How Construction Firms can Use Data

"How do we start to leverage AI with our own data?" I was part of an awesome dialogue about leveraging AI within a large civil and utility company. Everyone keeps saying “AI is the future of construction,” but if you’re a $100M+ civil contractor with a big, messy tech stack… you can’t use AI until you get your house in order. Here’s where you actually start: 1. Do a real data inventory. What systems do you use? Where does the data live? Who owns it? What’s clean and what’s garbage? If you don’t know this, AI will do nothing for you. 2. Build a central data warehouse. Pull ERP, field logs, telematics, trucking, estimating, safety, HR—everything—into one place. If your data stays siloed, AI can’t learn the whole business. 3. Clean and standardize the data. Construction data is chaos. AI can’t fix inconsistent job names, missing fields, or 12 versions of the same Excel file. Create rules, structure, and governance. 4. Integrate before you “AI.” (This is one of my favorite steps 😁) Connect your systems with real APIs and ETL pipelines. A connected ecosystem beats a pile of point solutions every time. 5. Start small with high-impact use cases. Think: – Predicting cost overruns – Forecasting trucking needs – Auto-building daily reports – Catching production slowdowns – Improving bid accuracy Small wins get buy-in fast. Bottom line: AI isn’t the first step. The first step is building the infrastructure so AI actually has something to work with. Clean data + solid integrations = a company that can truly leverage AI. If you are still reading this and are in a seat responsible for AI strategy and implementation I would love to talk. #dirtworld #constructiontech #AI #AIstrategy

We collect more data than ever before. But in construction, we hardly touch it: Programme data, cost codes, production logs, fuel burn, cycle times, weather sensors, surveys, trackers. Even WhatsApp thread records. The digital age has given us more information than we know what to do with. The Dutch have a phrase: "Meten is weten" - to measure is to know. Yet in civil and marine construction, we often build first and understand later. Recently I was involved in a post-project review for a major marine project. With basic cost analysis, the pattern became crystal clear: Preliminaries were consistently underpriced across every tender. Site setup, security, permits, design support, contract management, head office overhead - they always needed more time and people than budgeted. But the truth is price prelims honestly based on actual data and you might not win the work. That's where data analysis becomes strategic - not just for pricing smarter, but for building smarter. Because data doesn't just show you where you're bleeding money and time. It shows you where you're strongest. That site team who finishes quay walls early every time? That dredger who hits top production curves despite weather delays? Double down on what works. The most commonly quoted phrase applies here: "Insanity is doing the same thing over and over expecting different results." Einstein probably never said it, but it rings true. Most contractors I know have the data sitting in spreadsheets, databases, and project files.  They just don't analyse it systematically. They repeat the same estimating mistakes. Ignore the same warning patterns. Miss the same optimisation opportunities. In a world full of assumptions and inherited practices, data is your sharpest tool. But only if you actually pick it up and use it.

Let's Talk About Data... No, Really. For years, we've talked about "digital transformation" in construction. But what does it actually mean? Too often, it's just a buzzword. The real transformation, the one that will define the next decade of our industry, is building a data-driven culture. We're great at generating data. It's everywhere – on our jobsites, in our trailers, across dozens of apps, and people are adding more and more tools every day that they "need" to do their job. As my friend Rob Sloyer cleverly pointed out to me one time, "if you give a worker a leaf blower to help clean a deck, they’ll see that as a benefit, but only if they consider cleaning the deck part of their job" - I paraphrased this in a recent article with On-Site Magazine. People need to understand that tools to collect, analyze and drive decision with data are part of their job and are what will make them better at it. The reality is that construction teams lose a staggering amount of time—some studies say up to 18%—just looking for information. That’s a massive, self-inflicted wound on productivity and profitability. The Vision: Imagine a jobsite where every decision, from schedule sequencing to change order approval, is based on a single source of truth. Where disputes are resolved not by who shouts loudest, but by looking at a clear, objective data record. That's not a fantasy; it's where we're heading and it sounds like a tool I would have used when I was in the field. The Reality: I get it. The biggest barriers aren't technology; they're about people and process. There are legitimate concerns about trust, training, and liability. Moving from "gut feel" to data-driven insights feels like a huge leap. Seeing data as part of our jobs is a new mind set. The Path Forward: It starts with "data maturity". You don't have to boil the ocean. 1️⃣ Integrate: Focus on creating a connected platform where data can flow seamlessly between the tools you already use. 2️⃣ Clean & Standardize: Invest in getting your data house in order. Good AI and analytics are useless if they're fed bad data. 3️⃣ Empower: Give your teams the tools and training to see data not as a burden, but as their most valuable asset. This isn't just about efficiency. It's about making our industry safer, more predictable, and more profitable. It's about building a better future. What's one step your team is taking to build a better data culture? #DataDriven #DigitalTransformation #FutureofConstruction #DataMaturity

On most projects today, we are surrounded by data. Schedules with thousands of activities. Cost systems tracking every budget line. Procurement logs, engineering progress, resource histograms, invoices, commitments. The data is everywhere. Yet, in many project reviews, teams still ask the same question: “Why did we only notice this problem now?” The truth is that the problem was rarely invisible. The signals were often already there. They were just buried inside spreadsheets, reports, and disconnected systems. This is where data analytics in project controls starts to make a real difference. It is not about building fancy dashboards or adding more charts to a report. Most projects already have plenty of those. The real value comes when we start connecting the dots between schedule, cost, progress, and procurement. Sometimes the first signal is small. Engineering productivity slows down slightly. Procurement approvals start taking a bit longer than usual. Actual progress begins to diverge from the planned curve. Individually, these things may not look alarming. But when you look at them together, a pattern starts to appear. And that pattern often tells you where the project is heading before the delay or cost overrun becomes visible. That is the real purpose of data analytics in project controls. Not to describe what already happened, but to help the team see what is starting to happen. When used properly, it turns project controls into an early warning system. It gives project managers and leadership the chance to react while there is still time to correct the trajectory. Because in complex projects, the biggest problems rarely appear suddenly. They usually start as small signals that only become obvious when it is already too late. #ProjectControls #DataAnalytics #ProjectManagement #ConstructionProjects #DecisionMaking

Most construction companies describe themselves as "fast followers" when it comes to innovation. And historically that's made sense. Construction operates on tight margins. When you're making 3% profit, you can't invest 5% in R&D that might not work out. The math doesn't work. Silicon Valley might be able to afford to fail fast. But construction companies can't. BUT: the rules of fast following are changing dramatically. In the past, the consequences of being a fast follower were minimal. You could let other folks bear the cost and waste of experimentation, and invest in proven solutions because the distance between leaders and followers was manageable. But the current landscape is different. It’s a flywheel, driven by one’s data foundation. The more data you have, the more you can learn, which is itself more data, which helps you learn, and so on. By the time you see robotics working on someone else's job site and decide "now it's ready," you'll be facing a massive catch-up challenge. Those robots work only because they have a massive data foundation in place. I’m not suggesting construction companies throw caution to the wind and start experimenting with all kinds of new tech. What I am suggesting is building the data layer now, even if it’s imperfect. As I mentioned in a previous post, the tools are now here to make meaning out of unstructured data. So the cost of getting it “wrong” is a lot lower. The only wrong answer is to have no data at all. Once you have that data, start implementing some digital workflows. Not exactly bleeding edge stuff, but likely to make a meaningful difference in your company. If you want to get fancy, task someone to mess around with that data using some 3rd party LLMs (or take a look at a platform like Precogs 😉). Perhaps most importantly, start getting your team comfortable with change and agility. The muscle of being able to move more quickly is going to become more important than ever. It’s time to start building that muscle now. Because by the time you can see the future clearly, it might be too late to catch up.

✅ 𝐂𝐨𝐧𝐬𝐭𝐫𝐮𝐜𝐭𝐢𝐨𝐧 & 𝐆𝐞𝐧𝐀𝐈 𝐃𝐨𝐧𝐞 𝐑𝐢𝐠𝐡𝐭 🚧 𝐈𝐧 𝐜𝐨𝐧𝐬𝐭𝐫𝐮𝐜𝐭𝐢𝐨𝐧, 𝐝𝐞𝐥𝐚𝐲𝐬 𝐚𝐫𝐞 𝐞𝐱𝐩𝐞𝐧𝐬𝐢𝐯𝐞. 𝐒𝐨 𝐚𝐫𝐞 𝐛𝐥𝐢𝐧𝐝 𝐬𝐩𝐨𝐭𝐬. We capture jobsite observations, RFIs, daily logs, photos, and safety notes—but most of it sits underutilized in silos. And when we do react, it’s already too late. That’s why we’ve taken a 𝐝𝐚𝐭𝐚-𝐟𝐢𝐫𝐬𝐭, 𝐫𝐢𝐬𝐤-𝐚𝐥𝐢𝐠𝐧𝐞𝐝 𝐚𝐩𝐩𝐫𝐨𝐚𝐜𝐡 𝐭𝐨 𝐆𝐞𝐧𝐀𝐈—one that’s grounded in: 🔹 A strong enterprise 𝐃𝐚𝐭𝐚 𝐏𝐥𝐚𝐭𝐟𝐨𝐫𝐦 🔹 Microsoft Azure’s secure AI ecosystem 🔹 NIST Cybersecurity Framework for governance 🔹 And a clear North Star: 𝐈𝐦𝐩𝐫𝐨𝐯𝐞𝐝 𝐟𝐢𝐞𝐥𝐝 𝐚𝐰𝐚𝐫𝐞𝐧𝐞𝐬𝐬 & 𝐟𝐚𝐬𝐭𝐞𝐫 𝐝𝐞𝐜𝐢𝐬𝐢𝐨𝐧-𝐦𝐚𝐤𝐢𝐧𝐠 💡 Our pilot? A 𝐆𝐞𝐧𝐀𝐈 𝐒𝐚𝐟𝐞𝐭𝐲 & 𝐑𝐅𝐈 𝐂𝐨-𝐏𝐢𝐥𝐨𝐭 that does what used to take hours— 🔹 Summarizes RFI backlogs 🔹 Flags safety issues from field reports 🔹 Converts field notes into proactive insights 🔹 All inside a secure, auditable, and compliant AI pipeline This isn’t about AI hype. This is 𝐝𝐚𝐭𝐚 𝐰𝐢𝐭𝐡 𝐩𝐮𝐫𝐩𝐨𝐬𝐞, 𝐀𝐈 𝐰𝐢𝐭𝐡 𝐚𝐜𝐜𝐨𝐮𝐧𝐭𝐚𝐛𝐢𝐥𝐢𝐭𝐲, and value delivered where it counts—𝐭𝐡𝐞 𝐣𝐨𝐛𝐬𝐢𝐭𝐞. 📊 Key to our approach? A resilient Data Platform to unify Procore, Viewpoint, and field reports—so GenAI has trustworthy, contextual information to reason with. 📢 If you're in AEC and wondering how to responsibly apply GenAI to field, safety, or subcontractor workflows—happy to share the roadmap, lessons learned, or partner up. #GenAI #ConstructionTech #AIinAEC #MicrosoftAzure #DataPlatform #NISTCSF #FieldOps #SafetyFirst #DigitalConstruction #ResponsibleAI #Procore #Viewpoint #AILeadership #DataStrategy #Dataleader #Procore

AI in construction only works when the data works In many construction projects today, the main challenge isn’t adopting AI - it’s the state of the data underneath it. Schedules, site updates, costs, drawings, and asset information still live in separate systems, owned by different teams. When the data is fragmented, even the smartest AI can’t do much. The impact is easy to see. Risks show up late, forecasts are unreliable, and teams end up reacting instead of planning. AI promises better predictions and coordination, but without connected, trusted data, those promises stay theoretical. What actually unlocks value is getting the foundations right: clean data flows between core systems, consistent structures across projects, and information that teams can rely on day to day. Once that’s in place, AI becomes practical - helping flag issues earlier, support scheduling and cost decisions, and reduce last-minute surprises on site. AI in construction isn’t about replacing experience. It’s about giving teams better signals, earlier. Where do you see AI already helping - or failing - on your projects? #Construction #ConstructionTech #ConTech #AIinConstruction #BuiltEnvironment

Celebrating Freedom to Build Better—With Technology That Works as Hard as We Do: Pre-Con & Estimating 📊 Preconstruction Isn’t Just About Pricing Anymore—It’s About Strategy, Speed, and Smarter Collaboration. At Cleveland Construction, Inc., we’ve built our preconstruction process around data, transparency, and agility—and it’s all powered by the right tech stack. Here’s a peek at the platforms we’re using to make precon faster, more collaborative, and more accurate: 🧠 JOIN - Aligning stakeholders around decisions with cost visibility and design options in real time. 📬 BuildingConnected – Streamlining bid management and subcontractor communication in one clean platform. ✅ TradeTapp (acquired by BuildingConnected) – Vetting subs with confidence using risk analysis and prequalification tools built right into our bidding process. 📁 Provision – Organizing bid packages with clarity and consistency, especially on complex scopes. 🔐 SmartBid Solutions – Secure, streamlined bid distribution and document management that scales with our workload. 📉 Unanet – Tying precon efforts directly to estimating, time tracking, and forecasting for a tighter operational loop. This tech isn’t just about replacing spreadsheets—it’s helping us: ✔️ Build better relationships with trade partners ✔️ Reduce risk earlier in the process ✔️ Get sharper on cost, faster ✔️ Give owners real-time insight into how design decisions impact budget We’re not just estimating numbers—we’re building alignment before a shovel hits the ground. #ConstructionTechnology #Preconstruction #Estimating #ClevelandConstruction #JoinBuild #BuildingConnected #TradeTapp #Provision #SmartBid #Unanet #ConTech #DigitalPrecon #BidManagement #GeneralContractor #PreconTools #ConstructionEstimating #RiskManagement #TechInConstruction #BuildingSmarter #ConstructionWorkflow #GCLeadership

Do you use Power BI? Transforming Construction in the US with Power BI The construction industry in the United States is evolving rapidly, and data-driven decision making is becoming essential to stay competitive. Tools like Microsoft Power BI are helping construction companies turn mountains of data into actionable insights. Here’s how Power BI is revolutionizing construction: 1. Real-Time Project Tracking Power BI allows project managers to visualize progress across multiple sites, monitor timelines, and spot potential delays before they escalate. This real-time insight helps teams keep projects on schedule and on budget. 2. Cost Control and Budgeting Construction projects often face cost overruns. Power BI dashboards consolidate financial data, track expenditures against budgets, and identify areas of overspend, allowing teams to take corrective action early. Safety and Compliance Monitoring 3. Safety is critical on every site. Power BI can integrate safety reports, incident logs, and inspection data to highlight trends and risks, ensuring compliance with OSHA regulations and promoting a safer working environment. 4. Resource Optimization From labor to materials, efficient resource management is crucial. Power BI helps visualize workforce allocation, equipment usage, and material consumption, enabling smarter planning and reducing waste. 5. Predictive Analytics for Future Projects By analyzing historical project data, Power BI can support predictive models for timelines, costs, and risks, helping companies plan more accurately for future projects. Bottom Line: Power BI empowers construction teams to transform raw data into insights that improve efficiency, reduce costs, and enhance safety. For US construction companies, embracing this tool isn’t just a competitive advantage it’s becoming a necessity.

💡In my AEC Business article, I discuss an innovative method for managing BIM data without relying on heavy models or specific software. Knowledge graphs offer a software-independent approach, adding meaning and context to data. 🏗️ These semantic networks can capture over 80% of the essential information needed for procurement and operations while allowing advanced queries and AI-based reasoning. Knowledge graphs present data as 'triples'—subject, predicate, object—linking entities like rooms, doors, and equipment. ⚙️ This approach supports interoperability and integration across various systems by utilizing predefined ontologies, such as IfcOWL and BOT. Knowledge graphs can connect with external data sources, EPDs, or IoT data, forming a "digital twin" for enriched data management. Specialized graph databases can store and analyze these relationships, facilitating efficient facility management, energy assessments, and other applications. Knowledge graphs, now emerging from research, are poised to become valuable tools in the construction and real estate industries. 👉 Read more at https://lnkd.in/d4FjjTPV #BIM #knowledgegraph #ontology #construction #contech #aecbusiness