I’ve spent the past few weeks on a mission: calling construction leaders to ask how they are using AI.
Not the futuristic stuff. Not the marketing promises. What’s actually running on job sites and in back offices right now?
What I found surprised me. Some companies are playing it safe, testing cautiously. Others are going all-in on automation. But the ones seeing real results? They share something that has nothing to do with their budget size or employee count.
Let me show you what’s actually working.
Construction Companies Don’t Use Tech Anymore. They Are Tech.
Every leader I talked to said some version of this: they stopped seeing themselves as construction companies that use technology. They’re technology companies that happen to build, landscape, or renovate.
Josh Menold from CHE told me about a Jamie Dimon quote from over a decade ago that rewired his brain: “We’re a technology company that provides financial services.”
That’s it. That single line changed how Josh approaches every business decision. Every hire. Every process. Every dollar spent.
Sounds like semantics, right? It’s not.
K&D, a landscaping company in California, created an entirely new C-level position: Head of Technology and Innovation. Not IT support. Not “the computer guy.” An executive role dedicated purely to preparing the business for what they call an “agentic and autonomous future.”
That doesn’t happen in industries that treat tech as optional.
What’s Actually Running Right Now
1) Robots Doing the Grunt Work
Forget the future. Autonomous equipment is already here.
K&D is testing autonomous mowers with two completely different strategies, and both are working.
Strategy one: The robot works alongside the crew. The machine handles the heavy repetitive mowing. Crew focuses on edging, cleanup, and detail work. Same labor hours, better output.
Strategy two: One operator runs a “milk route” with five or six autonomous units on a trailer. Drops a unit at site A, starts it, drives to site B, drops another, keeps looping. Comes back around to collect. Crews never touch the robots but show up to pre-completed work.
Sean Laux, K&D’s Head of Technology and Innovation, broke down the economics for me. You need fewer people for basic tasks. But the operators managing $100K robots? You’re paying them 50% more. One mistake is catastrophic.
The labor model isn’t disappearing. It’s evolving. And companies figuring this out first are winning talent their competitors can’t afford.
2) AI Is Quietly Transforming Back Office Operations
K&D runs agentic AI for customer service, lead management, and estimating. Not pilot programs. Production systems. Real results.
The operational AI stuff isn’t sexy. But it’s printing money.
My favorite application: K&D built internal GPT tools for quarterly departmental check-ins. Leaders answer questions about their challenges. AI synthesizes insights and suggests improvements.
Here’s what nobody expected. Employees started using those insights to redesign their own workflows. Without being asked. Without mandates. The AI just made the improvements so obvious that people took action voluntarily.
That’s the difference between forcing change and enabling it.
3) Process Automation Is Saving Hundreds of Thousands
Josh from CHE found something painful: hundreds of thousands of dollars bleeding out annually from PO and invoice mismatches.
Orders are placed manually. Invoices received manually. Humans checking humans. Errors compounding.
Not malicious. Just manual. And manual doesn’t scale.
Automation fixed it. Orders flow directly from internal systems to manufacturers. Invoices match automatically. Discrepancies flag before they become problems.
Josh’s quote stuck with me: “Without the automation piece, we’d still be hemorrhaging cash, and nobody would know exactly where.”
The Unsexy Truth: Data Governance Comes First
Before you can wrap AI around your business, you need your data house in order.
Every. Single. Leader. Said this.
Sean at K&D is spending two to three quarters (not weeks, quarters) organizing datasets before rolling out major AI features.
His quote: “It gets really messy if you try to wrap AI around your company without the infrastructure to support it.”
His framework: five lanes. Every piece of software fits exclusively in one lane and provides measurable value. Overlap? Cut. Redundancy? Gone. Doesn’t fit cleanly? Deleted.
Here’s the truth most companies don’t want to face: they fail with technology, not because the tech sucks. They fail because they try to cram new tools into old, broken processes.
The technology works fine. The process is the problem.
Data governance means being the bad guy. When the CFO wants to play fast and loose with data categorization, you shut it down. When leadership wants to skip documentation, you block it.
Sloppy data breaks everything downstream. Period.
How to Actually Start (Without Blowing It)
Based on everything Josh and Sean told me, here’s what works.
Be a student, not an expert. Download new apps constantly just to see what’s possible. You can’t implement what you don’t understand. Spend 30 minutes weekly exploring tools you’ll probably never use. The exposure matters.
Start stupidly small. Don’t overhaul everything. Find one workflow wasting 10 minutes daily. Fix that. Show the value. Expand from there.
Single point of ownership. Technology transformation can’t be a committee or someone’s side project. One person owns it in full, with executive backing and decision-making authority.
Five lanes, zero exceptions. Sean’s framework works because it’s ruthless. Software doesn’t fit cleanly in one lane with measurable value? It’s gone. The tech stack becomes a weapon, not a liability.
Build, buy, or partner. Not everything needs custom development. Not everything works off-the-shelf. Make sure vendors share your vision and offer open APIs for data integration.
Both Josh and Sean warned me: Don’t wait for competitors to prove it works. By then, you’ve already lost. Early movers compound advantages. Late adopters can’t catch up.
Let’s Build Something That Lasts
At Seisan, we work with construction companies navigating exactly this transformation. Data governance frameworks. AI implementation strategy. Custom automation that actually integrates with your existing systems.
We don’t do vibe coding. We don’t build prototypes that break under pressure. We build platforms that scale.
Whether you’re just starting to think about AI or ready to overhaul your entire operation, let’s talk. We’ll cut through the hype and focus on what actually works for your specific business.
Contact us today. Let’s build competitive advantages through technology, before your competitors do.
