Construction technology has been built around a single assumption: that this industry's core problem is design.

Architects draw. Engineers calculate. Contractors estimate. Builders build.

So the tools followed that sequence — design software, estimating software, scheduling software, field software — each one built to optimize its single step.

But somewhere in that sequence is the actual decision: what should this project cost, and where should the money go, before any of it is committed. None of those tools touch that moment.

This is not a design gap. It's a capital allocation gap.

Construction doesn't create buildings. It allocates capital.

Every commercial construction project starts with a question that has nothing to do with design: should we invest hundreds of millions of dollars to build this asset?

Everything that happens after that — the drawings, the systems, the schedules — is really just an attempt to answer that question with more confidence as the project moves forward.

A developer isn't paying for drawings. A hospital isn't paying for an HVAC system. A data center operator isn't paying for concrete and steel. They are making a capital allocation decision.

The building is simply the physical manifestation of that investment decision.

Design is a constraint, not the driver

We celebrate great architecture, and for good reason, but most projects are not driven by architectural ambition.

They're constrained by economics. The available budget determines:

  • Whether the building is feasible
  • How many floors it can support
  • Which structural system is selected
  • The quality of finishes
  • The construction schedule
  • The acceptable level of risk

These appear to be engineering or design decisions, but ask anyone who's actually sat in a preconstruction meeting; these all come down to the number on the page. 

It’s this part that often gets missed: the budget isn't a constraint you design around after the fact. It’s what shapes engineering and design from the very first number.

The missing operating system

By the time our best construction software gets involved, the decision that matters most has already been made. We use it to help draw the building, coordinate the work, or track progress on-site — which is all useful, but none of it comes in early enough to touch the first number that drives everything else.

Taking a step back, we see that every major industry has a system responsible for capital allocation.

Financial institutions have Bloomberg.

Manufacturing has ERP systems.

Supply chains have planning systems.

Construction has…spreadsheets.

Our industry's most important, most critical financial decisions are fragmented across disconnected estimates, value engineering exercises, old historical projects, and tribal knowledge.

There isn’t one shared cost intelligence layer to unite it all.

AI changes the equation

Most conversations about preconstruction AI start with automation. Automated takeoffs, automated estimates, automated scheduling…faster work means less work, no? 

These solve for speed, yes, but I’d argue that it doesn’t solve for the decisions that actually matter.

The more interesting question is whether AI can help us make better capital allocation decisions before the money is spent. 

Imagine being able to ask:

  • Which structural system maximizes long-term ROI?
  • Which design option delivers the highest value per dollar?
  • Where should we spend an additional $10 million?
  • What should we remove to stay within budget without sacrificing performance?
  • Which historical projects should influence this decision?

Right now, these are capital allocation questions that nobody can answer.

AI has the potential to answer them for us, but only if it has access to our cost data.

It needs to be:

  • In a structured database (not buried in a PDF or an email thread)
  • Connected to the decisions that produced the estimate, not just the estimate itself
  • Historical, so it can draw on what happened the last time a similar decision got made
  • Current, so it reflects real pricing instead of a number from two years ago

Without that, AI can still generate an answer — but it won't be one you can trust with a $10 million decision. 

The next generation of construction software

For decades, construction technology has focused on digitizing workflows and it’s successfully made the industry better at producing outputs.

But we started this piece with a different question: not how fast can we get to a number, but how good are the decisions behind it? 

That's the gap the next generation of construction technology has to close. And we’ll do it with a system that treats every cost decision as an asset, structures it, and makes it available the next time a similar decision has to be made.

Because construction was never really about the building. It's about the capital that gets allocated to build it.