Here's what the data says about construction right now.
Public infrastructure spending hit $463 billion in 2024. The highest level in over two decades. Manufacturing and data center megaprojects accounted for nearly two-thirds of all billion-dollar construction starts between 2020 and 2025. The Stargate AI data center project alone represents $400 billion in planned investment.
Money is flooding in. Projects are everywhere.
And yet 80-90% of contractors can't fill their open positions. The industry needs 439,000 additional workers just this year. There are one million fewer construction workers than there were at the 2007 peak. One in five current workers is over 55 and heading toward retirement.
The constraint isn't capital. It's not even GC office capacity.
It's labor. Specifically, it's subcontractors.
Subs are the ones who actually build things. And right now, they're drowning.
The Automation Paradox
AI is displacing white-collar workers at an unprecedented rate. Entry-level hiring in "AI-exposed" jobs has dropped 13% since large language models took off. Goldman Sachs estimates 6-7% of U.S. workers could lose their jobs to AI adoption. Ford's CEO predicts AI will halve the number of white-collar positions while demand for skilled trades keeps climbing.
The workforce is already responding. In 2024, 18 to 25 year olds made up nearly 25% of all new hires in skilled trades, despite being only 14% of the working population. Apprenticeship enrollments jumped 11%. More than half of Gen Z is now considering a trade career.
This is good news for construction long-term. But long-term doesn't help the electrical sub who's already turning down work because he can't staff his crew.
So when we build automation tools for pre-construction that let a GC blast out 500 bid requests instead of 50, what exactly are we accomplishing?
We're not solving a bottleneck. We're flooding one.
Unreasonable Hospitality
I've been reading a book called Unreasonable Hospitality by Will Guidara. He ran Eleven Madison Park, one of the best restaurants in the world. The book is about the small, obsessive details that transform a good experience into an unforgettable one.
One story stuck with me.
Early in his career, Guidara was GM of a museum cafeteria in New York. He convinced leadership to let him build a gelato cart. Nice cart, good gelato, the whole setup. But the thing he obsessed over? The spoons.
He imported tiny blue plastic spoons from Italy. The same ones you'd get at a gelateria in Rome.
His logic was simple. You could have the best gelato in the city and the most beautiful cart. But if you handed someone their cup with a generic white plastic spoon, the experience would fall flat. The spoon was the detail that made it feel real. Feel special. Feel like someone cared.
Reading that, something clicked for me about what we're building at Margo.
The Subcontractor Experience
A few days ago I got on a call with a GC friend of mine. I asked him point blank: what would make a subcontractor's life easier when they receive a bid request?
He gave me the standard answers. Clear scope. Realistic timeline. Contact info for questions.
I pushed harder. Tell me about this person. What's their day like? What are their priorities? What makes them delete an email versus actually open the drawings?
Here's what I learned.
The average sub is getting hammered with bid requests from BuildingConnected, PlanHub, and a dozen GCs they've worked with before. Their estimator might be reviewing 30 to 50 invitations a week. They have maybe 90 seconds to decide if something is worth their time.
Most of what lands in their inbox is noise. Incomplete scopes. Wrong drawings. Projects that don't match their capabilities. Requests from GCs they've never heard of for buildings three hours away.
They're not hurting for opportunities. They're hurting for time.
Finding the Blue Spoon
That conversation changed how I think about the tools we're building.
The goal isn't to help GCs send more emails. It's to make every email worth opening.
So I started asking: what's the blue spoon for a subcontractor?
Here's one thing I landed on. When a trade receives a bid request, they often get a link to a full drawing set. Hundreds of pages. The HVAC sub has to dig through architectural, structural, and civil sheets just to find the mechanical drawings that actually matter to them.
What if we intelligently broke apart the drawing set before it ever reached them? What if the mechanical contractor received only mechanical drawings, pre-organized, with the relevant specs highlighted?
It's not faster outreach. It's more thoughtful outreach.
It's expensive to build. It adds complexity. It would be easier to just blast the full set and let them figure it out.
But that's the point. The blue spoon wasn't efficient. Importing tiny plastic utensils from Italy made no sense on a spreadsheet. It made all the sense in the world when you handed someone their gelato and watched their face.
Building for the Ecosystem
When you automate a process, you're making a choice about who benefits.
Most construction tech is built for the buyer. The GC. The one writing the check. That makes sense from a business model perspective. But it creates tools that optimize for the GC's experience at the expense of everyone downstream.
More bid requests. More follow-up sequences. More automated reminders. All of it landing in the inbox of a sub who's already underwater.
I think there's a better approach.
What if we built automation that served the whole ecosystem? Tools that made the GC more efficient and made the sub's experience better at the same time?
Because here's the thing, GCs sometimes forget: your subs are your capacity. If they stop responding to your invites, your pipeline dries up. If they prioritize competitors because those competitors are easier to work with, you lose.
Winning the relationship with your sub base is a competitive advantage. And right now, most of the tech in this space is actively eroding those relationships.
The Question Worth Asking
I'm not saying every automation tool needs to be a work of art. Speed matters. Cost matters. Shipping matters.
But before you deploy something that touches another human being's workflow, it's worth asking one question:
Does this make their life easier or harder?
If the answer is harder, you might be solving your problem by creating theirs. And in a market where subs hold the cards, that's a losing strategy.
The construction industry is about to go through a massive transformation. AI is going to reshape how offices operate. New workers are going to enter the trades. Billions in infrastructure spending will break ground.
The companies that win won't just be the ones who automate fastest. They'll be the ones who automate thoughtfully. Who treat their partners like guests instead of targets. Who find the blue spoon in every process they touch.
That's what we're trying to build at Margo. Not just tools that work. Tools that care.