platform holding the drawings. A separate system for estimates. Another for submittals and RFIs. A field reporting tool on the site superintendent’s phone. A finance system in the back office.
And somewhere, almost certainly, a spreadsheet bridging the gaps between all of them.
This is not a failure of technology. It reflects how the built world works. A company we work with closely manages relationships with more than 2,000 vendors. Many of those relationships are project-mandated — an owner specifies a subcontractor, a municipality requires a particular system, a design firm has a preferred platform.
You cannot simply standardize your way out of that complexity. The tower is not going anywhere. Still, that does not mean you have to keep paying the full price of living inside it.
This is what I call the “tool tax” — or, more precisely, the inefficiency tax.
Every time a project team has to re-enter information from one system into another, they are paying for it. Every time a field supervisor cannot access the latest design revision because it lives on a platform, he does not have credentials for, they are paying for it. Every time a project manager calls a subcontractor to confirm something that should be visible in the data, they are paying for it.
They are the clicks that quietly drain productivity from every person on the project team. Integration is the antidote, and saving those clicks adds up.
Industry research has found that construction professionals lose more than 14 hours per week to nonproductive activities: searching for information, resolving coordination failures, re-keying data that already exists somewhere else.
Fourteen hours. Every week. Per worker.
I spent nearly a decade at Leica Geosystems selling the instruments that generate this data — laser distance meters, reality capture scanners. The hardware was never the bottleneck. What happened to the measurement after it left the device was.
I heard it put simply by a customer not long ago: the entire annual cost of their software subscription was recovered the moment one team member saved two hours of manual work. Two hours. That framing stuck with me.
Technology in construction is not just a productivity tool but a risk mitigation strategy. Every hour saved is an error avoided, a decision made on better data, a project that closes cleaner.
The consequences go well beyond wasted time. When project information is fragmented across systems that do not talk to one another, decision-making degrades. Teams act on stale data. Discrepancies between the drawing set in the field and the latest revision in the design platform generate conflict and confusion.
Approvals slow down. Errors go undetected. Bad data costs the construction industry enormous sums every year — much of it tied to rework, tearing out and redoing work because someone was operating from the wrong information.
This is not a problem of intelligence. It is a problem of infrastructure. The industry’s information flows are broken, and no amount of machine learning will fix that on its own.
Something is starting to change.
Across the industry, new ways to connect are emerging — point solutions are either connecting with relevant workflows or simply going away. Some solutions are expanding horizontally, building integrations that pull together data from design, field and finance.
Others are carving out niche workflows, but with a partner-first mindset. And some are building open ecosystems with published APIs, inviting third-party tools to connect. The goal in each case is the same: let information flow more freely across the project lifecycle.
One promising development in this space is Model Context Protocol, or MCP — a standard that allows AI tools to communicate directly with the software systems around them, pulling context and acting across platforms without requiring custom-built integrations for every connection.
For an industry managing hundreds of vendor relationships per project, that kind of connective capability is not a nice-to-have. It is a multiplier.
One dimension of construction coordination often gets lost in these conversations: construction communication is fundamentally visual. A contract describes a building in words. A schedule describes it in time.
Yet the way a project team understands what is being built — where the ductwork runs, how the structural steel connects, what the finished ceiling height should be — is through drawings. Through markups. Through shared visual context that allows dozens of trades to coordinate their work.
When teams cannot access the same visual information, in the same state, at the same time, everything downstream suffers. Interpretation diverges. Assumptions pile up. And by the time the discrepancy surfaces in the field, it has usually become expensive.
The most effective integration strategies in construction are not just about moving numbers between systems. They are about creating a shared layer where project information — especially visual project information — can be reviewed, marked up, approved and acted on by the people who need it. That coordination layer is the connective tissue the industry has been missing.
Building integration partnerships is literally my job, so discount my enthusiasm accordingly. But I arrived at this conviction selling hardware, not software.
This is where AI enters the picture — but not in the way most of the hype suggests.
The construction industry is right to be interested in artificial intelligence. AI systems that can surface schedule conflicts before they cause delays, flag design changes that will ripple through cost estimates or predict cash flow pressure on a subcontractor mid-project — that is genuinely valuable capability.
But every one of those applications depends on connected data. An AI model cannot flag conflicts between the schedule and the structural drawings if it cannot see both. A risk model cannot identify cost exposure if the finance system and the project management system are not integrated.
Artificial intelligence is only as useful as the information it can access, and right now, in most construction firms, that information is siloed, duplicated and out of sync.
This is the uncomfortable truth about the AI moment in construction: the tools that will make AI work are not the AI tools themselves. They are the integration infrastructure that must exist underneath them.
BIM adoption continues to accelerate, bringing with it more structured, machine-readable project data. Digital twins are moving from experimental to operational. Cloud-native platforms are making integration easier than ever.
And new connectivity standards like MCP are beginning to let AI tools work across the full stack of construction software rather than within a single platform.
The companies that will get the most out of AI are not necessarily the ones that move fastest to deploy AI features. They are the ones that invest now in connecting their systems, structuring their data and building the habits of digital coordination that AI requires to deliver on its promise.
That investment compounds. Every project run on connected systems generates better data. Better data makes risk models more accurate. More accurate risk models make decisions less expensive.
Construction has built something extraordinary — cities, bridges, hospitals, campuses, the physical infrastructure of modern life. The industry that built all of that can build the digital foundation the next generation of tools requires. The complexity is real, the scale is real and the opportunity is real.
The AI gold rush is real, too. But the companies that profit most from it will be the ones who laid the pipes before the rush began.
